CN109784346A - A kind of screening technique of sketch figure picture, electronic equipment and storage medium - Google Patents
A kind of screening technique of sketch figure picture, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN109784346A CN109784346A CN201811472041.5A CN201811472041A CN109784346A CN 109784346 A CN109784346 A CN 109784346A CN 201811472041 A CN201811472041 A CN 201811472041A CN 109784346 A CN109784346 A CN 109784346A
- Authority
- CN
- China
- Prior art keywords
- picture
- screened
- pictures
- sketch
- coordinate vector
- 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.)
- Granted
Links
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Abstract
A kind of screening technique of sketch figure picture provided by the invention, sketch figure picture is for making outline, include: the pictures to be screened obtained for making outline, the first gray proces are carried out to the picture to be screened in pictures to be screened and obtain the processed pictures to be screened containing multiple processed pictures to be screened；Processed pictures to be screened are input to default edge contour study collection model to identify, obtain the coordinate vector set containing multiple edge contour key point coordinate vector groups；The contour curve set containing multiple picture contour curves to be screened is obtained according to coordinate vector set, calculate separately the turning quantity of each picture contour curve to be screened, filter out picture to be screened corresponding with maximum turning quantity, will picture to be screened corresponding with maximum turning quantity as target sketch figure picture.A kind of screening technique of sketch figure picture of the invention, obtained target sketch figure picture are better able to embody the contour feature of corresponding personage or article.
Description
Technical field
The present invention relates to field of image recognition more particularly to a kind of screening technique of sketch figure picture, electronic equipment and deposit
Storage media.
Background technique
Traditional artifacts of the outline as the Chinese nation, it is very welcomed by the people now.According to face or human body
And the outline of the profile paper-cut forming meaning of other objects.When making outline, performance be people, object profile, make every effort to prominent master
Body characteristics, the most distinctive shape of performance main body, posture, chamfered shape.It is also many due to the unique artistic style of outline
Large-scale theme meeting is used as the souvenir card cover decoration for ceremony of registering.Usually obtain sketch figure picture for making outline and be all with
Under type obtains: inviting guest to shoot photo at meeting scene, generates sketch figure picture after automatically processing the photo taken
And it is fabricated to the background of meeting souvenir card.But the effect image of outline is made, shooting angle to image and selecting is wanted
It asks higher, needs to obtain the picture for best embodying the prominent contour feature of welcome guest of outline production effect, while can not also require every
A guest expends the time specially by the requirement adjustment posture of photographer and angular compliance shooting.At present for for making
The selection of the sketch figure picture of outline is substantially artificial selection, and the picture chosen is caused not to be able to satisfy wanting for production outline
It asks, the contour feature of the corresponding personage of sketch figure picture or article can not be embodied well by choosing sketch figure picture.
Summary of the invention
For overcome the deficiencies in the prior art, one of the objects of the present invention is to provide a kind of screening sides of sketch figure picture
Method can solve and be substantially artificial selection for the selection of the sketch figure picture for making outline at present, cause to choose
Picture is not able to satisfy the requirement of production outline, and the corresponding personage of sketch figure picture can not be embodied well by choosing sketch figure picture
Or article contour feature the problem of.
The second object of the present invention is to provide a kind of electronic equipment, can solve at present for for making cutting for outline
The selection of shadow image is substantially artificial selection, and the picture chosen is caused not to be able to satisfy the requirement of production outline, and selection is cut
Shadow image can not embody the problem of contour feature of the corresponding personage of sketch figure picture or article well.
The third object of the present invention is to provide a kind of computer readable storage medium, can solve at present for for making
The selection for making the sketch figure picture of outline is substantially artificial selection, and the picture chosen is caused not to be able to satisfy wanting for production outline
It asks, chooses the problem of sketch figure picture can not embody the contour feature of the corresponding personage of sketch figure picture or article well.
The present invention provides the first purpose and is implemented with the following technical solutions:
A kind of screening technique of sketch figure picture, the sketch figure picture is for making outline, comprising the following steps:
Picture to be screened is obtained, the pictures to be screened for making outline is obtained, contains in the pictures to be screened
Multiple pictures to be screened；
First gray proces carry out the first gray proces to the picture to be screened in the pictures to be screened and obtain
To the processed pictures to be screened containing multiple processed pictures to be screened；
Picture recognition to be screened, by the processed pictures to be screened be input to default edge contour study collection model into
Row identification, obtains the coordinate vector set containing multiple edge contour key point coordinate vector groups；
The acquisition of contour curve set is obtained according to the coordinate vector set containing multiple picture profiles to be screened
The contour curve set of curve, each picture contour curve to be screened are corresponding with picture to be screened described in every；
Turning quantity is calculated, calculates separately the turning quantity of each picture contour curve to be screened, and contained
The turning magnitude-set of multiple turning quantity, wherein each turning quantity is corresponding with picture to be screened described in every；
Target sketch figure picture is screened, the maximum turning quantity in the turning magnitude-set is filtered out, and is filtered out and institute
The corresponding picture to be screened of maximum turning quantity is stated, the picture to be screened corresponding with the maximum turning quantity is made
For target sketch figure picture.
Further, before the acquisition picture to be screened further include:
Summarize picture to be screened, the video image drawn by camera acquisition for clip creation, to the video image
It carries out extracting key frame processing, obtains multiple for making the picture to be screened of outline, multiple described pictures to be screened are summarized
For pictures to be screened.
Further, before the picture recognition to be screened further include: establish default edge contour study collection model, specifically
Are as follows:
Picture to be trained is collected, and the multi-angle of the personage of network collection various forms, animals and plants and familiar object is passed through
Picture to be trained obtains the pictures to be trained containing multiple pictures to be trained；
Key point set of coordinates is extracted, the edge contour of the picture to be trained is chosen, extracts the profile of the edge contour
Key point coordinate vector group, wherein reflect each profile key point coordinate vector group and the picture foundation to be trained
Penetrate relationship；
It is defeated to incite somebody to action the picture to be trained, the mapping relations and the profile key point coordinate vector group for model training
Enter to default learning model and be trained, obtains default edge contour study collection model.
Further, the model training specifically: using the pictures to be trained as the input of default learning model
Amount is trained, and is trained the mapping relations as the identified amount of the default learning model, and the profile is crucial
Point coordinate vector group is trained as the output quantity of default learning model, obtains default edge contour study collection model.
Further, the picture recognition to be screened specifically: be input to the processed pictures to be screened default
Edge contour study collection model, the default edge contour study collection Model Matching go out corresponding with the processed picture to be screened
The profile key point coordinate vector group, the default edge contour study collection model is by the profile key point coordinate vector
Group is exported as edge contour key point coordinate vector group, is finally obtained containing multiple edge contour key point coordinate vector groups
Coordinate vector set.
Further, the model training further includes the second gray proces, carries out the second gray scale to the picture to be trained
Processing.
Further, first gray proces specifically: according to weighted mean method in the pictures to be screened
The picture to be screened carries out the first gray proces and obtains the processed figure to be screened containing multiple processed pictures to be screened
Piece collection.
The present invention provides the second purpose and is implemented with the following technical solutions:
A kind of electronic equipment, comprising: processor；
Memory；And program, wherein described program is stored in the memory, and is configured to by processor
It executes, described program includes the screening technique for executing a kind of sketch figure picture of the application.
The present invention provides the third purpose and is implemented with the following technical solutions:
A kind of computer readable storage medium, is stored thereon with computer program, it is characterised in that: the computer program
It is executed by processor a kind of screening technique of sketch figure picture of the application.
Compared with prior art, the beneficial effects of the present invention are a kind of screening technique of sketch figure picture of the invention, outline
Image is for making outline, comprising the following steps: obtains the pictures to be screened for making outline, contains in pictures to be screened
There are multiple pictures to be screened；Gray proces are carried out to the picture to be screened in pictures to be screened and are obtained processed containing multiple
The processed pictures to be screened of picture to be screened；Processed pictures to be screened are input to default edge contour study collection mould
Type is identified, the coordinate vector set containing multiple edge contour key point coordinate vector groups is obtained；According to coordinate vector collection
Conjunction obtains the contour curve set containing multiple picture contour curves to be screened, each picture contour curve to be screened and every to
It is corresponding to screen picture；The turning quantity of each picture contour curve to be screened is calculated separately, and is obtained containing multiple turning quantity
Turning magnitude-set, wherein each turning quantity is corresponding with every picture to be screened；It filters out in the magnitude-set of turning most
Big turning quantity, and picture to be screened corresponding with maximum turning quantity is filtered out, it will be corresponding wait sieve with maximum turning quantity
Select picture as target sketch figure picture.By being identified to obtain edge contour key point coordinate vector group to picture to be screened,
To calculate turning quantity, picture to be screened is filtered out as target sketch figure picture, whole process according to the size of turning quantity
Intelligence screening, and obtained target sketch figure picture is better able to embody the contour feature of corresponding personage or article, more
It is suitable as the picture for making outline.
The above description is only an overview of the technical scheme of the present invention, in order to better understand the technical means of the present invention,
And can be implemented in accordance with the contents of the specification, the following is a detailed description of the preferred embodiments of the present invention and the accompanying drawings.
A specific embodiment of the invention is shown in detail by following embodiment and its attached drawing.
Detailed description of the invention
The drawings described herein are used to provide a further understanding of the present invention, constitutes part of this application, this hair
Bright illustrative embodiments and their description are used to explain the present invention, and are not constituted improper limitations of the present invention.In the accompanying drawings:
Fig. 1 is a kind of flow diagram of the screening technique of sketch figure picture of the invention.
Specific embodiment
In the following, being described further in conjunction with attached drawing and specific embodiment to the present invention, it should be noted that not
Under the premise of conflicting, new implementation can be formed between various embodiments described below or between each technical characteristic in any combination
Example.
As shown in Figure 1, a kind of screening technique of sketch figure picture of the invention, comprising the following steps:
Summarize picture to be screened, the video image drawn by camera acquisition for clip creation carries out video image
Key frame processing is extracted, obtains multiple for making the picture to be screened of outline, it is to be screened that multiple pictures to be screened, which are summarized,
Pictures.In the present embodiment, passes through setting camera and obtain the attending a meeting of meeting scene welcome guest, the photograph of the multiple angles of admission in real time
Piece suitably rotates the direction (random shooting can also with) of head or body in the welcome guest link guidance welcome guest that registers, just can easily
The video image of the multiple angles of welcome guest is got, and video image is carried out to extract key frame processing, extracts wherein each shooting
Image in angle with apparent contour feature, most suitable production outline, obtains multiple for making pictures to be screened of outline.
Picture to be screened is obtained, the pictures to be screened for making outline picture is obtained, contains in pictures to be screened more
Open picture to be screened.
First gray proces carry out the first gray proces to the picture to be screened in pictures to be screened and obtain containing more
Open the processed pictures to be screened of processed picture to be screened.The first gray scale is carried out to picture to be screened according to weighted mean method
Processing can remove extra variegated in image, improve treatment effeciency, make image by carrying out gray proces to picture to be screened
Black-white-gray state is all presented.
Default edge contour study collection model is established, specifically includes the following steps:
Picture to be trained is collected, by the animals and plants of network collection various forms and the multi-angle picture of familiar object,
Obtain pictures to be trained.In the present embodiment, pass through the common items such as the personage of network collection various forms, animal, plant
The picture of all angles, various angles, the form of the picture of each type will collect respectively, for example, the picture of personage, needs
It collects such as stance, sitting posture, various postures of bending over；Hair portion is also required to for various head dummys such as long hair, curly hair, horse hairs
It collects respectively；And animal, such as dog, frizzle and the king-sized kind of undercoat, appearance difference require to collect respectively.
Profile key point coordinate vector is extracted, the edge contour of picture to be trained is chosen, the profile for extracting edge contour closes
Key point coordinate vector group, wherein each profile key point coordinate vector group is established into mapping relations with picture to be trained.First
Picture is marked, and chooses out the edge contour of picture to be trained according to label, extracts each pass of edge contour in order
Key point coordinate, obtains key point set of coordinates；Profile key point corresponding with key point set of coordinates is obtained according to key point set of coordinates
Coordinate vector group.Every picture to be screened all corresponds to a profile key point coordinate vector group.
Picture, mapping relations and profile key point coordinate vector group to be trained are input to default study by model training
Model is trained, and default edge contour study collection model is obtained, using pictures to be trained as the input of default learning model
Amount is trained, and is trained mapping relations as the identified amount of default learning model, by profile key point coordinate vector group
Output quantity as default learning model is trained, and obtains default edge contour study collection model.In the present embodiment, it is inciting somebody to action
Training picture input value is preset before learning model, it is also necessary to be treated trained picture and be carried out the second gray proces, specially basis
Weighted mean method treats trained picture and carries out the second gray proces, can remove figure by treating trained picture progress gray proces
It is extra variegated as in, treatment effeciency is improved, makes image that black-white-gray state all be presented.
Processed pictures to be screened are input to default edge contour study collection model and known by picture recognition to be screened
Not, the coordinate vector set containing multiple edge contour key point coordinate vector groups is obtained.Specifically: by processed figure to be screened
Piece collection is input to default edge contour study collection model, presets edge contour study collection Model Matching and goes out and processed figure to be screened
The corresponding profile key point coordinate vector group of piece, preset edge contour study collection model using profile key point coordinate vector group as
Edge contour key point coordinate vector group output, finally obtain the coordinate containing multiple edge contour key point coordinate vector groups to
Duration set.
The acquisition of contour curve set obtains the wheel containing multiple picture contour curves to be screened according to coordinate vector set
Wide collection of curves, each picture contour curve to be screened are corresponding with every picture to be screened.According to obtained coordinate vector set
In edge contour key point coordinate vector group, by each edge contour key point in edge contour key point coordinate vector group
The corresponding key point of coordinate vector is attached, thus the picture region for the polygon being closed, as each is to be screened
The corresponding contour curve of personage's main body in picture.Each edge contour key point coordinate vector group corresponds to a contour curve.
Turning quantity is calculated, calculates separately the turning quantity of each picture contour curve to be screened, and obtain containing multiple
The turning magnitude-set of turning quantity, wherein each turning quantity is corresponding with every picture to be screened.Calculate each figure to be screened
The number of corners amount of piece contour curve specifically: the arbitrary point on picture contour curve to be screened starts to calculate as starting point, constantly
Judge the direction between adjacent point, successively takes clockwise a little, when the direction of next point and previous point is inconsistent
Then it is determined as an inflection point, in this manner until all the points on screening picture contour curve have been calculated, finally statistics is every
The quantity of the inflection point of a picture contour curve to be screened, the quantity of this inflection point are turning quantity.The inconsistent judgement mark in direction
It is quasi- are as follows: such as A is as starting point, i.e. first point, and B is as second point, and C is as third point, when the direction of B and the direction of A
When different, then B point is inflection point at this time, when C point direction and B direction difference when, using C point as inflection point.It is no
It then, then is not inflection point.
Target sketch figure picture is screened, filters out the maximum turning quantity in the magnitude-set of turning, and filter out and turn with maximum
The corresponding picture to be screened of angle quantity, will picture to be screened corresponding with maximum turning quantity as target sketch figure picture.Finally
Outline production is carried out to current sketch figure picture.
The present invention provides a kind of electronic equipment, comprising: processor；
Memory；And program, wherein program is stored in memory, and is configured to be executed by processor, journey
Sequence includes the screening technique for executing a kind of sketch figure picture of the application.
The present invention also provides a kind of computer readable storage mediums, are stored thereon with computer program, computer program
It is executed by processor a kind of screening technique of sketch figure picture of the application.
A kind of screening technique of sketch figure picture of the invention, sketch figure picture is for making outline, comprising the following steps: obtains
For making the pictures to be screened of outline, contain multiple pictures to be screened in pictures to be screened；To in pictures to be screened
Picture to be screened carry out the first gray proces and obtaining the processed picture to be screened containing multiple processed pictures to be screened
Collection；Processed pictures to be screened are input to default edge contour study collection model to identify, are obtained containing multiple edges
The coordinate vector set of profile key point coordinate vector group；It is obtained according to coordinate vector set containing multiple picture profiles to be screened
The contour curve set of curve, each picture contour curve to be screened are corresponding with every picture to be screened；Calculate separately it is each to
The turning quantity of picture contour curve is screened, and obtains the turning magnitude-set containing multiple turning quantity, wherein each turning
Quantity is corresponding with every picture to be screened；The maximum turning quantity in the magnitude-set of turning is filtered out, and filters out and is turned with maximum
The corresponding picture to be screened of angle quantity, will picture to be screened corresponding with maximum turning quantity as target sketch figure picture.Pass through
Picture to be screened is identified to obtain edge contour key point coordinate vector group, so that turning quantity is calculated, according to number of corners
The size of amount filters out picture to be screened as target sketch figure picture, whole process intelligence screening, and obtained target is cut
Shadow image is better able to embody the contour feature of corresponding personage or article, is more suitable as the picture for making outline.
More than, only presently preferred embodiments of the present invention is not intended to limit the present invention in any form；All current rows
The those of ordinary skill of industry can be shown in by specification attached drawing and above and swimmingly implement the present invention；But all to be familiar with sheet special
The technical staff of industry without departing from the scope of the present invention, is made a little using disclosed above technology contents
The equivalent variations of variation, modification and evolution is equivalent embodiment of the invention；Meanwhile all substantial technologicals according to the present invention
The variation, modification and evolution etc. of any equivalent variations to the above embodiments, still fall within technical solution of the present invention
Within protection scope.
Claims (9)
1. a kind of screening technique of sketch figure picture, the sketch figure picture is for making outline, it is characterised in that: including following step
It is rapid:
Picture to be screened is obtained, the pictures to be screened for making outline is obtained, contains multiple in the pictures to be screened
Picture to be screened；
First gray proces carry out the first gray proces to the picture to be screened in the pictures to be screened and are contained
There are the processed pictures to be screened of multiple processed pictures to be screened；
The processed pictures to be screened are input to default edge contour study collection model and known by picture recognition to be screened
Not, the coordinate vector set containing multiple edge contour key point coordinate vector groups is obtained；
The acquisition of contour curve set is obtained according to the coordinate vector set containing multiple picture contour curves to be screened
Contour curve set, each picture contour curve to be screened is corresponding with picture to be screened described in every；
Turning quantity is calculated, calculates separately the turning quantity of each picture contour curve to be screened, and obtain containing multiple
The turning magnitude-set of turning quantity, wherein each turning quantity is corresponding with picture to be screened described in every；
Screen target sketch figure picture, filter out the maximum turning quantity in the turning magnitude-set, and filter out with it is described most
The corresponding picture to be screened of big turning quantity, will the picture to be screened corresponding with the maximum turning quantity as mesh
Mark sketch figure picture.
2. a kind of screening technique of sketch figure picture as described in claim 1, it is characterised in that: it is described obtain picture to be screened it
Before further include:
Summarize picture to be screened, the video image drawn by camera acquisition for clip creation carries out the video image
Extract key frame processing, obtain multiple for making pictures to be screened of outline, by multiple described pictures to be screened summarize for
Screen pictures.
3. a kind of screening technique of sketch figure picture as described in claim 1, it is characterised in that: the picture recognition to be screened it
Before further include: default edge contour study collection model is established, specifically:
Picture to be trained is collected, and waits instructing by the multi-angle of the personage of network collection various forms, animals and plants and familiar object
Practice picture, obtains the pictures to be trained containing multiple pictures to be trained；
Key point set of coordinates is extracted, the edge contour of the picture to be trained is chosen, the profile for extracting the edge contour is crucial
Point coordinate vector group, wherein each profile key point coordinate vector group is established into mapping with the picture to be trained and is closed
System；
Picture, the mapping relations and the profile key point coordinate vector group to be trained are input to by model training
Default learning model is trained, and obtains default edge contour study collection model.
4. a kind of screening technique of sketch figure picture as claimed in claim 3, it is characterised in that: the model training specifically:
It is trained the pictures to be trained as the input quantity of default learning model, using the mapping relations as described default
The identified amount of learning model is trained, using the profile key point coordinate vector group as the output quantity of default learning model into
Row training obtains default edge contour study collection model.
5. a kind of screening technique of sketch figure picture as claimed in claim 3, it is characterised in that: the picture recognition tool to be screened
Body are as follows: the processed pictures to be screened are input to default edge contour study collection model, the default edge contour
It practises collection Model Matching and goes out the profile key point coordinate vector group corresponding with the processed picture to be screened, the default side
Edge profile study collection model is exported the profile key point coordinate vector group as edge contour key point coordinate vector group, most
The coordinate vector set containing multiple edge contour key point coordinate vector groups is obtained eventually.
6. a kind of screening technique of sketch figure picture as claimed in claim 3, it is characterised in that: the model training further includes the
Two gray proces carry out the second gray proces to the picture to be trained.
7. a kind of screening technique of sketch figure picture as described in claim 1, it is characterised in that: first gray proces are specific
Are as follows: the first gray proces are carried out to the picture to be screened in the pictures to be screened according to weighted mean method and are contained
There are the processed pictures to be screened of multiple processed pictures to be screened.
8. a kind of electronic equipment, characterized by comprising: processor；
Memory；And program, wherein described program is stored in the memory, and is configured to be held by processor
Row, described program include requiring method described in 1-7 any one for perform claim.
9. a kind of computer readable storage medium, is stored thereon with computer program, it is characterised in that: the computer program quilt
Processor executes the method as described in claim 1-7 any one.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811472041.5A CN109784346B (en) | 2018-12-04 | 2018-12-04 | Screening method of silhouette image, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811472041.5A CN109784346B (en) | 2018-12-04 | 2018-12-04 | Screening method of silhouette image, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109784346A true CN109784346A (en) | 2019-05-21 |
| CN109784346B CN109784346B (en) | 2023-10-13 |
Family
ID=66496595
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811472041.5A Active CN109784346B (en) | 2018-12-04 | 2018-12-04 | Screening method of silhouette image, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109784346B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116524592A (en) * | 2023-04-18 | 2023-08-01 | 凯通科技股份有限公司 | Gait sequence silhouette generation method and device, electronic equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101369309A (en) * | 2008-09-26 | 2009-02-18 | 北京科技大学 | Human ear image normalization method based on active apparent model and outer ear long axis |
| CN101464948A (en) * | 2009-01-14 | 2009-06-24 | 北京航空航天大学 | Object identification method for affine constant moment based on key point |
| CN101593277A (en) * | 2008-05-30 | 2009-12-02 | 电子科技大学 | A kind of complicated color image Chinese version zone automatic positioning method and device |
-
2018
- 2018-12-04 CN CN201811472041.5A patent/CN109784346B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101593277A (en) * | 2008-05-30 | 2009-12-02 | 电子科技大学 | A kind of complicated color image Chinese version zone automatic positioning method and device |
| CN101369309A (en) * | 2008-09-26 | 2009-02-18 | 北京科技大学 | Human ear image normalization method based on active apparent model and outer ear long axis |
| CN101464948A (en) * | 2009-01-14 | 2009-06-24 | 北京航空航天大学 | Object identification method for affine constant moment based on key point |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116524592A (en) * | 2023-04-18 | 2023-08-01 | 凯通科技股份有限公司 | Gait sequence silhouette generation method and device, electronic equipment and storage medium |
| CN116524592B (en) * | 2023-04-18 | 2024-02-06 | 凯通科技股份有限公司 | Gait sequence silhouette generation method and device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109784346B (en) | 2023-10-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107292813B (en) | A kind of multi-pose Face generation method based on generation confrontation network | |
| CN105118082B (en) | Individualized video generation method and system | |
| CN109670060A (en) | A kind of remote sensing image semi-automation mask method based on deep learning | |
| CN107844797A (en) | A kind of method of the milking sow posture automatic identification based on depth image | |
| CN109815846A (en) | Image processing method, device, storage medium and electronic device | |
| CN108460398A (en) | Image processing method, device, cloud processing equipment and computer program product | |
| CN107730457A (en) | A kind of image completion method, apparatus, electronic equipment and storage medium | |
| CN108537720A (en) | A kind of image processing method and device | |
| CN107944457A (en) | Drawing object identification and extracting method under a kind of complex scene | |
| CN109117877A (en) | A kind of Pelteobagrus fulvidraco and its intercropping kind recognition methods generating confrontation network based on depth convolution | |
| CN106803257A (en) | The dividing method of scab in a kind of crop disease leaf image | |
| CN109359527A (en) | Hair zones extracting method and system neural network based | |
| CN106447720A (en) | Method for constructing golden-ratio face | |
| CN116649159B (en) | Edible fungus growth parameter optimizing system and method | |
| CN106768081A (en) | A kind of method and system for judging fruits and vegetables growth conditions | |
| CN112990103A (en) | String mining secondary positioning method based on machine vision | |
| CN106651879A (en) | Method and system for extracting nail image | |
| CN109784346A (en) | A kind of screening technique of sketch figure picture, electronic equipment and storage medium | |
| CN113269191A (en) | Crop leaf disease identification method and device and storage medium | |
| CN109934256A (en) | One kind is based on GA-ANN Feature Dimension Reduction and the preferred weeds in paddy field recognition methods of SOM feature | |
| CN111079733A (en) | Data augmentation method based on underwater imaging environment model | |
| CN114155377A (en) | Poultry self-adaptive feeding method based on artificial intelligence and growth cycle analysis | |
| CN114239754A (en) | Pedestrian attribute identification method and system based on attribute feature learning decoupling | |
| CN103984927B (en) | Information processing method and electronic equipment | |
| CN111241908B (en) | Device and method for identifying biological characteristics of young poultry |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |