WO2006109423A1 - 物品位置推定装置、物品位置推定方法、物品検索システム、及び物品位置推定用プログラム - Google Patents
物品位置推定装置、物品位置推定方法、物品検索システム、及び物品位置推定用プログラム Download PDFInfo
- Publication number
- WO2006109423A1 WO2006109423A1 PCT/JP2006/305401 JP2006305401W WO2006109423A1 WO 2006109423 A1 WO2006109423 A1 WO 2006109423A1 JP 2006305401 W JP2006305401 W JP 2006305401W WO 2006109423 A1 WO2006109423 A1 WO 2006109423A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- article
- person
- information
- detection
- time
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
Definitions
- Article position estimation device article position estimation method, article search system, and article position estimation program
- the present invention relates to an article position estimation device, an article position estimation method, and an article search system for managing articles in a general house or office, and in particular, daily necessities and offices used for life in a general home.
- an article position estimation device for managing the position of various items such as portable items used in the system using R FID tag technology etc.
- the position of the item to be searched is appropriately displayed.
- the present invention relates to an article position estimation device, an article position estimation method, and an article search system.
- Patent Document 1 Japanese Patent Laid-Open No. 07-146362
- a tag an RFID tag
- a reading device for the tag hereinafter referred to as a tag reader
- a terminal for searching for an article is provided.
- each tag reader tries to communicate with the tag attached to the article, and the place of the article to be searched is determined by the place of the tag reader that receives the reply from the tag. It is.
- a tag is set for a book to be managed, and a tag reader is set on each shelf of a library book rack.
- each tag reader searches for a tag including the ID of the search target book. If there is no response from the tag reader in response to an inquiry from the tag reader, there is no corresponding book in the communicable range of the tag reader, and conversely, if a response from the tag is received in response to an inquiry from the tag reader. Is applicable When a book is found, it can be said that it is the approximate location of the book that the tag reader receiving the response from the tag is searching for. Therefore, the user can get to the target book by going to the location of the reader and searching for only the books around the reader.
- the conventional system can manage the location of the article while having a very simple configuration, and is therefore starting to be used in various fields mainly for business use.
- the disadvantage of this example is that to increase the accuracy of the search position, the tag attached to the article can communicate with any tag reader anywhere in the environment where the article is managed. This means that it is necessary to install a tag reader. Therefore, it is a big problem that it cannot be used for home use, where the cost of power is an important issue in business applications where costs are allowed to be somewhat high.
- Patent Document 2 Japanese Patent Laid-Open No. 2000-357251
- Patent Document 2 Japanese Patent Laid-Open No. 2000-357251
- an article management device a plurality of sensor units (tag readers), a tag unit attached to an article, and a reception sensitivity when receiving tag information transmitted from the tag unit through the plurality of sensor units.
- This technology uses an active tag with a built-in battery because it needs to reach the tag's radio waves to multiple sensor units that are relatively distant from each other.
- a method is adopted in which transmission is performed at a predetermined timing (once a day to once a month).
- Patent Document 3 Japanese Patent Laid-Open No. 2003-233715
- an observation device tag reader
- Living information such as human behavior history and location information of managed objects is collected and managed.
- Data tags are attached to objects to be managed, and people carry personal tags on their hands. These data tags and personal tags hold identification information of objects to be managed and persons, and can communicate information with an observation device wirelessly at a plurality of communication distances.
- Patent Document 1 Japanese Patent Application Laid-Open No. 07-146362
- Patent Document 2 Japanese Patent Laid-Open No. 2000-357251
- Patent Document 3 Japanese Patent Laid-Open No. 2003-233715
- Patent Document 2 since the tag does not transmit force at a predetermined timing, the position of the article is determined based on the last record of the tag attached to the article and the tag reader. Therefore, after communicating with the tag force S tag reader, the component where the article with the tag is carried is lost until the next communication with the tag force S tag reader. Furthermore, it goes without saying that the longer the time has passed since the last communication with the tag reader, the lower the possibility that the tagged article will be in the vicinity of the last tag reader with which communication was made. Therefore, in order for a user to find an article using such a system, it is necessary to remember and remember from the location of the article informed by the search query, such as how the user himself / herself moved and where the article was placed.
- Patent Document 3 it is necessary to install a plurality of tag readers in the same room as Patent Document 2, which is disadvantageous in terms of cost.
- data tags and personal tags can be transmitted wirelessly, there is a problem that the size of the tag itself is increased and the cost is increased.
- an object of the present invention is to provide a simple configuration as in the prior art, that is, a tag on an article.
- the present inventors first deal with goods in a home or office, mainly those who live in the home or work in the office. Therefore, we noticed that there should be an important correlation between the movement of the person and the location of the goods. Therefore, if human movement information could be obtained by some means, we thought that it would be possible to narrow down the location of goods using that information.
- the present invention relates to a technique for realizing this idea in order to solve the above-mentioned problems.
- an article detection device that is arranged at least near the entrance of a room where a person can enter and exit to identify and detect different articles.
- An article management database for storing article identification information, detection place information, and detection time information detected by the article detection device as article management information
- a person detecting device for detecting the position of the person in the room by distinguishing each person, a person management database for storing movement history information of the person detected by the person detecting device,
- an article position estimation device comprising: article presence area estimation means for estimating the subsequent movement area of the detected person as an article presence area of the article.
- an input device for inputting a target article to be searched
- the article position estimation device that estimates the article presence area of the article by the article presence area estimation unit, and searches for the search target article input by the input device from the estimated articles.
- a display device for displaying the article presence area where the search target article exists, using the article presence area estimation means, or the estimation result of the article presence area estimation means and the article position candidate weighting means;
- An article search system comprising:
- the article detection device detects an article, and stores the location and time in the article management database.
- the human detection device detects a person and stores the location information and time information in the human management database.
- the article presence area estimation means refers to the information stored in the article management database and the information stored in the person management database, and there is a possibility that an article exists in a route traveled by a person as the article existence area. A certain place can be estimated.
- the article position candidate weighting means uses information such as human movement history data and device operation information, so that there is a possibility of existence for each place where the article may exist. It is also possible to weight the height.
- the user of the system can prioritize the places to be searched by referring to the location where the weighted articles are present, and search for the search target articles in descending order of strength.
- the article position estimation apparatus and method, article retrieval system, and article position estimation program according to the present invention enable a user to easily and quickly find an article to be searched.
- an article can be detected only within the detection range of the article detection device. Therefore, if there is an article at a location other than that, the user can see the detection result of the article detection device. He was forced to perform a task that required thinking, such as searching for places while remembering them.
- the location where the article is present can be further narrowed down, and if necessary, weighting can be performed to indicate the high possibility of the presence or absence of the article at the narrowed-down location.
- the user can search for the target article more easily and quickly without the need for the thought work of remembering the user's past behavior.
- FIG. 1 is a block diagram showing a typical configuration example of an article search system according to first and second embodiments of the present invention.
- FIG. 2 is a sketch showing an example of an environment in which a tag reader, which is an example of an article detection device of the article search system that works on the first embodiment of the present invention, is installed at the entrance of each room. Yes,
- FIG. 3 is an explanatory diagram showing an example in which a gate-type tag reader, which is an example of an article detection device, is installed at the entrance of the “living room”.
- FIG. 4A is a conceptual diagram showing an example in which a tag is attached to a book as an example of an article.
- FIG. 4B is a conceptual diagram showing an example in which juice is tagged as another example of an article.
- FIG. 5A is a diagram showing an example of the article management information included in the article management database of the article search system in a tabular form
- FIG. 5B is a diagram showing an example of the article management information included in the article management database in a table format.
- FIG. 6 is a conceptual diagram showing a system configuration for detecting the position of a person using a weight sensor as an example of the human detection device of the article search system;
- FIG. 7 is a conceptual diagram showing a system configuration for detecting the position of a person using a tag as another example of the human detection device of the article search system
- FIG. 8 is an auxiliary diagram for specifically explaining the background subtraction method as still another example of the human detection device of the article search system.
- FIG. 9A is a diagram showing an input image taken at a certain point in time using the same camera that took the background image to specifically explain the background subtraction method
- FIG. 9B is a diagram showing an example of a background image for specifically explaining the background subtraction method.
- FIG. 9C in order to specifically explain the background difference method, a coordinate system of the environment is added to the background difference image obtained by subtracting the background image of FIG. 9B from the input image of FIG. 9A. It is a figure shown,
- FIG. 10 is an explanatory diagram for explaining a calculation for converting a position coordinate of a cut-out person in a camera image into a world coordinate system.
- FIG. 11A is a diagram showing an example of human management information included in a human management database in the article search system in a tabular form.
- FIG. 11B is a diagram showing an example of human management information included in the human management database in the article search system in a tabular format
- FIG. 12 is a diagram showing an example of human movement history information in the article search system in a tabular format
- FIG. 13 (a) and (b) in FIG. 13 refer to the human movement history information in the article search system, respectively, and a draft and a plot in which the movement amount (movement distance) at the time is plotted for each time.
- FIG. 3 is a diagram in which the person movement history information is plotted on a floor plan in which only “living room” is extracted from the floor plan in FIG. 2;
- FIG. 14 (a) and (b) of FIG. 14 are cases in which data for two persons are plotted in the article search system, respectively, and the person movement history information is referred to and the data is
- FIG. 3 is a graph in which the movement history information is plotted on a graph in which only the “living room” is extracted from the graph in which the movement amount (movement distance) at the time is plotted and the sketch in FIG. 2;
- FIG. 15A is a diagram showing a device having a storage function in the article search system.
- FIG. 15B is a diagram illustrating a device having a storage function in the article search system.
- FIG. 16A is a diagram showing an example of device management information included in a device management database in a table format in the article search system
- FIG. 16B is included in the equipment management database in the article search system. Is a diagram showing an example of device management information in a tabular format
- FIG. 17 is a flowchart showing a flow of processing in the article presence area estimating means in the article search system
- FIG. 18 is a diagram showing, in a tabular form, processing results in the article presence area estimation means in the article search system,
- FIG. 19 is a flowchart showing a process of performing weighting using the moving speed of a person in the article search system
- FIG. 20 is a diagram showing processing results in the article position candidate weighting means in the article search system in a tabular form
- FIG. 21 is a conceptual diagram showing an example in which an article search result is displayed in CG (computer graphic) on the display device in the article search system;
- FIG. 22 is a flowchart showing a process of weighting using device management information in the article search system
- FIG. 23 is a diagram showing a database for managing device locations in the article search system in a tabular format.
- Figure 24 is a sketch showing the equipment placed in the environment.
- FIG. 25 is a diagram for explaining a weighting process according to the distance between a person's staying position and the device in the article search system
- FIG. 26 is a diagram showing in tabular form the results of weighting processing according to the distance between the person's staying position and the device in the article search system,
- FIG. 27 is a conceptual diagram showing an example in which an article search result is displayed in CG (computer graphic) on the display device in the article search system that works on the second embodiment of the present invention
- FIG. 28 is a conceptual diagram showing a state in which the moving images in the article search system that are useful for the second embodiment of the present invention are given a time stamp and accumulated.
- FIG. 29 is a flowchart showing the flow of processing in image search means in the article search system according to the second embodiment of the present invention.
- FIG. 30A is a diagram of the article search system according to the second embodiment of the present invention. It is a conceptual diagram showing how the 19:31 image is retrieved and displayed from the image database in FIG.
- FIG. 30B is a conceptual diagram showing a state in which an image of 19:32 is retrieved from the image database of FIG. 28 and displayed in the article search system that is relevant to the second embodiment of the present invention. Yes,
- FIG. 31 shows information stored in an article management database in the article search system of the first embodiment, in which an article name and a tag ID assigned thereto are paired. It is an explanatory diagram of the state,
- FIG. 32 is a conceptual diagram showing a system configuration for detecting the position of a person using a weight sensor as an example of the person detection device of the article search system of the first embodiment.
- FIG. 33 is an explanatory diagram of article owner information in the article search system of the second embodiment.
- FIG. 34 is a sketch showing an example of an office environment in which the article detection device in the article search system is installed
- FIG. 35 is a flowchart of an article position estimation program capable of realizing a part of the article search system
- FIG. 36A is a graph showing the relationship between the time and the actual position of the person at that time.
- FIG. 36B is a graph showing the position P of the person recorded at the time interval At.
- FIG. 37 if the average moving speed in the predetermined time zone after staying is higher than the average moving speed in the predetermined time zone before staying, it is determined that the possibility of releasing the carried item is high.
- FIG. 38 is a graph showing the processing result in step 3705 of FIG. 37.
- FIG. 39A is a graph showing the time change of the moving speed of two persons H1 and H2.
- FIG. 39B shows the moving path of two persons HI and H2 on the floor plan of the room. It is a figure displayed on top of
- Fig. 40 explains the processing by the article presence area estimation means considering the possibility of delivery. Is a flowchart to explain,
- FIG. 41 is a diagram showing data used when processing by the article presence area estimation means considering the possibility of delivery,
- FIG. 42A is a diagram showing data representing changes in the moving speeds of two persons HI and H2, which are used in the article presence area estimation process in the article existence area estimation means when delivery of articles is considered.
- FIG. 42B shows the movement trajectory data of the two persons HI and H2 superimposed on the floor plan in the article existence area estimation process by the article existence area estimation means when the delivery of the article is considered. It is a figure that
- FIG. 42C is a diagram showing a movement trajectory presented on the display device in the article presence area estimation process by the article presence area estimation means when taking the article into consideration. Is a flowchart showing the processing in the article position candidate weighting means considering the possibility of delivery,
- FIG. 44A shows the movement speed of the two persons HI and H2 used for the article presence area estimation process in the article existence area estimation means when taking the article into consideration, as in FIG. 42A. It is a figure which shows the data showing a change,
- FIG. 44B shows the movement trajectory data of the two persons HI and H2 in the article presence area estimation process in the article presence area estimation means when the delivery of the article is considered. It is a figure expressed overlaid on the floor plan of the room,
- Figure 44C shows the article presence area estimation process performed by the article position candidate weighting means when the delivery of the article is taken into consideration, and the obtained article presence area is mapped onto the environment sketch by the article presence area estimation means. It is a figure which shows a result,
- FIG. 45A is a graph showing the time change of the moving speed of two persons HI and H2.
- FIG. 45B is based on the graph of FIG. 45A in time zones S1 to S5. It is a figure which shows the table
- FIG. 45C is a diagram in which the movement trajectories of two persons HI and H2 are superimposed on the floor plan of the room and displayed.
- FIG. 46 is a flowchart showing a process of reflecting the change in the moving speed before and after the staying state in the weighting. It is a chart. BEST MODE FOR CARRYING OUT THE INVENTION
- the first aspect of the present invention at least near the entrance of a room where people can go in and out, different articles (for example, different types of articles) are identified and detected so as to be distinguished.
- An article detection device At least near the entrance of a room where people can go in and out, different articles (for example, different types of articles) are identified and detected so as to be distinguished.
- An article management database for storing article identification information, detection place information, and detection time information detected by the article detection device as article management information
- a person detecting device for detecting the position of the person in the room by distinguishing each person, a person management database for storing movement history information of the person detected by the person detecting device,
- a predetermined time including the detection time of the article (this predetermined time) ) Means a predetermined time for estimating the presence area of the article.)
- the person detected in is specified, and the person and the person are identified based on the movement history information of the person management database.
- an article existence area for estimating a subsequent movement area of the detected person as an article existence area of the article
- An article position estimation device comprising: an estimation unit;
- the article presence area estimating means is configured to change the movement speed of the person obtained by referring to the movement history information stored in the person management database.
- the article position estimation device according to the first aspect for estimating the article existence area of the article is provided.
- the estimated plurality of article existence areas with reference to information stored in the article management database or the person management database, the estimated plurality of article existence areas
- the first or second aspect further comprising article position candidate weighting means for performing weighting so that the weight of the article existence area is high based on the possibility that the article exists and the weight of the article existence region is high.
- the article position estimation apparatus described in 1. is provided.
- the article position candidate weighting means includes
- a predetermined value for determining the staying state.
- the region where the average value of the person's moving speed is lower is extracted using the average value of the person's moving speed obtained by extracting the time information as the staying state and referring to each information constituting the staying state.
- An article position estimation apparatus according to a third aspect is provided that performs weighting so that the weight of an article existence area where the article is likely to exist is increased.
- the article position candidate weighting means includes
- the detection location information and detection in the movement history information having a movement speed equal to or lower than a predetermined value (a predetermined value for determining the staying state). Extracting the time information as a staying state, and using the change in the moving speed of the person obtained by referring to the information on the detection location and the detection time information before and after the staying state, the possibility that the article exists.
- the article position estimation apparatus which performs weighting so that the weight of a high article existence area is increased.
- the article position candidate weighting means includes
- An article position estimating apparatus When weighting the possibility that the article exists, weighting is performed so that the weight of the article existing area where the article is likely to exist is increased by using the time during which the staying state is continued.
- the article presence area estimation means includes
- the article position estimation device Provide a position.
- a device operation detection device that detects device operation of a device that stores and manages the article
- a device management database for storing operation information of each device detected by the device operation detection device; information stored in the device management database is information on presence / absence of operation of the device in the operation information
- the article position estimation apparatus which weights the article position candidate equipment so that the weight of the equipment used is increased.
- the article presence region estimation means includes the person and the article after the case where the person detection device and the article detection device are detected simultaneously.
- the detected movement area of the person is estimated as the article existence area of the article, and the estimation of the article existence area of the article is stopped after the detected person leaves the room.
- An article position estimation apparatus according to the second aspect is provided.
- the input device for inputting the target article to be searched for, and the article presenting area estimated by the article existing area estimating means while estimating the article existing area of the article is provided.
- the article position estimation device according to any one of the first to ninth aspects, which searches for an article to be searched input by the input device,
- a display device for displaying an article presence area in which the search target article exists, using the article presence area estimation unit, or estimation results in the article presence area estimation unit and the article position candidate weighting unit;
- An article search system comprising:
- an imaging device that captures an environment in which the article search is performed, an image database that stores image information captured by the imaging device, the article presence area estimation unit, The article existence region estimation means and the article position candidate weighting means estimate the place and time where the article requested to be searched is estimated to be placed, and photograph the place using the estimation result.
- image search means for extracting image information including the time, the image database.
- the display device provides the article search system according to the tenth aspect, characterized in that the image information searched by the image search means is displayed.
- the step of identifying and detecting with the article detection device so that different articles (for example, different types of articles) are distinguished at least in the vicinity of the entrance / exit of the room where people can go in and out.
- a predetermined time including the detection time of the article Means a predetermined time for estimating the presence area of the article.
- the person detected in is specified, and the person and the person are identified based on the movement history information of the person management database. Estimating the subsequent movement area of the detected person as the article presence area of the article when the article is detected simultaneously by the person detection device and the article detection device;
- An article position estimation method is provided.
- the article position estimating method according to the twelfth aspect is further provided.
- a computer includes:
- a predetermined time including the detection time of the article (this predetermined time)
- article detection is performed by identifying and detecting different articles (for example, different types of articles) that are arranged at least in the vicinity of a doorway where a person can enter and exit.
- articles for example, different types of articles
- a person detecting device for distinguishing and detecting the position of the person in the room for each individual; information on the detection time of the article stored in the article management database; and the person stored in the person management database Using human movement history information, a person detected within a predetermined time including the detection time of the article (this predetermined time means a predetermined time for estimating the article presence area). And when the person and the article are detected simultaneously by the person detection device and the article detection device based on the movement history information of the person in the person management database, the detected Article presence area estimation means for estimating a subsequent movement area of a person as an article existence area of the article;
- An article position estimation device is provided.
- FIG. 1 shows a typical configuration example of the article search system according to the first embodiment of the present invention.
- FIG. 1 is a block diagram, and FIG. 1 includes all the means or devices described in the various aspects of the present invention.
- the article retrieval system according to the first embodiment of the present invention basically has three parts as a whole, that is,
- An article position estimation device 140 that performs a process of estimating the position of the article, and searches for an article to be searched input by the input device 109 from the estimated articles;
- a display device 110 for displaying the location of the search target article input by the input device 109 using the estimation processing result by the article position estimation device 140;
- the article position estimation apparatus 140 can also search for time information in addition to the information on the article to be searched input from the input device. Is possible. That is, specifically, the force described later as the second embodiment of the present invention.
- An imaging device 111 that captures the environment in which the time information output from the timer means 120 is input and the article search is performed;
- An image database 112 for storing image information taken by the imaging device 111 together with time information
- Image search means 113 for estimating and extracting the image information including the above-mentioned time from the image information stored in the image database 112,
- a portion 140 surrounded by a solid line in FIG. 1 is a block diagram corresponding to the article position estimating apparatus 140 according to the first embodiment of the present invention. All means or apparatus and various databases described in the 1st to 15th aspects Shows the configuration including
- 'Timer means 120 to output current time information
- An article detection device 101 for identifying and detecting so that the time output from the timer means 120 is input and different articles are distinguished;
- the identification information (ID) of the article detected by the article detection device 101, the information on the detection location of the article, and the time information (detection time information) output from the timer means 120 at the time of detection are detected for each article detection.
- a person detection device 103 that receives the time information output from the timer means 120 and detects the position of the person for each person;
- the identification information (ID) of the person detected by the human detection device 103, the information on the detection location of the person, and the time information (detection time information) output from the timer means 120 at the time of detection are And a human management database 104 stored as human management information (human movement history information),
- ID Identification information of each device detected by the device operation detection device 105
- operation information information on the operation status of the device
- time information output from the timer means 120 when the device is operated operation time Information
- the article presence area estimation means 107 for searching for the search target article input by the input device 109 from the articles whose article existence area is estimated
- the article detection apparatus 101 receives the time information (detection time information) output from the timer means 120 and identifies and detects the different articles so that the identification information (ID) of the detected articles is detected. And the information on the detection location of the article and the time information (detection time information) output from the timer means 120 at the time of detection are output to the article management database 102 for each article detection.
- the article management database 102 for each article detection.
- FIG. 2 is a sketch showing an example of an environment in which the article detection apparatus 101 is installed (an example of a room in a general household).
- the XY coordinate system is set with the vertical direction in Fig. 2 as the Y-axis and the horizontal direction as the X-axis.
- a portion indicated as TGR is a tag reader as an example of the article detection apparatus 101, and includes “entrance”, “living room”, “study”, “bathroom”, “toilet”, and “bedroom”. It is installed near the doorway.
- FIG. 3 shows an example in which it is installed in the vicinity of an entrance where a person of “gate type tag reader TGR force ⁇ living room” can enter and exit.
- the gate type tag reader TGR can function alone, but for reasons such as to prevent mistakes in reading the tag TG, here the gate type tag reader TGR is installed in a pair as opposed to each other.
- Figure 3 shows an example.
- a tag TG for example, a passive tag with a relatively short distance (eg, near an entrance / exit) that does not detect force). It needs to be attached.
- Fig. 4A and Fig. 4B are conceptual diagrams showing an example in which a tag TG is attached to an article.
- Fig. 4A shows a tag for "book”
- Fig. 4B shows a tag for "juice”.
- An example in which TG is given is shown.
- the information of the tag TG read and read is stored in the article management database 102 together with the time of reading.
- the “living room” in FIG. 3 includes a bookshelf (BS), refrigerator (RF), kitchen system (KS), dining table (DT), sofa 1 (SF1) as shown in FIG. Sofa 2 (SF2) and Rotable (LT) are installed!
- the articles tagged in this way are stored in the article management database 102 as a pair of the article name and the tag ID assigned thereto.
- the tag reader TGR reads the ID of the tag TG, it is possible to determine what the article corresponds to the ID.
- other information may be added and stored in the article management database 102.
- the other information is information such as the category, weight, shape, and color of the article.
- the article management database 102 includes the article identification information (ID) detected by the article detection device 101, information on the detection location (detection position in the room) of the article, and the time output from the timer means 120 at the time of detection. This information (detection time information) is stored as article management information for each article detection.
- 5A and 5B are diagrams showing examples of the article management information included in the article management database 102 according to the first embodiment of the present invention in a tabular form.
- the article management database 102 is the article management information. Is a collection of “Goods” here refers to portable items that users use in daily life (or portable items that are normally used in offices, etc.), such as furniture and home appliances in homes (or offices). Desks and bookcases etc.) are distinguished from “articles” by the name of “equipment” or “equipment”.
- FIGS. 5A and 5B show article management information corresponding to “book” in FIG. 4A and “juice” in FIG. 4B, respectively.
- the article management information related to “book” in FIG. 4B shows article management information corresponding to “book” in FIG. 4A and “juice” in FIG. 4B, respectively.
- Each tag reader TGR detects the tag TG of each item, and each time the tag reader TGR obtains the ID of each item, the tag TG ID is paired with the tag TG ID. And the correspondence information in the correspondence database (see FIG. 31) of the article, the article corresponding to the ID of the tag TG is examined, and the result is written in the article management database 102 over time.
- the correspondence database between the tag TG ID and the article may be arranged separately from the article management database 102 as long as the tag reader TGR can be referred to, or the article management information is stored in the article management database 102. You may make it arrange
- the article management database 102 may be deleted from the article management database 102.
- the article management information may be deleted manually using the input device 109. If the tag reader TGR is installed in the trash box and the tag reader TGR detects the tag TG, the tag reader TGR The article management information corresponding to the tag TG may be deleted from the article management database 102.
- the human detection device 103 receives the time information output from the timer means 120 and detects the person's position by distinguishing each person. As a result, the human identification information (ID) and the detected person identification information (ID) are detected. Information on the detection location of the person and information on the time (detection time information) output from the timer means 120 at the time of detection are output to the human management database 103 for each unit time.
- human detection is a force that can be perceived in various ways, for example, when a person included in an image is cut out from the image, or where a person in a certain environment finds a place in the environment. Used to mean the position coordinates of a person at Various detection methods have been proposed for this purpose. In the first embodiment, three types of methods, a method using a weight sensor, a method using a tag TG, and a method using an image will be described below.
- FIG. 6 and 32 are conceptual diagrams showing a system configuration for detecting the position of a person using the weight sensor WSEN.
- the human detection device 103 includes a weight sensor WSEN and a sensor processing device 141 to which output information from the weight sensor WSEN is input and can be connected to the article management database 102.
- the weight sensor WSEN When detecting the position of a person using the weight sensor WSEN, as shown in FIGS. 6 and 32, the weight sensor WSEN is laid on the floor in the environment, for example, in a lattice pattern.
- Each of these weight sensors WSEN is configured to output a value when the weight is applied by a human foot or the like, and all of them are connected to the sensor processing device 141.
- the sensor processing device 141 registers the coordinates of each weight sensor WSEN in advance and always senses the data of the weight sensor WSEN. If a weight sensor WSEN is loaded, a load is applied. If this is detected, the sensor processing device 141 immediately obtains the coordinates where the weight sensor WSEN is placed.
- the sensor processing device 141 may add processing such as grouping sensing data having substantially equal weights of the weight sensor WSEN and arranging them in time order. Thereby, the accuracy of human detection can be improved.
- FIG. 7 is a conceptual diagram showing a system configuration for detecting the position of a person using a tag TG.
- the human detection device 103 includes a tag TG and a tag reader TGR that can detect the tag TG and can be connected to the article management database 102 by radio or the like.
- the tags TG are spread on the floor in the environment, for example, in a lattice pattern.
- the footwear 142 is provided with a tag reader TGR for reading the HD information of the tag TG laid on the floor, and the tag reader TGR includes the tag TG having the ID information and the tag TG having the ID information.
- Correspondence information (not shown) corresponding to the position information (for example, position coordinate information) of the place where the object is placed is stored.
- the footwear 14 2 and the tag reader TGR force at the moment the foot is stepped on the floor communicates with the tag TG below the tag TG, and the tag TG
- the person's position can be specified by reading the ID information and comparing the value of the information with the association information.
- this is only an example, and a configuration in which a tag reader TGR is attached to a place other than footwear 142 and a tag TG is attached to a place other than a person's foot may be used.
- a tag TG with an individual ID is embedded in a wrist watch attached to a wrist, it is possible to detect a person with the tag reader TGR for the article detection apparatus 101 described above.
- Step 1 Extracting a person from the image
- Step 2 Converting the position coordinates of the extracted person in the camera image to the world coordinate system.
- step 1 the force that is the clipping of a person with a medium image power, various methods have been developed. Here, the simplest background difference method will be described.
- the background subtraction method prepares a model image as a background in advance, This is a method of obtaining an object to be processed from an image by taking a difference from the model image.
- a background image is created. For example, when there is no environmental change, a single image in which no humans exist in the environment may be used. It is also possible to use images obtained by averaging images taken continuously.
- 8 and 9A to 9C are auxiliary diagrams for specifically explaining the background subtraction method.
- Fig. 8 is a conceptual diagram showing that the camera 143 is installed in the environment and the world coordinate system is set in the environment.
- Fig. 9B is a diagram showing an example of the background image.
- Fig. 9A is a diagram showing Fig. 9B.
- Figure 9C shows an input image (captured image) taken at a certain point in time using the same camera 143 that was taken.
- Figure 9C shows the background obtained by subtracting the background image of Figure 9B from the input image of Figure 9A.
- FIG. 6 is a diagram showing a difference image with a coordinate system of the environment added.
- the input image in Fig. 9A includes person 14 4 and the background image in Fig. 9B does not include person 144, which is the difference between these two images. Person 144 also appears in the background differential image power of Fig. 9C.
- the coordinates of the foot of the person 144 in the image can be obtained by image processing.
- the human detection device 103 since what is ultimately desired by the human detection device 103 is the coordinates of the person 144 in the world coordinate system, the method will be described with reference to FIG.
- FIG. 10 is a diagram for explaining calculation for converting the position coordinates of the clipped person 144 in the camera image into the world coordinate system.
- O O
- the coordinate system composed of the X axis, Y axis, and Z axis with w as the origin is the world coordinate system.
- O is the camera coordinate system whose origin is the lens center of the camera 143.
- the position coordinate of the person 144 is represented by (x, y, z).
- (U, V) is the camera e e e
- the rotation matrix around one axis, and (t, t, t), is the origin of the camera coordinate system and the origin of the world coordinate system
- the above is the outline of the method for detecting the person 144 using the camera image. If a plurality of people are detected in the camera 143, the above process may be performed individually.
- the human detection device 103 when using a camera image for the human detection device 103, it does not matter if the image captured by the imaging device 111, which is another component of the first embodiment of the present invention, is shared.
- the human detection device 103 is connected to the camera 143, the image information from the camera 143, and the arithmetic processing such as the background difference processing and the coordinate conversion processing, and is connected to the article management database 102. It comprises possible computing means 145.
- any of these methods can detect a person completely. It is not guaranteed. Therefore, in order to perform human detection with higher accuracy, it is possible to use these methods in combination, use other methods, or combine these methods with other methods. Absent.
- the method used in the article detection apparatus 101 and the method used in the person detection apparatus 103 may be shared with each other.
- only an example of a method suitable for each means has been introduced, and the methods may be used in combination according to the environment and cost to which the present invention is actually applied.
- the human management database 104 stores the identification information (ID) of the person detected by the human detection device 103, information on the detection location of the person, and the time output from the timer means 120 at the time of detection.
- Information is stored as person movement history information (person management information) every unit time.
- FIG. 11A and FIG. 11B are diagrams of person management information (for example, person movement history information including information such as person identification information, position coordinates and time) included in the person management database 104 according to the first embodiment of the present invention.
- the figure shows an example in the form of a table.
- the human management database 104 is a collection of such human management information.
- person management information (person movement history information) corresponding to the father in FIG. 11A and the mother in FIG. 11B is shown.
- the human management information (person movement history information) related to Dad in Figure 11A is
- the difference between the human management information (human movement history information) and the article management information is that the human management information detects the person detection location and the detection time for each unit time. Is included as human movement history information.
- This human movement history information is a detailed detection of how a person is actually moving, and the detected information (information such as the person's identification information and location (location coordinates) and its time) is stored. Next, the details of the human movement history information will be described with reference to FIG.
- FIG. 12 shows an example of human movement history information in a table format.
- the example of FIG. 12 shows the history data MF — Data02 in “living room” included in the father's person movement history information (FIG. 11A).
- This history data MF—Data02 is the history of human movement until the father passes the tag reader TGR of “living room” at 22:29 the next time after his father passed the “living room” tag reader TGR at 19:30. Is described.
- Each person's movement history information in the table consists of three elements: time, X-coordinate value, and Y-coordinate value. Time is the elapsed time since recording started, and X and Y coordinate values are shown in Fig. 2.
- the coordinate values in the coordinate system of the sketch shown in Fig. 1 are shown in mm.
- this table shows that the father was at the coordinates (5766, 23 04) at time 1 after passing the tag reader TGR of "living room" at 19:30. .
- the power of omitting the time unit This is because the interval of information recording differs depending on the system capability, and it should be decided appropriately according to the system capability (eg 1 second interval).
- the upper graph (a) in FIG. 13 refers to the human movement history information, calculates the movement distance between two consecutive times (hereinafter referred to as unit time), and the movement. It is a graph plotting distance on the vertical axis and time on the horizontal axis. Therefore, in the upper graph (a) of FIG. 13, the larger value on the vertical axis represents the larger moving amount per unit time and the smaller value represents the unit. It represents that the amount of movement in time is small. In other words, if the place where the value is not small continues, the time is moving.If the place where the value is small continues, the time stays! /, Ru ((in the house! , Te) stays locally (for a short time).
- a dotted line A is drawn in parallel with the horizontal axis slightly above the horizontal axis, but if the movement amount per unit time is less than this dotted line A, If the definition is defined as “Staying” with the dotted line A and subsequent parts as a lump, the three “Stalling” as shown in the upper graph (a) in FIG. It can be seen that there are parts (ie “bookcase”, “refrigerator and kitchen system”, “sofa 1”).
- the criteria for judging whether it is “staying” or moving will change, but the simplest method is to change the value on the vertical axis of dotted line A.
- a method of fixing to a fixed value is conceivable.
- the value may be changed depending on the person, or may be changed depending on what kind of goods the person is handling. For example, if the person to be detected is an elderly person, or if the article to be detected is a heavy article or a fragile article, it means that it is not “staying” but moving, even if the moving distance is continuous.
- the value on the vertical axis of dotted line A is It is desirable to make it smaller than the case of goods.
- Information on the articles handled can be obtained from the article detection device and the article management database.
- changing the standard for each detection target means determining the standard according to the normal movement speed of the detection target. In other words, this is equivalent to focusing on the difference between the normal moving speed and the moving speed at a certain point in time.
- the reference value may be changed according to the position where a person exists (for example, room unit). For example, considering the entire house shown in Fig. 2, the corridor usually has no obstacles so that it is easy to move. Easy to grow.
- FIG. 36A shows a graph PH showing the relationship between time and the actual position of the person at that time.
- Fig. 36B shows the position coordinates P of the person accumulated at time interval At.
- the movement speed of the person is represented by the slopes of the lines in FIGS. 36A and 36B, respectively, but the line segment (P ⁇ P) in FIG. 36B is smaller than the actual movement speed. Therefore, accumulation k k + 1
- the interval may be determined according to the movement speed of a person who has been measured in advance. In addition, if you want to include the case where a person repeatedly moves in a local area in the “Stay” intentionally, the storage interval may be increased.
- the accumulation interval may be reduced.
- two people, a mother and a daughter take food from a bag or shopping cart that has been purchased and packed near the kitchen system.
- the product is placed in a refrigerator or food storage, or miscellaneous goods are placed on a table.
- the mother and daughter repeatedly move between the position of the nog or shopping cart and the position of the refrigerator or food storage table.
- the accumulation time interval may be made sufficiently small.
- the lower plan view (b) of FIG. 13 is a plan view in which only “living room” is extracted from the sketch diagram of FIG. 2, and the human movement history information in the upper graph (a) of FIG. FIG. Since the person movement history information originally includes time information, it can be associated with the upper graph (a) in FIG. In this example, if the “dwell” part of the upper graph (a) in FIG. 13 is mapped to the floor plan in the lower plan view (b) of FIG. 13 (see dotted line with arrows), the earliest time In order, it can be determined that the object is staying in the vicinity of “Bookcase”, “Refrigerator and Kitchen System”, and “Sofa 1”.
- FIG. 13 shows only an example relating to one person's movement history information.
- the same processing may be performed for two or more persons.
- the upper graph (a) in Figure 14 shows two people
- one person is indicated by the same solid line B as the data in FIG. 13 and the other data is indicated by the dashed-dotted line C.
- This dash-dotted line C data shows that there is only movement at the beginning and no movement after that, but the corresponding movement in real space is the same in the bottom plan view (b) of Fig. 14.
- This is indicated by the alternate long and short dash line C.
- the other person indicated by the alternate long and short dash line C is sitting on "Couch 2" immediately after entering the room. From the above, it can be said that the results correspond to the results in the upper graph (a) in Fig. 14!
- the feature of the first embodiment of the present invention is that the position of the article is narrowed down by using the result of analyzing the person movement history information that the person has moved in this way, and the specific processing thereof is as follows. This is performed by the article presence area estimation means 107 and the article position candidate weighting means 108 which are components of the first embodiment of the present invention. Details will be described later.
- the device operation detection device 105 receives the time information (operation time information) output from the timer means 120 and detects the device operation, and as a result, the identification information (ID) of each device detected and the device Information on operation status (operation information) and time information (operation time information) output from the timer means 120 at the time of operation are output to the device management database 106 every time device operation is detected.
- Equipment as used herein is defined as having a function of storing articles in a confined manner, and other items that are not articles are called “facility”.
- “Bookshelf” and “Refrigerator” are “Equipment”
- “Kitchen System” “Dining Table”, “Sofa 1”, “Sofa 2”, and “Low Table” are “Equipment”. It is.
- “kitchen system” can be further subdivided into “kitchen system” storage shelves and storages as “equipment”, and the cooking table as “equipment”.
- FIG. 15A and FIG. 15B are conceptual diagrams showing examples of devices in which the device operation detection device 105 is installed for devices having a storage function.
- a portion indicated as TSEN is a sensor as an example of the device operation detection device 105, and here, an example using the contact sensor TSEN is shown.
- FIG. 15A there are two Type of open / close type door 150, 151 force, contact force sensor TSEN installed on each of the door 150 and fixed support frame 153 force
- the sliding door type door 152 There is a force
- the contact force TS EN is installed on each of the door 152 side and the fixed frame 154 side.
- the force shown as an example of the contact sensor TSEN may be used, or a method other than the sensor may be used.
- the device management database 106 includes the identification information (ID) of each device detected by the device operation detection device 105, the operation status information (operation information) of the device, and the time output from the timer unit 120 during operation.
- Information (operation time information) is stored as device management information for each device operation detection.
- 16A and 16B are diagrams showing examples of device management information included in the device management database 106 of the article search system according to the first embodiment of the present invention.
- FIG. 16A and FIG. 16B give examples of “book shelf” and “refrigerator” as devices, each of which is independent device management information.
- As the contents of the device management information information on the time when the device was operated and information on opening / closing as operation status information (operation information) are described. This example is the simplest example.
- device management information should be prepared independently for each device. ! ⁇ .
- the article presence area estimation means 107 receives the article management information stored in the article management database 102 and the person management information stored in the person management database 104 (if necessary, the equipment management database 106 further stores the equipment management information stored in the equipment management database 106. (Also management information)
- the movement area is estimated to be the article existence area of the article to be searched. As described in the section “Means for Solving Problems”, the existence area of an article should be estimated with a focus on the fact that there should be an important correlation between the movement of a person and the place where the article exists. More concretely, it can be said that the position of an article can be estimated based on the following two principles.
- FIG. 17 is a flowchart showing the flow of processing in the article presence area estimation means 107. In the following, the flow of processing for estimating the presence area of an article will be described according to this flowchart.
- step S1701 the user designates a detection target article whose position is to be detected by the input device 109, and the detection target article designated by the input device 109 is input to the article presence region estimation means 107.
- the article presence area estimation means 107 refers to the article management information in the article management database 102 based on the designated detection target article, and the designated article is finally detected. Get time and location. Specifically, the article presence area estimation means 107 may refer to the last line of the article management information of the article (the line in which the latest information is described in the tabular article management information). This is a process performed in accordance with the above (Principle 2).
- the article existence area estimation means 107 refers to the person management information in the person management database 104, and a predetermined time including the time (the same time as the time and the time before and after the time). Among them, the article presence area estimating means 107 acquires all persons detected at the place as candidates for handling the designated article. The reason for searching not only for the person detected at the same time as the above time but also for the person detected before and after that time is the reason for the detection process itself by the article detection apparatus 101 and the person detection apparatus 103.
- the predetermined time may be made variable according to the identification information (ID) of the article or the identification information (ID) of the person. For example, if the person is an elderly person, or if the article is a heavy article or a large article, there is a possibility that the moving speed may be reduced. Therefore, the time width may be increased.
- the person and the article are a predetermined time including the time when the article was last detected by the person detection device and the article detection device, that is, the article was last.
- the person and the article are detected simultaneously by the person detection apparatus and the article detection apparatus, the case where the person and the article are detected at the same time as the time detected at Paraphrased as necessary.
- the article existence area estimation means 107 refers to the person management information in the person management database 104, and determines the time and the movement of the person.
- the person movement history information associated with is acquired.
- the movement path (in other words, the movement area of the person) is the estimation result of the article existence area.
- the result is in accordance with the above (Principle 1).
- the input device 109 is used to specify an article whose position is to be detected.
- the input apparatus 109 whether the article name is input by voice or a keyboard using a portable terminal or a personal computer.
- a conventional input method in the article search such as displaying the previous article list and selecting from the list may be used.
- the user is a father and a user is designated as a detection target article to be searched.
- a state in which the processing in the article presence area estimation unit 107 is performed will be specifically described.
- step S1701 the father detected the position using the input device 109.
- the product is input to the article presence area estimation means 107 so that juice is designated as the detection target article.
- step S1702 with reference to the last line of FIG. 5B which is the article management information on the juice in the article management database 102, the time “19:30” when the designated juice was last detected. And the location “living room” information are obtained by the article existence area estimation means 107.
- step S1703 the person management information in FIG. 11A and FIG. 11B of the person management database 103 is referred to by the article existence area estimation means 107, and within a predetermined time centering on the time “19:30”, All persons detected in the “living room” are acquired by the article presence area estimation means 107 as candidates for handling the designated article.
- the predetermined time is 3 minutes
- persons detected in the “living room” between 19:27 and 19:33 are identified from the person management information in FIGS. 11A and 11B of the person management database 104.
- the article existence area estimation means 107 only the “dad” detected in the place “living room” at time “19:30” is acquired by the article existence area estimation means 107.
- step S 1704 the history data “MF-Data02j” of FIG. 12 corresponding to the detected person management information is acquired by the article presence region estimation means 107.
- FIG. 18 shows the result of the article presence area estimation by the article existence area estimation means 107 obtained in this way in a tabular form. Although there is only one history data acquired in this example, if there are multiple people who may have moved with the article, all the history data corresponding to those people should be acquired.
- the article existence area estimation processing in the article existence area estimation means 107 is the article existence area estimation processing in the article existence area estimation means 107, and the history data obtained thereby is mapped onto the environment sketch by the article existence area estimation means 107 (for example, as shown in FIG. 13).
- Just displaying the lower plan view (b) image) on the display device 110 is useful enough.
- the user can look for his / her past movement path shown on the screen and search for an article while looking at the power to trace the path in order or the equipment or facilities in the vicinity of the path. That is why. Since the place where the target article is searched is narrowed down to the vicinity of the past movement route that is not in the whole room, it can be expected that the time and effort for searching will be greatly improved.
- the article position candidate weighting means 108 refers to the article management information stored in the article management database 102 or the person management information stored in the person management database 104 or the equipment management information stored in the equipment management database 106.
- the plurality of article existence areas estimated by the article existence area estimation means 107 are weighted so that the possibility that the article exists is high and the weight of the article existence area is high. In other words, it can be said that the article existence area having a higher possibility of existence is narrowed down from the article existence area estimation means 107 described immediately before, narrowing down the position of the article to some extent. In the first embodiment of the present invention, this low possibility of article existence is referred to as “weighting of article position candidates”.
- the first method is that the more slowly a person moves (if it stops in extreme cases)
- FIG. 19 shows the movement speed of the person by the article position candidate weighting means 108. It is the figure which showed the process which weights using a degree in the format of the flowchart.
- step S1901 since the history data of the person who handled the search target article can be obtained by the article presence area estimation means 107, the movement speed of the person is obtained by the article position candidate weighting means 108 using the history data. calculate. For example, in the history data in the example of FIG. 12, the moving speed can be calculated by calculating the distance moved by the position coordinate value force at each time in the unit time. By dividing this distance by the article position candidate weighting means 108 in the unit time, It can be easily obtained.
- step S1902 a group of places where the moving speed obtained by the article position candidate weighting means 108 is equal to or less than a predetermined value (threshold value) is extracted by the article position candidate weighting means 108 as a person's staying state. Then, the article position candidate weighting means 108 registers, for example, as a weighting list in the temporary storage unit in the article position candidate weighting means 108 or the article management database 102.
- a predetermined value threshold value
- step S1903 the article position candidate weighting means 108 first selects one place from the weighting list.
- step S1904 the article position candidate weighting means 108 at the place selected in step S1903,
- the article position candidate weighting means 108 may obtain, for example, an average of the position coordinates in the staying state below the threshold value.
- the first average moving speed is used as a weighting index
- the other center position coordinate is used when presenting the location of the article to the user.
- the calculation result is registered by the article position candidate weighting unit 108, for example, in the temporary storage unit in the article position candidate weighting unit 108 or the weighting list in the article management database 102.
- step S1905 for each location extracted in step S1902, if there is a location where step S1904 has not been performed yet, proceed to step S1906, and still perform step S1904! / If there is no place, go to step S1907. [0107] In step S1906, for each location extracted in step S1902, step S1904 is still performed! /, Na! /, One location is selected from the list, and step S1904 is repeated. Proceed to step S1905.
- step S1907 for all the locations extracted in step S1902, the article position candidate weighting means 108 increases the priority from the one with the lower average moving speed calculated in step S1904. , Weight. This is based on the idea that the more people are moving slowly, the more likely they are to stop!
- the weighting result is registered by the article position candidate weighting means 108, for example, in the temporary storage unit in the article position candidate weighting means 108 or the weighting list in the article management database 102.
- the process will be described using a specific example.
- the history data MF-Data02 of FIG. 12 is obtained by the article existence area estimation process in the article existence area estimation means 107, and the article position candidate weighting means 108 Let's take a look at the process of weighting the article position candidates.
- step S1901 the movement speed of the person is calculated by the article position candidate weighting means 108 from the person history data MF-Data02.
- the calculation result is shown in the above graph (upper graph (a) in FIG. 13).
- step S1902 a group of places whose moving speed obtained by the article position candidate weighting means 108 is equal to or less than a predetermined value (threshold value) is extracted by the article position candidate weighting means 108 as a person's staying state.
- the article position candidate weighting means 108 registers, for example, as a weighting list in the temporary storage unit in the article position candidate weighting means 108 or the article management database 102. In this example, it can be determined from the upper graph (a) in FIG. When these are related to the bottom plan view (b) plotted in the floor plan in Fig.
- the calculation result is registered by the article position candidate weighting unit 108, for example, in the temporary storage unit in the article position candidate weighting unit 108 or the weighting list in the article management database 102.
- step S 1907 the article position candidate weighting unit 108 weights “bookcase”, “refrigerator and kitchen system”, and “sofa 1” based on the calculation result.
- FIG. 20 shows the result of the weighting of the article position candidates thus obtained in a tabular form.
- the table shows the three stays extracted in order from the left
- the center position coordinates of the location (in this example, the location name indicating where the coordinates are in the room)
- steps S 1904 and S 1907 in FIG. 19 are respectively obtained by the article position candidate weighting means 108, for example, in the temporary storage unit in the article position candidate weighting means 108 or the weighting list in the article management database 102. Stored.
- the residence time of a person may be used together as information for weighting. This is based on the knowledge that when a person puts or picks up an article, the speed of movement of the person decreases or becomes zero, as well as the knowledge that it takes some time to handle the article. is there. For example, even when the average moving speed is minimum, if the staying state lasts for a short time, the candidate ranking force may be removed because it is highly likely that the article has not been handled.
- the threshold for this dwell time can be determined according to the ID and location of the article handler (detection target), the handling article (detection target article).
- the threshold value is increased.
- the staying time is long V (if the staying time on the sofa is long), there is a possibility that the operator may rest, so there is no need to use it as information.
- Weighting can also be performed according to the difference in average moving speed before and after the handler stays.
- the movement speed of the operator often decreases. Therefore, if the average moving speed in the predetermined time zone after staying is higher than the average moving speed in the predetermined time zone before staying, it is likely that the operator has released the carried item. be able to.
- An algorithm performed by the article position candidate weighting means 108 using this idea will be described with reference to the flowchart of FIG.
- steps S3701 to S3704 are the same as the processing of steps S1901 to S1904 in FIG.
- step S3705 the article position candidate weighting means 108 calculates average moving speeds V 1 and V 2 in a predetermined time zone immediately before and after the location.
- step 3705 the article position candidate weighting means 108 calculates average moving speeds V 1 and V 2 in a predetermined time zone immediately before and after the location.
- FIG. 38 shows a graph VH of a change in moving speed with respect to time for a person as an example of a handler.
- the article history candidate weighting means 108 determines that the movement history data corresponding to the time below the threshold TH of the movement speed is “staying”.
- the length of the staying time is used as the time width for determining the predetermined time zone immediately before and immediately after the staying by the product position candidate weighting means 108.
- it may be a predetermined value.
- step S3706 for each location extracted in step S3702, if there is a location where steps S3704 and S3705 by the article position candidate weighting means 108 have not yet been executed, the procedure proceeds to step S3707, where it has not been executed. If not, go to Step S370 8. [0120] In step S3707, the article position candidate weighting means 108 selects one of the above-described squirrel repulsive forces for each location extracted in step S3702 and has not performed steps S3704 and S3705, and performs steps S3704 and S3705. After repeating, go to step S3706.
- step S3708 for each location extracted in step S3702, (V — V
- the weights are weighted by the article position candidate weighting means 108 which increases the priority from the one with the larger value of).
- the absolute value of aft bef may be calculated by the article position candidate weighting means 108.
- the residence time is the residence time! If it is less than the value TH, it can be used by removing it from the candidate ranking.
- the evaluation S is large by using, for example, the following equation (Equation 1). This can be realized by setting the candidate rank in order by the article position candidate weighting means 108 in order.
- V average moving speed
- V average moving speed after residence time
- V average before residence time
- the uniform moving speed ⁇ , j8 is a weighting factor. Note that j8 is determined in advance by experiments.
- the graph of Fig. 39A shows the time change of the moving speed of two persons Hl and H2 as an example of the handler.
- Person HI is selected as a candidate for handling the search target article in step S1703.
- the selected person, person H2 is assumed to be a powerful person who has not been selected as a candidate for handling the search target article in step S1703.
- TH is a threshold for the moving speed for determining whether or not the force is staying by the article position candidate weighting means 108. Both the persons HI and H2 are staying at a certain time at a moving speed below the threshold TH.
- the article position candidate weighting means 108 can determine this.
- Figure 39B shows the movement path of two persons HI and H2 superimposed on the floor plan of the room.
- both persons HI and H2 are near sofa 1 in Figure 39B. Is determined by the article position candidate weighting means 108. That is, the person position HI and H2 are staying at the same time and in the same place, and at this time, the article position candidate weighting means 108 determines that the delivery of the article may have occurred. Since the person HI is a candidate for handling the article to be detected, the article position candidate weighting means 108 determines that the person HI may have passed to H2. If person H2 is not a candidate for handling, it will be necessary to newly handle the candidate as a handling candidate, and there is a possibility that the search target article exists even in the movement route after the delivery of person H2. This is determined by the article position candidate weighting means 108.
- step S4001 The processing from step S4001 to step S4003 is the same as the processing from step S1701 to step S1703.
- step S4001 the user designates a detection target article whose position is to be detected by the input device 109, and the search target article designated by the input device 109 is input to the article presence region estimation means 107.
- step S4002 the article presence region estimation means 107 refers to the article management information in the article management database 102 based on the designated search target article, and the designated article is finally detected. Get the time and location (position in the room).
- FIG. 41 shows the detection results at the place where the designated article was last detected in time series, and the time when the designated article was last detected is Tx. Is shown.
- the article presence area estimation unit 107 refers to the person management information in the person management database 104 and is detected at the place within the predetermined time zone TZ1 including the time Tx. All persons are acquired as candidates for handling the designated article (hereinafter referred to as handling candidates).
- handling candidates are acquired as candidates for handling the designated article.
- person ⁇ 3 enterers the room to be detected at time T1
- person ⁇ 4 enters the room at time ⁇ 2 are acquired as goods handling candidates.
- step S4004 the article presence region estimation means 107 refers to the human management information in the human management database 104, and the candidate (person ⁇ 3, human ⁇ 4) force acquired in step S4003. Detected at the location acquired in step S4002 between the time detected at the first time (ie, when entering the room) (in the time zone TZ1) and the time detected at the second time (ie when exiting) All persons are acquired as candidates who have received delivery of goods (hereinafter referred to as delivery candidates), and a delivery candidate list is created.
- delivery candidates who have received delivery of goods
- the article presence area estimation means 107 is the candidate (person ⁇ 3, person ⁇ 4) detected during the time zone TZ1, and the time when the person ⁇ 3 last exited during the time zone ⁇ 2 until the time ⁇ 5.
- the detected person ⁇ 5 (regardless of entering or leaving the room) is acquired as a delivery candidate of the article and put in the delivery candidate list.
- step S4005 the article presence region estimation means 107 stores all of the handling candidates acquired in step S4003 and all the delivery candidates acquired in step S4004 in the human management database 104.
- the person management information person movement history information
- the person movement history information history data in which the time and the person movement are associated with each other is acquired.
- step S4006 the moving speed of the person (handling candidate) is calculated by the article presence area estimating means 107 from the history data of the handling candidate acquired in step S4005.
- the calculation of the moving speed is the same as the processing in step S1901, and thus the description is omitted.
- step S4007 the article presence area estimation means 107 registers the combination data in which the moving speed of the handling candidate is equal to or lower than (predetermined time, place) as a staying point in the staying point list.
- the element in other words, the movement history information element
- 3D data time, X coordinate value, Y coordinate value
- step S4008 the article presence area estimation means 107 selects one candidate from the delivery candidate list created in step S4004.
- step S4009 the article presence region estimation means 107 refers to the history data acquired in step S4005 for the candidate selected in step S4008, and the previous reporter (selected candidate) Calculate the moving speed.
- the calculation of the moving speed is the same as the process in step S1901, and thus the description thereof is omitted.
- step S4010 the article presence region estimation means 107 extracts all the combination data (time, place) where the moving speed of the candidate is equal to or less than a predetermined value as a stay point, and Each is compared with each of the residence points registered in the residence point list. In this comparison, the article existence region estimation means 107 determines whether or not the locations at the same time are within a predetermined distance. If there are locations where the locations at the same time are within the predetermined distance, Proceed to S4011. If there are no locations within the predetermined distance at the same time, the process proceeds to step S4012.
- the predetermined distance can be determined based on a criterion for determining whether or not the distance between the persons is that the article can be delivered.
- the length of a person's arm can be used as a reference because the arm is extended when the person delivers the article.
- the average length of a person's arm can be used, and if a person ID is acquired by the person detection device 103, it is stored in the person management database 104 based on the acquired ID. It can be determined based on the length of the person's arm.
- an error may be superimposed on the sensing result of the person's position, and therefore the threshold value may be determined in consideration of the error.
- the time when recording the movement history is synchronized between a plurality of persons has been described. If the time is not synchronized, the time difference is a predetermined threshold value (the time is It is only necessary to compare data within a range that can be regarded as.
- step S4011 the article presence area estimation means 107 compares the history data of the selected candidate with the stay points registered in the stay point list, and the location person at the same time has a predetermined location. If there are multiple dwell points within the distance, the dwell point with the earliest time A non-use flag is assigned to all previous history data. This is a result of reflecting the knowledge that the search target article does not exist in the place before the time when the delivery may have occurred in the delivery candidate history data, in the article presence area estimation process. It is. Thereafter, the process proceeds to step S4013.
- step S4012 the article presence area estimation means 107 deletes the currently selected candidate from the delivery candidate list. This is because the history data of the currently selected candidate does not have the same location and time as the history data of the handling candidate, so that the knowledge that the delivery of the article does not occur is reflected in the article existing area estimation process. It is a thing. Thereafter, the process proceeds to step S4013.
- step S4013 the article presence area estimation unit 107 checks whether or not the processing of S4010 has been completed for all the persons registered in the candidate list. If Yes (when the process of S4010 is completed), the present article presence area estimation process is terminated. If No (the process of S4010 is not completed), the process proceeds to step S4014.
- step S4014 the article presence area estimation means 107 selects one person who has not yet been processed in step S4010 from the delivery candidate list for each person registered in the delivery candidate list, and step S4009. Return to.
- the above is the article existence area estimation processing in the article existence area estimation means 107 when the delivery of the article is taken into account, and the history data of the handling candidates and the delivery candidates obtained thereby is the environment.
- the display device 110 presents the result of mapping by the article presence area estimation means 107 on the floor plan, and the user searches for the detection target article in the vicinity of the movement route (in other words, the movement area of the person). Can do.
- FIG. 42A, FIG. 42B, and FIG. 42C show examples of data and display results used in the article presence area estimation process.
- FIG. 42A shows data representing changes in the moving speed of two persons HI and H2 (the solid line in FIG. 42A is data for person HI and the dotted line is data for person H2).
- Person HI is a person who is selected as a candidate for handling goods in step S4003
- person H2 is a power who is not selected as a candidate for handling articles in step S4003.
- Step S 4004 Selected as a delivery candidate.
- TH is a threshold relating to the moving speed for determining whether or not the staying force is detected by the article position candidate weighting means 108.
- the data contained in the time zones T, T, and T are the residence points,
- FIG. 42B is a diagram in which the movement trajectory data of the two persons HI and H2 are also superimposed on the floor plan of the room.
- step S4010 if the article existence region estimation means 107 determines that the stay points of the persons HI and H2 included in the place group P corresponding to the same time are within a predetermined distance from each other, The movement history data is also finally selected by the article presence area estimation means 107 as there is a possibility that the article exists.
- the movement trajectory presented on the display device 110 is as shown in FIG. 42C.
- the display device 110 can hide the data so that there is no possibility of the presence of the article! You can avoid presenting it to the user.
- the movement path of a person who may have received the delivery of goods (in other words, the person's movement area) is also displayed at the same time. It is also possible to search for a movement route (in other words, another person's movement area).
- weighting of article position candidates in consideration of the possibility of delivery can be performed.
- the weighting process in the article position candidate weighting means 108 considering the possibility of delivery will be described using the flowchart of FIG. Note that, unless otherwise specified, the operation subject of each step is the article position candidate weighting means 108.
- step S4301 the article position candidate weighting means 108 Prepared by the area estimation means 107 (flow chart in Fig. 40) and the staying point data of the handling candidate and the staying point data of the delivery candidate (excluding those with a non-use flag) Data sets that are human data and are close in location and time are registered in the weighted list as stagnant status.
- the weighting list is provided, for example, in the temporary storage unit in the article position candidate weighting means 108 or the article management database 102. To determine whether the place and time are close to each other, for example, whether the distance between the places is within a predetermined threshold, and whether the time difference between the times is within a predetermined threshold. Please use it.
- step S4302 the article position candidate weighting means 108 first selects one staying state from the weighting list.
- step S4303 the article position candidate weighting means 108 enters the staying state selected in step S4302, and
- an average of the position coordinates in the staying state below the threshold value may be obtained by the article position candidate weighting means 108.
- the first average moving speed is used as a weighting index, and the other center position coordinate is used when presenting the location of the article to the user.
- the calculation result is registered in, for example, the temporary storage unit in the article position candidate weighting unit 108 or the weighting list in the article management database 102.
- step S4304 the article position candidate weighting means 108 proceeds to step S4305 if there is a staying state that has not yet been performed in step S4303 for each staying state registered in the weighting list. If step S4303 has not been performed yet and there is no staying state, the process proceeds to step S4306.
- step S4305 the article position candidate weighting means 108 selects one staying state that has not yet been subjected to step S4303 for each staying state registered in the weighting list from the weighting list, and Return to S4303.
- step S4306 the article position candidate weighting means 108 For all the staying states registered in the weighting list, weighting is performed so that the priority is higher from the lowest average moving speed calculated in step S4303. This is based on the idea that the more slowly a person is moving (in the extreme case, when the person is stopped), the more likely that the person is handling the goods there.
- the weighting result is registered in, for example, the temporary storage unit in the article position candidate weighting unit 108 or the weighting list in the article management database 102.
- FIG. 44A, FIG. 44B, and FIG. 44C show examples of data and display results used for weighting processing by the article position candidate weighting means 108 in consideration of the possibility of delivery.
- 44A and 44B are the same as FIGS. 42A and 42B, respectively.
- FIG. 44C shows the priority order obtained by executing the processing in step S4306 at the corresponding location.
- ⁇ average movement speed in place group P ⁇ average movement speed in place group R ⁇ average movement speed in place group Q ''.
- Place group P has the highest ranking, followed by place group R and place group Q.
- place group R is If the item HI is not included in the movement history of the person HI, but the delivery of the article is considered, the vicinity of the place group R on the movement history of the person H2 who is the candidate for delivery of the article is also selected (the place group R Is selected because the person HI is moving in the vicinity of the place group R, though it is not in the movement history of the person HI).
- the above is the article presence area estimation process in the article position candidate weighting means 108 when taking the article into consideration.
- the result of mapping the article existence area obtained by the article existence area estimation means 107 on the sketch of the environment (for example, FIG. 44C) is displayed on the display device 110, so that the route the user has moved is displayed.
- the goods can be searched not only for the above candidate locations but also for candidate locations on the route of the person who may have been delivered.
- FIG. 45A shows data representing the time change of the moving speed of two persons HI and H2 (the solid line in FIG. 45A is the data of person HI and the dotted line is the data of person H2).
- TH is a threshold value regarding the moving speed for determination by the staying and article position candidate weighting means 108.
- the article position candidate weighting means 108 determines that the time zones S2, S4, and S5 are in a staying state. Time zones Sl and S3 will be described later.
- FIG. 45B shows the average moving speeds V and V of the persons HI and H2 in the time zones S1 to S5 based on the graph of FIG. 45A.
- Fig. 45C shows the movement trajectories of the two persons HI and H2 superimposed on the floor plan of the room.
- the article position candidate weighting means 108 can determine that the locations of the humans HI and H2 in the time zone S2 correspond to the location group P in common. Since there is a time zone S2 between the two people that is in the staying state that overlaps at the time and place, there is a possibility that the person HI was transferred to the person H2 in the time zone S2 in the staying state. It can be determined by the article position candidate weighting means 108.
- step S4601 the article position candidate weighting means 108 acquires the handling candidate's stay point data and the delivery candidate's stay obtained by the article presence area estimation means 107 (flowchart in FIG. 40). Prepare point data (excluding items with a non-use flag) and use the same person's data and the location and time close to each other as a staying state. By 108, it is registered in the weighting list. Since this processing is the same as the processing in step S4301 in the flowchart of FIG. 43, description thereof is omitted.
- step S4602 the history data of all handling candidates and delivery candidates corresponding to the stay point data acquired in S4601 is acquired by the article position candidate weighting means 108. Since the history data acquisition is the same as step S4005 in the flowchart of FIG. 40, the description thereof is omitted.
- step S4603 the article position candidate weighting means 108 selects one staying state from the weighting list created in step 4601. This process is the same as step S4302 in the flowchart of FIG.
- step S4604 the staying state selected in step S4603 by the article position candidate weighting means 108 is as follows.
- step S4605 the article position candidate weighting means 108 determines whether or not the selected staying state is common to the handling candidate and the delivery candidate. In this determination, it is possible to use whether or not the stay point included in both stay states satisfies the determination criterion used in step S4010 in the flow chart of FIG. If it is determined by the article position candidate weighting means 108 that the handling candidate and the delivery candidate are common, the process proceeds to step S4606, and if it is determined that they are not common, the process proceeds to step S4607.
- step S4606 the article position candidate weighting means 108 calculates the average moving speed before and after the time period of the staying state in the handling candidate and delivery candidate history data.
- the history data in Fig. 45 change in movement speed over time
- the history data of person HI and person ⁇ 2 is the same as the length of time zone S2, which is a common staying state.
- the time zone S1 is set by the article position candidate weighting means 108 before and after the time zone S2, and the time zone S3 is set after the time zone S2.
- the length of these time zones may be a predetermined value that is not the same as the length of the common residence time zone S2.
- the article position candidate weighting means 108 calculates the average moving speed of the handling candidates and all the delivery candidates (persons HI and H2 in FIG. 45A) in the set time zones S1 and S3. The calculated results are shown in Figure 45B as V, V,
- V is calculated by the article position candidate weighting means 108, and similarly to the person H2
- the average moving speeds in the predetermined time zones S1 and S3 immediately before and immediately after the time zone S2 in the staying state are the article position candidate weighting means 108 as V and V, respectively.
- step S4607 the article position candidate weighting means 108 determines whether or not the processing of S4604 has been completed for each staying state registered in the weighting list. When the processing power of S4604 is finished! / When the process is completed, go to step S4609. If the processing of S4604 is finished, go to step S4608. [0168] In step S4608, the article position candidate weighting means 108 selects one staying state that has not yet been processed in S4604 for each staying state registered in the weighting list from the weighting list. Return to S4604.
- step S4609 the article position candidate weighting means 108 weights each staying state registered in the weighting list using the average moving speed and the average moving speed before and after the common staying state. I do.
- step S4306 in Fig. 43 described earlier the ranking is based on the one with the smallest average moving speed that does not distinguish between handling candidates and delivery candidates. Here, in addition to the average moving speed, ranking is performed in consideration of the possibility of delivery.
- the average moving speeds in the time zones S2 and S4 in the staying state for the person HI are V and V, respectively, and in the time zones S2 and S5 in the staying state for the person H2.
- the evaluation value Z (V -V) is the possibility that the goods were delivered from person H 1 who is a candidate for handling goods to person H 2 who is a candidate for delivery of goods in band S 2 + (V
- the article position candidate weighting means 108 determines that there is a high possibility that the article has been delivered from the person HI to the person H2.
- Time zone in which people HI and H2 are in a common staying state Each person's staying state after S2 (in the case of Figure 45A, time zone S4 in which the person HI stays and time zone S5 in which the person H2 is staying.
- the evaluation value Z is reflected by the article position candidate weighting means 108.
- the person HI who is a candidate for handling may have delivered the article to the person H2 with respect to the average moving speed V in a certain staying state after the time zone S2 which is a common staying state.
- the V force is also reduced by the article position candidate weighting means 108 by a constant a times the evaluation value Z. Delivery candidate Because the person H2 who is a person may have received the goods from the person HI,
- the article position candidate weighting means 108 adds a constant y times the evaluation value Z. Therefore, the scores for the three states of the time zones S2, S4, and S5, which are staying states that are the existence region candidates of the article,
- the score is calculated using the average moving speed V of the article handler HI as the article position candidate weight.
- the article position candidate weighting means 108 can perform ranking in consideration of the delivery possibility of articles as well as the average moving speed in the staying state.
- FIG. 21 is a conceptual diagram showing an example in which the article search result is displayed on the display device 110 by CG (computer graphic).
- Figure 21 shows the entire location of the goods. In the bird's-eye view, the place where the article may actually exist is overwritten with the numerical value indicating the possibility (rank obtained as a result of the article position candidate weighting). .
- the user looks at the result on the display device 110 and first searches for the vicinity including the “bookcase”. If the user does not find it, the user next searches for the “refrigerator and kitchen system”. As in the vicinity of “1”, it is possible to narrow down the search place in advance and efficiently search for articles.
- FIG. 22 is a flowchart showing the process of weighting using the operation information of the device.
- device management information for managing the location of the device is prepared in advance in order to search for the location of the device that is sufficiently close to the position of the person.
- Fig. 23 is a table showing the database that manages the location of equipment (including facilities) in tabular form. The table shows how the positions of the devices in the environment are simply managed in a rectangular shape, and the upper left and lower right position coordinates of the rectangle are given.
- the occupied area of the device may be determined by a polygon, and the vertex may be represented by vector data. Oh ,.
- Step S2201 to Step S2203 are the same as Step S1901 to Step S1903 in the flowchart of FIG. .
- step S2201 the movement speed of the person is calculated by the article position candidate weighting means 108 from the history data of the person who handled the search target article.
- step S2202 the article position candidate weighting means 108 is used.
- a group of places whose moving speed is equal to or less than a predetermined value (threshold value) is extracted by the article position candidate weighting means 108 as a person's staying state, and the article position candidate weighting means 108, for example, in the article position candidate weighting means 108 Is registered as a weighted list in the temporary storage unit or the article management database 102.
- step S2203 the article position candidate weighting means 108 first selects one place from the weighting list.
- step S2204 the article position candidate weighting means 108 uses the article position candidate weighting means 108 to determine the time at the location and the center position coordinate of the place at the location selected in step S1903. calculate.
- the article position candidate weighting means 108 searches for a device group within a predetermined range with respect to the center position.
- a device group within a predetermined range with respect to the center position.
- FIG. Fig. 24 is a sketch showing the equipment placed in the environment, and the X in Fig. 24 is the center position.
- the device area is included in the circle.
- the device may be extracted by the article position candidate weighting means 108 as a search result.
- the only device that has been sought by the article position candidate weighting means 108 is the “book shelf”. Of course, if a plurality of devices are found by the article position candidate weighting means 108, all of them are extracted.
- step S2206 if at least one device is found, the process proceeds to step S2207, and if not found, the process proceeds to step S2208.
- step S2207 the device management information is referred to by the article position candidate weighting means 108 for each device that has looked at, and there is an overlap between the time calculated in step S2204 and the time until the device is opened and the force is also closed.
- the equipment that does not exist is deleted by the article position candidate weighting means 108 from, for example, the temporary storage unit in the article position candidate weighting means 108 or the weight list in the article management database 102 in which the weighting result is stored. This is a process that reflects the fact that even if a person stays in a certain location, if there is a powerful device that does not store the item, that device is not likely to be present. Yes, it is possible to narrow down the devices where people actually stay and store articles.
- step S2208 if all the place groups extracted in step 2202 have been completed, the process proceeds to step S2210, and if not, the process proceeds to step S2209.
- step S2209 for each location extracted in step S2202, the location not yet stored in step S2204 is processed by the article position candidate weighting means 108, for example, the temporary storage unit or the article in the article position candidate weighting means 108.
- the article position candidate weighting means 108 for example, the temporary storage unit or the article in the article position candidate weighting means 108.
- One is selected from the weighting list in the management database 102, and Step S2204 is repeated.
- step S2210 for all devices extracted in step S2207, according to the distance between the person's staying position and the device, the priority is set higher as the distance force S is smaller by the product position candidate weighting means 108. Weight as follows.
- step S2201 the movement speed of the person is calculated by the article position candidate weighting means 108 from the person history data MF-Data02. The calculation result is shown in the above graph (upper graph (a) in FIG. 13).
- step S2202 a group of places whose moving speed obtained by the article position candidate weighting means 108 is equal to or less than a predetermined value (threshold value) is extracted by the article position candidate weighting means 108 as a person's staying state.
- the article position candidate weighting means 108 registers, for example, as a weighting list in the temporary storage unit in the article position candidate weighting means 108 or the article management database 102. In this example, it can be determined that there are three staying parts. When these are related to the floor plan on the lower side of Figure 13 plotted in the floor plan, they stay in the vicinity of “Bookcase”, “Refrigerator and Kitchen System”, and “Sofa 1” in order from the earliest time.
- each staying state is called “book shelf”, “refrigerator and kitchen system”, and “sofa 1”.
- the time of each staying state is as follows.
- step S2203 by performing the processing from step S2203 to step S2209, two devices, "bookshelf” and “refrigerator”, which have the possibility of the presence of an article, are obtained.
- the specific processing process is shown below.
- step S2204 the article position candidate weighting means 108 calculates the staying time and center position coordinates in the "book shelf”, "refrigerator and kitchen system", and "soft 1".
- step S2205 a device group within a predetermined range is searched for the center position by the article position candidate weighting means 108, and "book shelf” and “refrigerator” are searched by the article position candidate weighting means 108. Extract.
- step S2206 since at least one device has been found, the process proceeds to step S2207, and the device management information is referred to by the article position candidate weighting means 108 for each device that has been found. Open / close operation with the time calculated in step S2204 If there is no pair, the device is deleted by the article position candidate weighting means 108 (steps S2208 to S2204 to S2208).
- step S2210 as a result of the above processing, two devices of "book shelf” and “refrigerator” were obtained as juice storage locations, so depending on the distance from the position where the device stayed near the device. These two devices are weighted by the article position candidate weighting means 108.
- FIG. 27 is a conceptual diagram showing an example in which the article search result is displayed in CG (computer graphic) on the display device 110.
- CG computer graphic
- the bird's-eye view showing the entire location of the article is displayed by color-coding the equipment as the place where the article may actually exist, and the numerical value indicating the high ranking is overwritten. It shows how it is.
- the difference from Fig. 21 is that the location of goods is narrowed down to the equipment level. If there is a plurality of storage parts in one device, and device management information can be stored for each storage part, it is possible to further narrow down this process.
- “Refrigerator” shown as the second candidate in the example of FIG. 27 distinguishes and displays any storage partial force of “Refrigerator” that is not a single device (in FIG. 27).
- the top storage part of the refrigerator is displayed. This allows the user to narrow down the search place and search for articles more efficiently.
- the article presence area estimation means 107 and the article position candidate weighting means 108 are the major features of the first embodiment of the present invention.
- the position of the article can be narrowed down using the result of analyzing the operation information.
- the first other useful method is that when an area where an article exists is estimated by the article existence area estimation means 107, the article can be placed in the area. This is a method of weighting the places with the characteristics in descending order (not shown).
- normal location information that stores one or more places where each item is normally placed is also stored in advance in the item management database 102 as item management information. In some cases, a person handles the item. Update the normal location information according to the situation, and use the updated normal location information.
- the second other useful method is a method of using the article owner information on the owner of an article by the article existence area estimation means 107, and the history acquired by the article existence area estimation means 107. This is particularly effective when the data is for multiple people.
- the fact that historical data for multiple people has been acquired means that a certain article has been brought to a location and at the same time multiple people have entered the location. In this case, it is impossible to distinguish which person has the goods, so it is possible to obtain historical data of all people. If any of the multiple people is likely to have the goods If the product owner information is obtained in advance, the acquired history data itself can be weighted with certainty.
- the article owner information is, for example, as shown in FIG.
- the narrowing-down method may be as follows.
- the historical data for two people shown in Fig. 14 is obtained, and these are the historical data for ⁇ Dad '' and ⁇ Mom ''.
- this is a “technical book”.
- the location of the “technical book” is not 100% of the “dad” who has a “technical book” that does not examine all the historical data for two people. It is only necessary to take out history data and weight it. More specifically, in step S1907 in the flow of FIG. 19, the possibility of who owns the value obtained by simply weighting the priority higher from the lower average moving speed. Just add a value and weight it!
- the weight value obtained for the historical data of “dad” is multiplied by 100%.
- the weighted value obtained from the historical data is multiplied by 0% (ie, no weighting), which is the same as being excluded from the candidates.
- the following effects can be obtained.
- the article detection apparatus 101 detects an article, and stores the detected article location information and time information in the article management database 102.
- the person detection device 103 detects a person and stores the detected person location information and time information in the person management database 104. Further, the device operation detecting device 105 detects the device operation of the device that stores and manages the articles, and stores the detected operation information of each device in the device management database 106.
- the article presence area estimation means 107 refers to the information stored in the article management database 102 and the information stored in the person management database 104 (also refer to the information stored in the equipment management database 106 as necessary). As the article presence area, a place where an article may exist is estimated on a route traveled by a person.
- the article position candidate weighting means 108 can use information such as the movement of a person and knowledge of the place where the article is normally placed, so that it can exist among places where the article may exist. It is possible to display the information on the display device 110 by weighting the high and low characteristics.
- the present invention is not limited to the first embodiment, and can be implemented in various other modes.
- a simple example of a display where a user searches for an article is shown with a location on a bird's-eye view of the entire environment.
- it is also effective to be able to search for articles by looking at these display screens.
- an article search system according to the second embodiment of the present invention an example using images as other screen display method examples. Indicates.
- the display method in the article search system shown below is an example of the display method based on such an idea, and relates to the article search system according to the eleventh aspect of the present invention. More specifically, if the time and place where the article was handled in the past can be detected, the video of the place at that time can be shown to remind you how the article was handled at that place at that time. Let t be realized.
- an imaging apparatus 111 that captures an environment in which the article search is performed; Using the image database 112 that stores the photographed image information and the processing result of the article existence area estimation means 107 or the article position candidate weighting means 108, it is estimated that the article requested to be searched is placed.
- the image search means 113 for capturing the estimated place and time, photographing the estimated place, and extracting image information including the estimated time from the image database 112, and the image search means 113
- a configuration is shown that includes a display device 110 that displays image information together with information from the article presence region estimation means 107 or the article position candidate weighting means 108 as necessary.
- the imaging device 111 captures an environment in which the article search is performed.
- a camera using an imaging device such as a CMOS or a CCD is generally used in practice, but a special camera such as near infrared may be used depending on the location.
- a plurality of imaging devices may be prepared, and the camera used for the human detection device 103 may be used in combination. I do not care.
- information is added to indicate which position in the real world the image captured by each image corresponds to! /. As the simplest description mode of this information, the coordinates of each pixel of each imaging device and the real world are
- each imaging device 110 is associated with vector data of the range of the floor surface being copied,
- Corresponding Method 2 if one vector data represents one floor surface area, a force that requires at least one vector data for one image pickup device, for example, two or more in one image of one image pickup device 111 If there is a floor area, a plurality of areas may be prepared as necessary. Also, if the real-world model is known, and the position and orientation of the camera in the real world are known, the information power is also calculated to obtain the real-world coordinates of each location in the real world reflected in the camera. It is also possible. Note that information about which position in the real world the image captured by the imaging device corresponds to should be changed once it is first created unless the imaging device is powered! /.
- an image information power time stamp taken by each imaging device is added (according to time information output from the timer means 120) and stored.
- Picture The image may be a moving image or a still image, and may be used depending on the performance of the system.
- FIG. 28 is a conceptual diagram showing how a moving image is accumulated with a time stamp. Fig. 28 shows an example in which time stamps are given every minute. Of course, it can be determined according to the required system specifications.
- the image search means 113 obtains the place and time when it is estimated that there is an article requested to be searched, estimated by the article presence area estimation means 107 or the article position candidate weighting means 108, and then the place is photographed.
- the image information including the time is extracted from the image database 112 and displayed on the display device 110.
- FIG. 29 is a flowchart showing the flow of the image search process in the image search means 113. The flow of the process will be described below with reference to this figure.
- step S2901 the location and time estimated that there is an article requested to be searched, which is estimated using the processing result in the article position candidate weighting means 108, is obtained.
- step S2902 image information including the location and time included in the information or data is searched and downloaded for all information or data acquired.
- the method of selecting the imaging device including the place differs depending on the description mode of the information about the imaging device and which position in the real world the image captured by the imaging device corresponds to.
- image information including the time is retrieved from the image database 112 that stores image information captured by the image capturing devices. ,to download.
- image information to be downloaded include For example, moving image information including 2 minutes before and after the time or one still image information per second should be determined according to the required specifications of the system.
- Fig. 3 OA and Fig. 30B are conceptual diagrams showing how to retrieve and display the image information from the image database 112 of Fig. 28 at 19:31 and 19:32 respectively.
- the image information may be displayed suddenly or in stages.
- To display in stages first, only the processing result of the position candidate weighting means is displayed. Specifically, for example, the screen as shown in Fig. 21 or Fig. 27 is displayed. Then, the position candidate where the article shown there is specified is designated, and related image information is displayed.
- the flow of the article position estimation process in the article position estimation apparatus in the first and second embodiments of the present invention exemplified in the first and second embodiments is merely an example. It is not limited to. In other words, if the process includes the concept of using historical data including human movement speed and equipment operation information for the estimation of the article position, the flow of the process may be different or the required data description mode may be different. It doesn't matter if they are different.
- the article position estimation device includes an input device 109, an article detection device 101, a human detection device 103, among the devices constituting the article position estimation device,
- the device operation detection device 105 and the display device 110 may be installed in an optimum place according to the system requirements.
- the article management database 102, the person management database 104, and the equipment management database 106 may be arranged near the article detection device 101, the person detection device 103, and the device operation detection device 105 that acquire information stored therein.
- the article detection device 101, the human detection device 103, and the equipment operation detection device 105 are arranged in the vicinity of the article presence region estimation means 107 and the article position candidate weighting means 108, which are the remaining means of the apparatus. Send the information to each database via the network. Even if V is different, there is no particular restriction on the installation location of each means or device as long as it is installed in the optimal location according to the system requirements.
- the article search system of the second embodiment of the present invention is an image search means 113, an image database 112, and an imaging device, which are other components other than the device or means described above as the components of the system. There is no particular restriction on the installation location, etc., as long as 111 is installed in an optimal location according to the system requirements.
- the article search system of the first and second embodiments of the present invention has been described mainly for home use, but of course, it is not limited to home use, for example, used in offices, etc. You can do it.
- FIG. 34 is a sketch showing a state in which a tag reader TGR as an example of the article detection apparatus 101 is installed in an office.
- a gate-type tag reader TGR is installed at the entrance 200 of the room 205 where the office is located, and the room 205 has a work space 202 separated by a party 201 for eight persons.
- a tag reader TGR is installed in the open / close section of each bookcase 203 installed in the work space 202 (similar to FIG. 15B). The user can detect that the article is in the room 205 by bringing the article into the room 205.
- a human detection device such as a camera 204 or a floor sensor capable of distinguishing and detecting human movements
- historical data of the human 206 can be obtained thereby (for example, FIG.
- the solid line with an arrow shown in 34 is an example of the detection result of the person 206)
- a room 205 is obtained. This makes it possible to estimate the location of the article where the article brought in is likely to be in the room 205 with a weight.
- the force at which an article is placed can be specified to some extent by combining the detection result of the article with human history data.
- an article has an owner, and the article is considered to be the place where the owner carries it or the force in the owner's work space 202 is correct. Therefore, conversely, the entrance to room 205 indicates that such an item was taken out by another person.
- tag readers TGR detect it, it is possible to prevent goods from being stolen by determining that this is an illegal take-off and taking security measures such as sounding a buzzer.
- a part of the article search system including the article existence area estimation means 107 excluding the various detection devices may be configured such that different articles are distinguished by a computer as shown in the flowchart of FIG.
- Step S3001 for storing the article detection location information and time information detected in the article detection apparatus in the article management database, and a person detected by the person detection apparatus by distinguishing a person's position for each individual.
- Step S3002 for storing the movement history information of the person management database, the information stored in the article management database and the person management database, the movement history information of the person management database, and the detection location of the article management database
- Step S 3003 is performed for associating the person with the article based on the time and estimating an article existence area of the article.
- CD- ROM keep recording medium serial is recorded, such as, if desired, can also be used in reading also CD- ROM force.
- a display that is an example of a display device, a keyboard that is an example of an input device, a hard disk and a memory that can store the various databases and the various means, for example, a CD-ROM drive Is connected to a hard disk via a CD-ROM drive, the system for estimating the article position is recorded on a hard disk via a CD-ROM drive.
- the article search system may be executable.
- the date and time may be used instead of the power time using the time information.
- the article position estimating operation according to the present invention can be performed across different days.
- the article presence area estimation means may determine the movement area of the detected person after the case where the person and the article are simultaneously detected by the person detection device and the article detection device. This is estimated as the article presence area of the article, but after the detected person has exhausted the room force, the estimation of the article existence area of the article is stopped. Please do it.
- An article position estimation apparatus, article position estimation method, article search system, and article position estimation program according to the present invention include an article position estimation apparatus and article position estimation for managing articles in a general house or office.
- the present invention relates to a method, an article search system, and an article position estimation program, in particular, positions of various articles such as daily goods used in daily life in general households and portable articles used in offices, RFID tag technology, etc.
- the user can appropriately present the location of the search object when inquiring about the search object, making it easier to search for articles than in the past and greatly reducing the effort.
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Quality & Reliability (AREA)
- Entrepreneurship & Innovation (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Development Economics (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
Description
Claims
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2006519678A JP4006471B2 (ja) | 2005-04-01 | 2006-03-17 | 物品位置推定装置、物品位置推定方法、物品検索システム、及び物品位置推定用プログラム |
| US11/796,047 US7545278B2 (en) | 2005-04-01 | 2007-04-26 | Article position estimating apparatus, method of estimating article position, article search system, and article position estimating program |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2005-105923 | 2005-04-01 | ||
| JP2005105923 | 2005-04-01 |
Related Child Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| US11/796,047 Continuation US7545278B2 (en) | 2005-04-01 | 2007-04-26 | Article position estimating apparatus, method of estimating article position, article search system, and article position estimating program |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2006109423A1 true WO2006109423A1 (ja) | 2006-10-19 |
Family
ID=37086714
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/JP2006/305401 Ceased WO2006109423A1 (ja) | 2005-04-01 | 2006-03-17 | 物品位置推定装置、物品位置推定方法、物品検索システム、及び物品位置推定用プログラム |
Country Status (3)
| Country | Link |
|---|---|
| US (1) | US7545278B2 (ja) |
| JP (1) | JP4006471B2 (ja) |
| WO (1) | WO2006109423A1 (ja) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2010140754A (ja) * | 2008-12-11 | 2010-06-24 | Panasonic Electric Works Co Ltd | 照明システム |
| US7852217B2 (en) | 2005-12-28 | 2010-12-14 | Panasonic Corporation | Object detecting device, object detecting method and object detecting computer program |
| WO2011108055A1 (ja) * | 2010-03-03 | 2011-09-09 | パナソニック株式会社 | 物体位置推定装置、物体位置推定方法、及び、物体位置推定プログラム |
| JP2014108892A (ja) * | 2012-12-04 | 2014-06-12 | Fujitsu Ltd | 物品の配置位置管理装置、プログラム及び方法 |
| JP2015155345A (ja) * | 2014-02-20 | 2015-08-27 | 大和ハウス工業株式会社 | 収納管理システム |
| JP2018073012A (ja) * | 2016-10-26 | 2018-05-10 | 株式会社東芝 | 管理システム |
| JP2023110357A (ja) * | 2022-01-28 | 2023-08-09 | 積水ハウス株式会社 | 情報処理装置 |
| WO2024195059A1 (ja) * | 2023-03-22 | 2024-09-26 | 日本電気株式会社 | 移動履歴管理装置、移動履歴管理方法、および記録媒体 |
Families Citing this family (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080273754A1 (en) * | 2007-05-04 | 2008-11-06 | Leviton Manufacturing Co., Inc. | Apparatus and method for defining an area of interest for image sensing |
| JP4983505B2 (ja) * | 2007-09-25 | 2012-07-25 | ブラザー工業株式会社 | 無線タグ通信装置 |
| JP5088143B2 (ja) * | 2008-01-09 | 2012-12-05 | 富士通株式会社 | 位置判定方法 |
| JP4569663B2 (ja) * | 2008-04-25 | 2010-10-27 | ソニー株式会社 | 情報処理装置、情報処理方法、及びプログラム |
| JP5458802B2 (ja) * | 2008-10-23 | 2014-04-02 | リコーイメージング株式会社 | デジタルカメラ |
| US8600118B2 (en) * | 2009-06-30 | 2013-12-03 | Non Typical, Inc. | System for predicting game animal movement and managing game animal images |
| US9058732B2 (en) * | 2010-02-25 | 2015-06-16 | Qualcomm Incorporated | Method and apparatus for enhanced indoor position location with assisted user profiles |
| CA3147683C (en) | 2010-11-19 | 2023-09-05 | Isolynx, Llc | Associative object tracking systems and methods |
| US20120320204A1 (en) * | 2011-06-20 | 2012-12-20 | 3M Innovative Properties Company | Asset assessment system |
| JP5959923B2 (ja) * | 2012-04-26 | 2016-08-02 | キヤノン株式会社 | 検出装置、その制御方法、および制御プログラム、並びに撮像装置および表示装置 |
| JP6065911B2 (ja) * | 2012-08-06 | 2017-01-25 | 日本電気株式会社 | 配置情報登録装置、配置情報登録方法および配置情報登録プログラム |
| JP6021937B2 (ja) * | 2012-11-13 | 2016-11-09 | 三菱電機株式会社 | 空気調和システム及び中央管理装置 |
| JP6049448B2 (ja) * | 2012-12-27 | 2016-12-21 | キヤノン株式会社 | 被写体領域追跡装置、その制御方法及びプログラム |
| US20160088262A1 (en) * | 2013-04-10 | 2016-03-24 | Lg Electronics Inc. | Method For Managing Storage Product In Refrigerator Using Image Recognition, And Refrigerator For Same |
| US10929661B1 (en) * | 2013-12-19 | 2021-02-23 | Amazon Technologies, Inc. | System for user identification |
| JP5830706B2 (ja) * | 2014-01-29 | 2015-12-09 | パナソニックIpマネジメント株式会社 | 店員作業管理装置、店員作業管理システムおよび店員作業管理方法 |
| WO2017149582A1 (ja) * | 2016-02-29 | 2017-09-08 | 三井造船株式会社 | データ処理方法及び計測装置 |
| WO2018061328A1 (ja) * | 2016-09-30 | 2018-04-05 | 三菱電機ビルテクノサービス株式会社 | 所在人数予測装置、設備管理システム及びプログラム |
| CN114743326A (zh) * | 2022-04-07 | 2022-07-12 | 武汉东湖学院 | 一种设置有智能识别的智能制造车间防盗预警系统 |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2003233715A (ja) * | 2002-02-08 | 2003-08-22 | Sharp Corp | 生活情報管理システムおよび方法ならびに生活情報処理装置 |
| JP2004249389A (ja) * | 2003-02-19 | 2004-09-09 | Matsushita Electric Ind Co Ltd | 物品管理システム |
| JP2005037365A (ja) * | 2003-06-23 | 2005-02-10 | National Institute Of Information & Communication Technology | 物体配置図作成方法およびそのプログラムと記憶媒体、ならびに物体配置図作成システム |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH07146362A (ja) | 1993-11-25 | 1995-06-06 | Nippon Dry Chem Co Ltd | 物品探索システム |
| EP1027689A4 (en) * | 1997-11-03 | 2003-04-02 | Arial Systems Corp | Personnel and asset tracking method and apparatus |
| US6154139A (en) * | 1998-04-21 | 2000-11-28 | Versus Technology | Method and system for locating subjects within a tracking environment |
| US6084517A (en) * | 1998-08-12 | 2000-07-04 | Rabanne; Michael C. | System for tracking possessions |
| JP2000357251A (ja) | 1999-06-14 | 2000-12-26 | Sharp Corp | 物品管理システム |
| GB2380638B (en) * | 2000-05-22 | 2004-05-12 | Avery Dennison Corp | Trackable files and systems for using the same |
| US6300872B1 (en) * | 2000-06-20 | 2001-10-09 | Philips Electronics North America Corp. | Object proximity/security adaptive event detection |
| US7248933B2 (en) * | 2001-05-08 | 2007-07-24 | Hill-Rom Services, Inc. | Article locating and tracking system |
| US6933849B2 (en) * | 2002-07-09 | 2005-08-23 | Fred Sawyer | Method and apparatus for tracking objects and people |
-
2006
- 2006-03-17 WO PCT/JP2006/305401 patent/WO2006109423A1/ja not_active Ceased
- 2006-03-17 JP JP2006519678A patent/JP4006471B2/ja active Active
-
2007
- 2007-04-26 US US11/796,047 patent/US7545278B2/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2003233715A (ja) * | 2002-02-08 | 2003-08-22 | Sharp Corp | 生活情報管理システムおよび方法ならびに生活情報処理装置 |
| JP2004249389A (ja) * | 2003-02-19 | 2004-09-09 | Matsushita Electric Ind Co Ltd | 物品管理システム |
| JP2005037365A (ja) * | 2003-06-23 | 2005-02-10 | National Institute Of Information & Communication Technology | 物体配置図作成方法およびそのプログラムと記憶媒体、ならびに物体配置図作成システム |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7852217B2 (en) | 2005-12-28 | 2010-12-14 | Panasonic Corporation | Object detecting device, object detecting method and object detecting computer program |
| JP2010140754A (ja) * | 2008-12-11 | 2010-06-24 | Panasonic Electric Works Co Ltd | 照明システム |
| WO2011108055A1 (ja) * | 2010-03-03 | 2011-09-09 | パナソニック株式会社 | 物体位置推定装置、物体位置推定方法、及び、物体位置推定プログラム |
| JP4880805B2 (ja) * | 2010-03-03 | 2012-02-22 | パナソニック株式会社 | 物体位置推定装置、物体位置推定方法、及び、物体位置推定プログラム |
| JP2014108892A (ja) * | 2012-12-04 | 2014-06-12 | Fujitsu Ltd | 物品の配置位置管理装置、プログラム及び方法 |
| JP2015155345A (ja) * | 2014-02-20 | 2015-08-27 | 大和ハウス工業株式会社 | 収納管理システム |
| JP2018073012A (ja) * | 2016-10-26 | 2018-05-10 | 株式会社東芝 | 管理システム |
| JP2023110357A (ja) * | 2022-01-28 | 2023-08-09 | 積水ハウス株式会社 | 情報処理装置 |
| WO2024195059A1 (ja) * | 2023-03-22 | 2024-09-26 | 日本電気株式会社 | 移動履歴管理装置、移動履歴管理方法、および記録媒体 |
Also Published As
| Publication number | Publication date |
|---|---|
| JPWO2006109423A1 (ja) | 2008-10-16 |
| US20070247321A1 (en) | 2007-10-25 |
| JP4006471B2 (ja) | 2007-11-14 |
| US7545278B2 (en) | 2009-06-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP4006471B2 (ja) | 物品位置推定装置、物品位置推定方法、物品検索システム、及び物品位置推定用プログラム | |
| JP7260022B2 (ja) | 店舗装置、店舗システム、店舗管理方法、プログラム | |
| US9443414B2 (en) | Object tracking | |
| US7908237B2 (en) | Method and apparatus for identifying unexpected behavior of a customer in a retail environment using detected location data, temperature, humidity, lighting conditions, music, and odors | |
| US7908233B2 (en) | Method and apparatus for implementing digital video modeling to generate an expected behavior model | |
| US8099427B2 (en) | Search article estimation apparatus and method and server for search article estimation apparatus | |
| CN109753865B (zh) | 对象历史关联的系统和方法 | |
| KR102699484B1 (ko) | 냉장고 및 냉장고의 물품 관리 방법 | |
| CN112307864B (zh) | 用于确定目标对象的方法、装置、人机交互系统 | |
| CN103761505A (zh) | 对象跟踪 | |
| JP2017174272A (ja) | 情報処理装置及びプログラム | |
| JP6029622B2 (ja) | 情報管理サーバ、情報管理方法、および情報管理プログラム | |
| CN111488831B (zh) | 一种食材联想识别方法及冰箱 | |
| CN113326816A (zh) | 一种线下顾客行为识别方法、系统、存储介质及终端 | |
| CN109344680A (zh) | 物品登记系统 | |
| WO2019051167A1 (en) | SYSTEM FOR IDENTIFYING AND ANALYZING CUSTOMER INTERACTION | |
| US11561750B2 (en) | Retrieving personalized visual content items in real time for display on digital-content-display devices within a physical space | |
| US20210334758A1 (en) | System and Method of Reporting Based on Analysis of Location and Interaction Between Employees and Visitors | |
| Konstantinidis et al. | A deep network for automatic video-based food bite detection | |
| Ayub et al. | Don’t forget to buy milk: Contextually aware grocery reminder household robot | |
| CN114648385A (zh) | 一种智能货架的信息交互方法、设备和介质 | |
| US10726378B2 (en) | Interaction analysis | |
| Eno et al. | Virtual and real-world ontology services | |
| JP7647427B2 (ja) | 接客検出プログラム、接客検出方法および情報処理装置 | |
| Ling et al. | RFID-based user profiling of fashion preferences: blueprint for a smart wardrobe |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| WWE | Wipo information: entry into national phase |
Ref document number: 2006519678 Country of ref document: JP |
|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
| WWE | Wipo information: entry into national phase |
Ref document number: 11796047 Country of ref document: US |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| WWP | Wipo information: published in national office |
Ref document number: 11796047 Country of ref document: US |
|
| NENP | Non-entry into the national phase |
Ref country code: RU |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 06729390 Country of ref document: EP Kind code of ref document: A1 |