CN114005136A - End-to-end alarm condition identification and disposal recommendation system - Google Patents
End-to-end alarm condition identification and disposal recommendation system Download PDFInfo
- Publication number
- CN114005136A CN114005136A CN202111189240.7A CN202111189240A CN114005136A CN 114005136 A CN114005136 A CN 114005136A CN 202111189240 A CN202111189240 A CN 202111189240A CN 114005136 A CN114005136 A CN 114005136A
- Authority
- CN
- China
- Prior art keywords
- module
- police
- situation
- warning
- police situation
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/34—Browsing; Visualisation therefor
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Databases & Information Systems (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Animal Behavior & Ethology (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Alarm Systems (AREA)
Abstract
The invention discloses an end-to-end warning condition identification and handling recommendation system, which comprises a robot module, a video acquisition and processing module, a warning condition identification module, a warning condition degree classification module and a warning condition handling module, wherein the robot module is used for acquiring a video; the robot module is used for carrying subsequent modules and providing travelling power; the video acquisition and processing module is used for acquiring videos of surrounding environments through the RGB camera and processing the videos to obtain human skeleton points; the warning degree classification module is used for classifying the warning results identified by the warning identification module according to a given threshold value; and the alarm condition processing module is used for searching the processing mode and the related information corresponding to each alarm condition and displaying the processing mode and the related information for the reference of the police officers. The end-to-end warning condition identification and treatment recommendation system has the advantages of long endurance time of the robot, saving of warning power, reduction of police errors, capability of solving the problems of long treatment time and irregular treatment method caused by serious dependence on the experience of the police in the conventional warning condition treatment method and the like.
Description
Technical Field
The invention relates to an alarm information processing system, in particular to an end-to-end alarm identification and disposal recommendation system.
Background
End-to-end is a network connection. To communicate, a network must establish a connection, no matter how far away it is, how many machines there are, between both ends (source and destination), and once the connection is established, it is said to be an end-to-end connection, i.e., the end-to-end is a logical link.
In the years, economic society develops rapidly, law enforcement environments are increasingly complex, various policemen are complicated, various media are ubiquitous, the policemen are not disposed in place, and great public sentiments can be caused by carelessness. In recent years, most of police-related opinions have been associated with the disposition of the opinions.
In recent years, security robots have become popular with the development of robotics. However, most of existing security robots encapsulate a computer with high power consumption inside, which greatly affects endurance and causes more neutral time in security process due to frequent charging. Moreover, after the abnormal behavior recognition is completed, the alert handling method cannot be autonomously generated. The alarm needs to be notified to the control center, and the control center personnel notifies the police officers or the handling devices carried by the remote control robot to handle the alarm. In addition, most of the alarm information is stored in a file or database mode, the processing result is not displayed intuitively, the processing flow time is long, the processing flow depends on the abundant experience of the policemen seriously, and the policemen with insufficient experience are difficult to provide a correct processing mode.
The chinese patent application with application number 201410444499.5 discloses a multifunctional alarm processing method and system, the system includes a video alarm subsystem, a telephone alarm subsystem and a receiving and processing alarm subsystem, wherein: the video alarm subsystem is used for alarming to the alarm receiving and processing subsystem when the abnormal condition occurs in the monitoring area so as to send first alarm information to the alarm receiving and processing subsystem; the telephone alarm subsystem is used for alarming to the alarm receiving and processing subsystem when the alarm telephone is connected so as to send second alarm information to the alarm receiving and processing subsystem; and the alarm receiving and processing subsystem is used for receiving the first alarm information and/or the second alarm information and sending the alarm condition in the first alarm information and/or the second alarm information to an alarm terminal. The scheme can improve the alarm management efficiency from a certain angle, but cannot solve the problems that the existing alarm handling method is long in handling time and irregular in handling method due to the fact that the existing alarm handling method depends on the experience of policemen seriously.
Disclosure of Invention
The invention aims to avoid the defects in the prior art, and provides an end-to-end alarm condition identification and disposal recommendation system to solve the problems of long disposal time and irregular disposal method caused by serious dependence on the experience of police officers in the conventional alarm condition disposal method.
The invention adopts the following technical scheme to solve the technical problem.
An end-to-end warning condition identification and handling recommendation system is characterized by comprising a robot module, a video acquisition and processing module, a warning condition identification module, a warning condition degree classification module and a warning condition handling module; the robot module, the video acquisition and processing module, the warning condition recognition module, the warning condition degree classification module and the warning condition processing module are sequentially connected with one another;
the robot module is used for carrying subsequent modules and providing travelling power;
the video acquisition and processing module is used for acquiring videos of surrounding environments through the RGB camera and processing the videos to obtain human skeleton points;
the warning condition identification module is used for classifying the human skeleton points obtained by the video acquisition and processing module into warning condition actions;
the warning degree classification module is used for classifying the warning results identified by the warning identification module according to a given threshold value;
the alarm handling module is used for searching the handling mode and the related information corresponding to each alarm, displaying the handling mode and the related information for the police to refer to, and storing the handling information of the alarm.
The end-to-end warning condition identification and treatment recommendation system of the invention is also characterized in that:
preferably, the robot module includes a robot, a USB camera, and a wireless connection module.
Preferably, the video acquisition and processing module comprises a video acquisition unit and a human body posture estimation unit.
Preferably, the alert identification module comprises a self-built alert data set and a motion identification model.
Preferably, the self-built alarm condition data set comprises four categories of smashing, fighting, falling, illegal gathering and the like.
Preferably, the warning degree classification module comprises a warning degree classification model.
Preferably, the warning degree classification model classifies the warning into a short warning or a long warning according to the duration of the warning.
Preferably, the alert handling module includes a knowledge graph construction unit, a knowledge graph search unit, a result visualization unit, and an update unit.
Preferably, the knowledge map construction unit extracts the warning situation knowledge contained in the relevant authoritative law in an artificial extraction mode, and stores the extracted warning situation knowledge by using the map database to obtain the warning situation knowledge map.
Preferably, the warning situation knowledge graph comprises six types of warning situation nodes, namely a fine-grained warning situation node, a warning situation category node, a disposal mode node, a law and regulation node, a robot behavior node and a policeman node.
Compared with the prior art, the invention has the beneficial effects that:
the invention discloses an end-to-end warning condition identification and handling recommendation system, which comprises a robot module, a video acquisition and processing module, a warning condition identification module, a warning condition degree classification module and a warning condition handling module, wherein the robot module is used for acquiring a video; the robot module is used for acquiring a first video of an alarm condition; the video acquisition and processing module is used for acquiring videos of the surrounding environment through the RGB camera to obtain a second video and processing the two videos; the warning condition identification module is used for classifying the human skeleton points obtained by the video acquisition and processing module into warning condition actions; the warning degree classification module is used for classifying the warning results identified by the warning identification module according to a given threshold value; and the alarm condition processing module is used for searching the processing mode and the related information corresponding to each alarm condition, displaying the processing mode and the related information for the police officer to refer to, and storing the processing information of the alarm condition.
The end-to-end alarm situation identification and treatment recommendation system has the following advantages.
1. The robot module is separated from other modules, the power consumption in the robot is reduced, the endurance time of the robot is prolonged, and the security and protection neutral gear time caused by robot charging is reduced.
2. The warning condition degree classification model provided by the invention can classify the warning conditions in a length mode with finer granularity, and can adopt different disposal modes aiming at the warning conditions with different degrees, so that the disposal modes are more reasonable, the short warning conditions are completely solved by the robot, and the warning power is saved.
3. The disposal recommendation unit and the update unit can recommend a more appropriate disposal mode to the police officers, and update the warning situation knowledge graph according to the feedback of the police officers after the warning situation is finished, so that the knowledge graph can be continuously learned, is more comprehensive and accurate, visually displays the disposal measures and related attributes of the warning situation, provides reference and standard for the actions of the police officers, and reduces errors of the police officers caused by insufficient experience.
The end-to-end warning condition identification and treatment recommendation system has the advantages of long endurance time of the robot, saving of warning power, reduction of police errors, capability of solving the problems of long treatment time and irregular treatment method caused by serious dependence on the experience of the police in the conventional warning condition treatment method and the like.
Drawings
Fig. 1 is a block diagram of an end-to-end alarm identification and handling recommendation system according to the present invention.
Fig. 2 is a flow chart of the alert level classification of an end-to-end alert identification and disposition recommendation system of the present invention.
The present invention will be further described with reference to the following detailed description and accompanying drawings.
Detailed Description
The following detailed description of the preferred embodiments of the present invention is provided to enable those skilled in the art to more readily understand the advantages and features of the present invention, and to clearly and unequivocally define the scope of the present invention.
Referring to fig. 1-2, an end-to-end warning situation recognition and handling recommendation system of the present invention includes a robot module, a video acquisition and processing module, a warning situation recognition module, a warning situation degree classification module, and a warning situation handling module; the robot module, the video acquisition and processing module, the warning condition recognition module, the warning condition degree classification module and the warning condition processing module are sequentially connected with one another;
the robot module is used for carrying subsequent modules and providing travelling power;
the video acquisition and processing module is used for acquiring videos of surrounding environments through the RGB camera and processing the videos to obtain human skeleton points;
the warning condition identification module is used for classifying the human skeleton points obtained by the video acquisition and processing module into warning condition actions;
the warning degree classification module is used for classifying the warning results identified by the warning identification module according to a given threshold value;
the alarm handling module is used for searching the handling mode and the related information corresponding to each alarm, displaying the handling mode and the related information for the police to refer to, and storing the handling information of the alarm.
The robot module comprises a robot, a USB camera and a wireless connection module.
As shown in fig. 1, the robot module includes an autolobor robot, a USB camera, and a wireless connection module. The robot module is separate from the other modules but all exist within the same local area network. Due to the arrangement mode, the power consumption of the robot body is reduced, and the endurance time is prolonged. The robot adopts an Autolabor robot of the university of Qinghua as a carrier, the robot carries a USB camera, and the USB camera transmits videos through a wireless connection module.
The video acquisition and processing module comprises a video acquisition unit and a human body posture estimation unit.
As shown in fig. 1, the video capture and processing module is configured to process RGB videos captured by the USB camera into human skeleton point data. The video acquisition and processing module comprises a video acquisition unit and a human body posture estimation unit. The system comprises a video acquisition unit, a video processing unit and a video processing unit, wherein the video acquisition unit acquires videos of an environment in real time by using a common RGB camera; the human body posture estimation unit extracts 18 skeletal points of a human body in the RGB video by using an OpenPose algorithm and a COCO model to obtain coordinates and confidence.
The warning condition identification module comprises a self-built warning condition data set and an action identification model.
The self-built alarm data set comprises smashing, supporting, falling and illegal gathering.
As shown in fig. 1, the warning identification module is configured to classify the warning actions of the human skeleton points obtained by the video acquisition and processing module. The alarm condition identification module comprises an action identification model and a self-built alarm condition data set. In specific implementation, the motion recognition model can be based on deep learning, and an ST-GCN algorithm is adopted to train a self-established warning situation data set to obtain model weight. The self-built alarm data set comprises four types of artificial alarms including smashing, supporting, falling and illegal gathering. And extracting time characteristics through GCN, extracting space characteristics through TCN, and classifying the characteristics to obtain a classification result of the alarm condition.
The warning condition degree classification module comprises a warning condition degree classification model.
The warning condition degree classification model classifies the warning condition into a short warning condition or a long warning condition according to the duration of the warning condition.
As shown in fig. 1, the warning degree classification module is configured to classify each warning situation into a short warning situation or a fine-grained warning situation of a long warning situation according to the duration of the warning situation and according to a given threshold value, the warning situation result identified by the warning situation identification module.
Specifically, in the warning degree classification model, each warning situation is subdivided into two types of length according to the duration, and the given threshold value can be adjusted according to the actual operation condition.
In one specific embodiment, a short alert decision threshold a of 1800 frames, a long alert decision threshold B of 5400 frames, a maximum duration threshold T of 180 seconds, a long alert decision window L of 45 seconds, and a short alert decision window S of 180 seconds are given. The starting points of the two windows are the same, L and S belong to an inclusion relationship, a plurality of alarm condition duration time lists are maintained, and the alarm condition duration time lists store the time of each occurrence of the alarm condition. And the alarm condition duration time is smaller than the threshold A in the window S, and the alarm condition duration time is considered as a short alarm condition, and the alarm condition duration time is larger than the threshold B in the window L, and the alarm condition duration time is considered as a long alarm condition. And resetting after the long warning duration time exceeds a threshold value T, and outputting the picture and the warning at the current moment after the warning is recognized, as shown in fig. 2.
The warning situation handling module comprises a knowledge graph construction unit, a knowledge graph searching unit, a result visualization unit and an updating unit.
The knowledge map construction unit extracts the warning situation knowledge contained in the relevant authoritative law in an artificial extraction mode, and the warning situation knowledge map is obtained by storing the extracted warning situation knowledge by using the map database.
The warning condition knowledge graph comprises six types of warning condition nodes which are respectively a fine-grained warning condition node, a warning condition category node, a disposal mode node, a law and regulation node, a robot behavior node and an policeman node.
As shown in fig. 1, the alert handling module is configured to construct an alert situation knowledge graph in a knowledge graph construction unit for the fine-grained alert situation obtained by the alert degree classification module, search a handling manner and related information corresponding to the alert situation in the alert situation knowledge graph through a knowledge graph search unit, display the information for reference of an alert through a result visualization unit, and finally update the knowledge graph through the feedback of the alert situation handling of the alert situation in the update unit.
Specifically, the knowledge graph construction unit artificially extracts the alert knowledge contained in the related authoritative laws "criminal law", "security management punishment law" and "party tourism westernization law". And storing the extracted knowledge by using the graph database Neo4j to obtain an alarm situation knowledge graph, and storing the alarm situation knowledge graph in a triple mode of entity-relation-entity and entity-attribute values.
Defining 6 types of nodes, fine-grained warning condition nodes, warning condition category nodes, disposal mode nodes, law and regulation nodes, robot behavior nodes and police officer nodes in a warning condition knowledge graph.
More specifically, in the 6 types of nodes, the fine-grained warning node is used for classifying the long and short fine-grained warning output by the warning degree classification module.
The warning situation type node is a warning situation type which can be identified in the warning situation identification module.
The disposal mode node is a disposal mode specified in laws and regulations, and the disposal mode node contains a recommendation weight attribute and represents the recommendation degree of the disposal mode.
Legal nodes are the source of disposal means.
The robot behavior node is a reaction to the warning condition which can be completed by the patrol robot, and comprises sound and light warning and police officer notification, and the police officer node comprises all police officers in the cooperative security database.
The police officer node contains a position attribute and displays the position of the current time.
More specifically, the following 6 relationships exist between the 6 types of nodes:
(1) a fine-grained alert node belongs to an alert category node;
(2) alert category node-take-dispose mode node;
(3) disposal mode node-source-law and regulation node;
(4) fine-grained alert node-adopting-robot behavior node;
(5) robot behavior node-notification-police node;
(6) the method comprises the steps of police officer node-control-robot behavior node.
Specifically, the knowledge graph search unit searches the warning situation obtained by the warning situation degree classification module as a fine-grained warning situation node in the warning situation knowledge graph constructed by the knowledge graph construction unit, and outputs a search result containing the relationship between the 6-class node and the 6-class node.
Specifically, the disposal recommendation unit performs preliminary disposal on the alarm condition according to a search result provided by the knowledge graph search unit, and takes a sound-light warning mode for the short alarm condition. If the warning is invalid, namely the warning is upgraded to a long warning situation, the search result is visually provided for the police. The police gives the recommended weight of a plurality of disposal methods by referring to the search result, and decides to select the type of the disposal method, such as controlling the robot to carry out remote warning or carrying out the disposal methods provided by laws and regulations.
Specifically, the updating unit updates the alert situation knowledge graph according to the feedback of the police officer after the alert situation ends, increases the weight of the police officer if the police officer selects a certain disposal mode, and stores the police officer in the alert situation knowledge graph if the police officer adopts a new disposal mode.
The invention provides an end-to-end warning condition identification and disposal recommendation system, which can replace the policeman to carry out long-time patrol operation, reduce the danger of the policeman on duty in dangerous areas and unnecessary police force waste,
the robot module is separated from other modules, the power consumption in the robot is reduced, the endurance time of the robot is prolonged, and the security and protection neutral gear time caused by robot charging is reduced.
The warning condition degree classification model provided by the invention can classify the warning conditions in a length mode with finer granularity, and can adopt different disposal modes aiming at the warning conditions with different degrees, so that the disposal modes are more reasonable, the short warning conditions are completely solved by the robot, and the warning power is saved.
The disposal recommendation unit and the update unit can recommend a more appropriate disposal mode to the police officers, and update the warning situation knowledge graph according to the feedback of the police officers after the warning situation is finished, so that the knowledge graph can be continuously learned, is more comprehensive and accurate, visually displays the disposal measures and related attributes of the warning situation, provides reference and standard for the actions of the police officers, and reduces errors of the police officers caused by insufficient experience.
The invention discloses an end-to-end alarm condition identification and disposal recommendation system, which aims to increase the endurance time of a security robot and solve the problems of long disposal time and irregular disposal method caused by the fact that the existing alarm condition disposal method depends on the experience of an alarm seriously, can liberate the alarm strength to a certain degree, shorten the alarm condition feedback time, and simultaneously visually display the disposal mode, improve the alarm condition processing efficiency, and can work in a dangerous area to reduce the danger of the alarm.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned.
Furthermore, it should be understood that although the present description refers to embodiments, not every embodiment may contain only a single embodiment, and such description is for clarity only, and those skilled in the art should integrate the description, and the embodiments may be combined as appropriate to form other embodiments understood by those skilled in the art.
Claims (10)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111189240.7A CN114005136A (en) | 2021-10-12 | 2021-10-12 | End-to-end alarm condition identification and disposal recommendation system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111189240.7A CN114005136A (en) | 2021-10-12 | 2021-10-12 | End-to-end alarm condition identification and disposal recommendation system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114005136A true CN114005136A (en) | 2022-02-01 |
Family
ID=79922765
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111189240.7A Pending CN114005136A (en) | 2021-10-12 | 2021-10-12 | End-to-end alarm condition identification and disposal recommendation system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114005136A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115755817A (en) * | 2022-11-30 | 2023-03-07 | 苏州艾科瑞思智能装备股份有限公司 | Production management platform of integrated circuit chip mounter |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103984315A (en) * | 2014-05-15 | 2014-08-13 | 成都百威讯科技有限责任公司 | Domestic multifunctional intelligent robot |
| CN107943276A (en) * | 2017-10-09 | 2018-04-20 | 广东工业大学 | Based on the human body behavioral value of big data platform and early warning |
| CN110139598A (en) * | 2016-12-29 | 2019-08-16 | 关健强 | Monitoring and tracking system, method, article and equipment |
| CN110427834A (en) * | 2019-07-10 | 2019-11-08 | 上海工程技术大学 | A kind of Activity recognition system and method based on skeleton data |
| CN111444353A (en) * | 2020-04-03 | 2020-07-24 | 杭州叙简科技股份有限公司 | Construction and use method of a police knowledge map |
-
2021
- 2021-10-12 CN CN202111189240.7A patent/CN114005136A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103984315A (en) * | 2014-05-15 | 2014-08-13 | 成都百威讯科技有限责任公司 | Domestic multifunctional intelligent robot |
| CN110139598A (en) * | 2016-12-29 | 2019-08-16 | 关健强 | Monitoring and tracking system, method, article and equipment |
| CN107943276A (en) * | 2017-10-09 | 2018-04-20 | 广东工业大学 | Based on the human body behavioral value of big data platform and early warning |
| CN110427834A (en) * | 2019-07-10 | 2019-11-08 | 上海工程技术大学 | A kind of Activity recognition system and method based on skeleton data |
| CN111444353A (en) * | 2020-04-03 | 2020-07-24 | 杭州叙简科技股份有限公司 | Construction and use method of a police knowledge map |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115755817A (en) * | 2022-11-30 | 2023-03-07 | 苏州艾科瑞思智能装备股份有限公司 | Production management platform of integrated circuit chip mounter |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117749995B (en) | Video monitoring method and system based on multi-scene recognition and voice interaction | |
| CN118486152B (en) | Security alarm information data interaction system and method | |
| AU2017436901B2 (en) | Methods and apparatus for automated surveillance systems | |
| CN115620208A (en) | Power grid security early warning method, device, computer equipment and storage medium | |
| US10860865B2 (en) | Predictive security camera system | |
| CN207820108U (en) | A kind of transmission line long-distance video monitoring system | |
| CN116846059A (en) | An edge detection system for power grid inspection and monitoring | |
| CN105100724A (en) | A smart home remote security monitoring method and device based on visual analysis | |
| CN110457984A (en) | Pedestrian attribute recognition method based on ResNet-50 in surveillance scene | |
| CN104394361A (en) | Pedestrian crossing intelligent monitoring device and detection method | |
| CN112863100B (en) | Intelligent construction safety monitoring system and method | |
| CN112699750A (en) | Safety monitoring method and system for intelligent gas station based on edge calculation and AI (Artificial Intelligence) | |
| CN108540310A (en) | A kind of behavior prediction method based on wireless network collaborative perception | |
| CN115103155A (en) | A public video surveillance system based on big data | |
| CN112487961B (en) | Traffic accident detection method, storage medium and device | |
| CN107123242A (en) | A kind of intelligent building Video Surveillance Alarm System | |
| CN114187541A (en) | Intelligent video analysis method and storage device for user-defined service scene | |
| CN108629037A (en) | A kind of smart city service platform and its management method | |
| CN117290431A (en) | System and method for fusing and visualizing multidimensional data of intelligent park | |
| CN114841312B (en) | A weakly supervised video anomaly detection method based on adaptive graph convolutional network | |
| CN105959621A (en) | Quasi-real time control distribution system and method based on multi-source video structural data | |
| CN117453835A (en) | Intelligent community management system | |
| EP3850502B1 (en) | Method for detection of temporal pattern anomalies in video streams | |
| CN113642403B (en) | Intelligent security detection system for crowd movement based on edge computing | |
| CN119893264A (en) | Camera monitoring system based on wireless communication and communication protocol optimization method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |