CN108447087A - The method that the article of suitable object size is looked for using object image - Google Patents
The method that the article of suitable object size is looked for using object image Download PDFInfo
- Publication number
- CN108447087A CN108447087A CN201810162574.7A CN201810162574A CN108447087A CN 108447087 A CN108447087 A CN 108447087A CN 201810162574 A CN201810162574 A CN 201810162574A CN 108447087 A CN108447087 A CN 108447087A
- Authority
- CN
- China
- Prior art keywords
- image
- skin
- group
- colour
- label
- 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.)
- Withdrawn
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/903—Querying
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M1/00—Substation equipment, e.g. for use by subscribers
- H04M1/72—Mobile telephones; Cordless telephones, i.e. devices for establishing wireless links to base stations without route selection
- H04M1/724—User interfaces specially adapted for cordless or mobile telephones
- H04M1/72403—User interfaces specially adapted for cordless or mobile telephones with means for local support of applications that increase the functionality
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M1/00—Substation equipment, e.g. for use by subscribers
- H04M1/72—Mobile telephones; Cordless telephones, i.e. devices for establishing wireless links to base stations without route selection
- H04M1/724—User interfaces specially adapted for cordless or mobile telephones
- H04M1/72403—User interfaces specially adapted for cordless or mobile telephones with means for local support of applications that increase the functionality
- H04M1/7243—User interfaces specially adapted for cordless or mobile telephones with means for local support of applications that increase the functionality with interactive means for internal management of messages
- H04M1/72439—User interfaces specially adapted for cordless or mobile telephones with means for local support of applications that increase the functionality with interactive means for internal management of messages for image or video messaging
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30196—Human being; Person
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Human Computer Interaction (AREA)
- Signal Processing (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computer Networks & Wireless Communication (AREA)
- Databases & Information Systems (AREA)
- Geometry (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Business, Economics & Management (AREA)
- General Business, Economics & Management (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of methods of article that looking for suitable object size using object image, include the following steps：A label is respectively set at 4 angles of a rectangle plane；It is located at the object in plane with mobile phone shooting to obtain its image, which must contain at least three label；Make image procossing；K means algorithms are executed, wherein k=3 and profile value are 6；It obtains group's quantity and calculates per group of central value, the wherein central value is average value；The colour of skin in image is made into binary conversion treatment, wherein by image each pixel and the distance between immediate colour of skin group center be set as 255, and the pixel and distant the distance between other colour of skin group center are set as 0；Obtain object size；The object size of acquirement is inputted into item database；And a Item Number for complying fully with the object size is looked for out in item database using retrieval mode.
Description
Technical field
The present invention relates to image processing field, particularly relate to a kind of method, 4 feet of this method using rectangle plane fall as
Label, foot's (i.e. object) steps on rectangle plane, and the image of foot is shot with mobile phone, as long as containing at least three mark in image
Label and foot and label are in the same plane, you can and judge the just size of foot, then size value is inputted into item database,
To look for the shoes for being suitble to the size.
Background technology
Long-range (network) looks for lower single pattern of shoes (or clothes etc.) at present, is typically found most by normalized size
It is suitably sized, the purchase that places an order then is determined again；But since there are the mode of various different codings, each ethnic groups for the size of shoes
Foot shape be also not quite similar, the coding of shoes code can only generally also stipulate the ruler of shoes with the width of the length of sole and heelpiece at present
Code specification, it is therefore evident that its at least miss the internal diameter size of the thickness of sole and shoes cylinder, thus occur buying shoes
When even if shoes are long and shoes are wide suitable, but the average case that whole shoes do not fit；It is such to perplex while may occur at shoemaking
Factory opens shoe-making die and makes in the link of last；Therefore, how perfect remote measuring method and accuracy are provided,
Search is compared from the merchandising database by image procossing, calculation, and connection merchandising database again, so as to long-range
Can find the shoes more to fit, the present invention emphasis and the problem to be solved in the present invention；Certainly the present invention is not
It is only confined to look for shoes and this narrow concept of clothes, but can be widely applied to as any object from long-range zhang
It measures and is compared in database and found, to obtain a predetermined target item for being used for wearing or arranging in pairs or groups, finally can smoothly be solved
Dimensionally inappropriate technical problem.
Invention content
To solve the disadvantage that the aforementioned article for being not easy to look for suitable size, present invention offer is a kind of to utilize object image
The method for looking for the article of suitable object size comprising following steps：
(1) label is respectively set at 4 angles of a rectangle plane；
(2) it is located at the object in plane with mobile phone shooting to obtain its image；
(3) by image diminution and smoothing processing；
(4) with the noise in a default threshold value filtering image；
(5) by image compared with an adaptability threshold value, to remove the partial image less than adaptability threshold value；
(6) using accurate edge detection to strengthen the edge of image；
(7) the profile stratum number of any label is counted；
(8) silhouette of object in image is calculated to obtain number of tags；
(9) if number of tags is more than 2, next step is carried out, otherwise flow returns to step (2)；
(10) coordinate of not shown 4th label in image is found out with extrapolation；
(11) the distance between arbitrary 2 points in image in same level are restored using perspective projection；
(12) image is made into the conversion of Lab color spaces；
(13) by image make it is appropriate zoom in or out, and the two dimensional character of image is converted into one-dimensional characteristic；
(14) k-means algorithms are executed, wherein k=3 is to indicate a colour of skin group, a white group and other background colors
Group and profile value are 6；
(15) it obtains group's quantity and calculates per group of central value, the wherein central value is average value；
(16) skin tone value (L is used：90, a：12, b：12) immediate group center and distinguishing as the colour of skin are calculated
Knowledge value；
(17) colour of skin in image is made into binary conversion treatment, wherein by each pixel and immediate colour of skin group in image
The distance between center is set as 255, and the pixel and distant the distance between other colour of skin group center are set as 0；
(18) length and width of object in image are measured；
(19) object size is made into projection transform；
(20) object size is obtained；
(21) object size of acquirement is inputted into item database；And
(22) a Item Number for complying fully with the object size is looked for out in item database using retrieval mode.
In one embodiment of this invention, the label in step (1) is in the surface of the plane.
Description of the drawings
Fig. 1 is that the schematic diagram on rectangle plane is stepped on by first embodiment of the invention foot；
Fig. 2 is the method flow diagram for the article that first embodiment of the invention looks for suitable object size using object image；
Fig. 3 is the schematic diagram that label includes 6 layers of profile；
Fig. 4 is the schematic diagram for calculating the 4th label；
Fig. 5 is the schematic diagram of foot length；
Fig. 6 is the schematic diagram of foot's width；
Fig. 7 is that the schematic diagram on rectangle plane is stepped on by second embodiment of the invention foot；
Fig. 8 is the method flow diagram for the article that second embodiment of the invention looks for suitable object size using object image；
And
Fig. 9 is the schematic diagram that third embodiment of the invention people stands beside rectangle plane.
Specific implementation mode
The technological means for reaching above-mentioned purpose is used about the present invention, hereby coordinates icon with specifically with following implementation type state
It is bright.
Referring to FIG. 1, it shows that four corners of a rectangle plane 1 are respectively arranged with a label 11 and foot (makees herein
For object) it steps on rectangle plane 1, as the first step for implementing first embodiment of the invention.
Referring to FIG. 2, it shows that first embodiment of the invention looks for the article of suitable object size using object image
Method flow diagram, each step are as described below in detail.
(A1) foot for using client shown in mobile phone shooting figure 1 obtains its image, and foot and label 11 are in the same plane；
Then image is uploaded into the present invention and makees label detection below.
(A2) image is made into diminution appropriate and smoothing processing, so that mobile phone can execute rapidly the present invention.
(A3) since the label 11 set by the present invention is black and white label, therefore take a default threshold value to filter in image
Other noises, and because image brightness is inconsistent, then by image and an adaptability threshold value (adaptive threshold)
Compare, the partial image removal of adaptability threshold value is will be less than, to obtain clearer image.
(A4) using accurate edge detection method (canny edge detection) to strengthen the edge of image.
(A5) the contoured quantity of statistics institute and stratum, specifically, from the label of the present invention known to 6 arrows of Fig. 3
Stratum's profile is 6 layers, therefore takes 6 layers of profile.
(A6) silhouette of foot in image is calculated to obtain number of tags.
(A7) if number of tags is more than 2, next step is carried out；If number of tags is not more than 2, flow returns to step
Suddenly (A1).
(A8) coordinate of 4 label positions of foot in image is determined, specifically, Fig. 4, which is shown, is located at 3 angles in image
3 labels fallen, position can use coordinate representation, and the wherein label in the upper left corner is the first label；Utilize the second of the lower left corner
2 points of the right side of the third label of 2 points of the lower section of label and the upper right corner is obtained with extrapolation positioned at bottom right after online extension
The coordinate of 4th label position at angle.
(A9) after the coordinate for obtaining four label positions, you can learn position of the image in real space, recycle saturating
Depending on projection (perspective transformation) formula, restore between any two points in image in same level away from
From.
Flow then makees colour of skin detecting
(B1) image is made into the conversion of Lab color spaces, specifically, since rgb color space is for the identification of the colour of skin
It is relatively low, so video conversion to be promoted to the susceptibility of color at Lab color spaces.
Flow then executes training mode
Training mode
(B2) by image make it is appropriate zoom in or out, and the two dimensional character of image is converted into one-dimensional characteristic.
(B3) k-means algorithms (k=3) are executed, specifically color is divided into three group's (i.e. colours of skin by k-means algorithms
Group, white group and other background colours group), and parameter when by profile value 6 as operation.
(B4) obtain group's quantity and calculate the central value of each group, specifically after having executed k-means algorithms incited somebody to action
Color is divided into different groups, and each group can obtain group center's value with an average value, is equal to the 3 i.e. value of k, then make
With the skin tone value (L that Asia is general：90, a：12, b：12) identifier of the immediate group center as the colour of skin is calculated.
(B5) colour of skin in image is made into binary conversion treatment, specifically in the new striograph subsequently obtained, obtains image
In each the distance between pixel (pixel) and immediate colour of skin group center, such as will pixel in immediate colour of skin group
The distance between heart is set as 255, and pixel and distant the distance between other colour of skin group center are set as 0.
(B6) length and width of foot in image are measured, obtained binaryzation profile passes through the foot degree of different shootings
It can measure the length (Fig. 5) and width (Fig. 6) of foot.
Flow then executes the conversion of image physical length.
(C1) foot size is made into projection transform.
(C2) foot size is obtained.
Flow finally looks for the article of suitable foot size.
(D1) foot size of acquirement is inputted into item database, wherein there are many variety classes and the shoes of size for storage
Archives.
(D2) a suitable shoes for complying fully with the foot size are looked for out in item database using retrieval mode
Item Number.
Advantages of the present invention is：Any 2 points distance in photo quickly and is accurately measured, can be obtained just with measurement result
True size, and can be with automatic storage measurement result.
Manufacturer can accurately produce a shoes for complying fully with customer requirement size using the Item Number.It is somebody's turn to do
Solution, although above-mentioned illustrated using shoes as object, this is appreciated that, although above-mentioned is using shoes as object
Illustrate, but the method for the present invention is also applied for other objects, such as clothes.
Referring to FIG. 7, it shows that four corners of a rectangle plane 1 are respectively arranged with a label 11, foot (in this as
Object) it steps on rectangle plane 1, and the suctorial pillar of tool 12 is resisted against toe as Z axis, and a mark is provided on the pillar 12
Label 11, as the first step for implementing second embodiment of the invention.
Referring to FIG. 8, it shows that second embodiment of the invention looks for the article of suitable object size using object image
Method flow diagram, each step is identical with step shown in Fig. 2, therefore description is omitted.
Referring to FIG. 9, it shows that a rectangle plane 1 is attached to the wall left side, four corners of the plane 1 are respectively arranged with one
Label 11, people's (in this as object) then stand on the right of wall, as the third embodiment of the present invention, purpose explanation when pair
As it is parallel with plane 1 when, the present invention can also be implemented.
Claims (2)
1. a kind of method of article that looking for suitable object size using object image, which is characterized in that include the following steps：
(1) label is respectively set at 4 angles of a rectangle plane；
(2) it is located at the object in plane with mobile phone shooting to obtain its image；
(3) by image diminution and smoothing processing；
(4) with the noise in a default threshold value filtering image；
(5) by image compared with an adaptability threshold value, to remove the partial image less than adaptability threshold value；
(6) using accurate edge detection to strengthen the edge of image；
(7) the profile stratum number of any label is counted；
(8) silhouette of object in image is calculated to obtain number of tags；
(9) if number of tags is more than 2, next step is carried out, otherwise flow returns to step (2)；
(10) coordinate of not shown 4th label in image is found out with extrapolation；
(11) the distance between arbitrary 2 points in image in same level are restored using perspective projection；
(12) image is made into the conversion of Lab color spaces；
(13) by image make it is appropriate zoom in or out, and the two dimensional character of image is converted into one-dimensional characteristic；
(14) k-means algorithms are executed, wherein k=3 is to indicate a colour of skin group, a white group and other background colors group, with
And profile value is 6；
(15) it obtains group's quantity and calculates per group of central value, the wherein central value is average value；
(16) skin tone value (L is used：90, a：12, b：12) immediate group center and the identifier as the colour of skin are calculated；
(17) colour of skin in image is made into binary conversion treatment, wherein by each pixel and immediate colour of skin group center in image
The distance between be set as 255, and the pixel and distant the distance between other colour of skin group center are set as 0；
(18) length and width of object in image are measured；
(19) object size is made into projection transform；
(20) object size is obtained；
(21) object size of acquirement is inputted into item database；And
(22) a Item Number for complying fully with the object size is looked for out in item database using retrieval mode.
2. the method for looking for the article of suitable object size using object image as described in claim 1, it is characterised in that：Step
Suddenly the label in (1) is in the surface of the plane.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810162574.7A CN108447087A (en) | 2018-02-27 | 2018-02-27 | The method that the article of suitable object size is looked for using object image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810162574.7A CN108447087A (en) | 2018-02-27 | 2018-02-27 | The method that the article of suitable object size is looked for using object image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108447087A true CN108447087A (en) | 2018-08-24 |
Family
ID=63192616
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810162574.7A Withdrawn CN108447087A (en) | 2018-02-27 | 2018-02-27 | The method that the article of suitable object size is looked for using object image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108447087A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100111370A1 (en) * | 2008-08-15 | 2010-05-06 | Black Michael J | Method and apparatus for estimating body shape |
| CN103098076A (en) * | 2010-09-17 | 2013-05-08 | 索尼公司 | Gesture recognition system for TV control |
| CN105118051A (en) * | 2015-07-29 | 2015-12-02 | 广东工业大学 | Saliency detecting method applied to static image human segmentation |
| CN106097354A (en) * | 2016-06-16 | 2016-11-09 | 南昌航空大学 | A kind of combining adaptive Gauss Face Detection and the hand images dividing method of region growing |
| CN106127181A (en) * | 2016-07-02 | 2016-11-16 | 乐活无限（北京）科技有限公司 | One is virtual tries manicure method, system on |
| CN107481267A (en) * | 2017-08-14 | 2017-12-15 | 华南理工大学 | A kind of shooting projection interactive system and method based on binocular vision |
-
2018
- 2018-02-27 CN CN201810162574.7A patent/CN108447087A/en not_active Withdrawn
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100111370A1 (en) * | 2008-08-15 | 2010-05-06 | Black Michael J | Method and apparatus for estimating body shape |
| CN103098076A (en) * | 2010-09-17 | 2013-05-08 | 索尼公司 | Gesture recognition system for TV control |
| CN105118051A (en) * | 2015-07-29 | 2015-12-02 | 广东工业大学 | Saliency detecting method applied to static image human segmentation |
| CN106097354A (en) * | 2016-06-16 | 2016-11-09 | 南昌航空大学 | A kind of combining adaptive Gauss Face Detection and the hand images dividing method of region growing |
| CN106127181A (en) * | 2016-07-02 | 2016-11-16 | 乐活无限（北京）科技有限公司 | One is virtual tries manicure method, system on |
| CN107481267A (en) * | 2017-08-14 | 2017-12-15 | 华南理工大学 | A kind of shooting projection interactive system and method based on binocular vision |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109635818B (en) | Seal anti-counterfeiting inspection method, seal anti-counterfeiting inspection device and computer-readable storage medium | |
| KR102783839B1 (en) | Method for detecting skin quality, method for classifying skin quality grades, device for detecting skin quality, electronic device and storage medium | |
| US10013803B2 (en) | System and method of 3D modeling and virtual fitting of 3D objects | |
| EP2359313B1 (en) | Method and system for item identification | |
| CN110546651B (en) | Method, system and computer readable medium for identifying objects | |
| US20190180432A1 (en) | Color Quality Assessment Based on Multispectral Imaging | |
| CN109977886A (en) | Shelf vacancy rate calculation method and device, electronic equipment, storage medium | |
| GB2430735A (en) | Object detection | |
| CN109271937A (en) | Athletic ground Marker Identity method and system based on image procossing | |
| WO2017040338A1 (en) | System for accurate remote measurement | |
| CN110008968B (en) | Automatic triggering method for robot settlement based on image vision | |
| US20170079599A1 (en) | Moisture feeling evaluation device, moisture feeling evaluation method, and moisture feeling evaluation program | |
| KR102071410B1 (en) | Smart mirror | |
| CN109330106A (en) | A kind of subscript dimension measurement method based on mobile phone photograph | |
| CN115880297A (en) | Quilt cover dyeing quality evaluation method based on machine vision | |
| CN119919867A (en) | Vending machine product visual recognition system based on machine learning model | |
| JP7767798B2 (en) | Verification device and program | |
| CN108447087A (en) | The method that the article of suitable object size is looked for using object image | |
| CN106485265B (en) | An image recognition method and device | |
| JP5411092B2 (en) | Fashion check system using mobile devices | |
| WO2015136716A1 (en) | Image processing device, image sensor, and image processing method | |
| CN109859228A (en) | Recommend the method, apparatus and computer equipment of shoes money type based on image recognition | |
| CN111429394B (en) | Image-based detection methods, devices, electronic equipment and storage media | |
| TW201935414A (en) | Method of selecting an article for covering a body part by processing the image of the body part | |
| CN117152787A (en) | A method, device, equipment and readable storage medium for character clothing recognition |
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 | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20180824 |
|
| WW01 | Invention patent application withdrawn after publication |