CN109711390A - Face scratches the preferred method and device of figure picture - Google Patents
Face scratches the preferred method and device of figure picture Download PDFInfo
- Publication number
- CN109711390A CN109711390A CN201910043304.9A CN201910043304A CN109711390A CN 109711390 A CN109711390 A CN 109711390A CN 201910043304 A CN201910043304 A CN 201910043304A CN 109711390 A CN109711390 A CN 109711390A
- Authority
- CN
- China
- Prior art keywords
- face
- scoring
- picture
- area
- calculates
- 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
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses preferred method and device that a kind of face scratches figure picture, wherein this method comprises: carrying out Face datection to video original image, obtains multiple face frame coordinates;Human face region is detected according to default face position, obtains multiple face key points;Facial angle scoring is calculated according to the face key point and the positive face of standard that detect, face area is calculated according to the face frame coordinate detected and the scoring of face area is calculated according to face area and video artwork area;Face center is calculated according to the face frame coordinate position detected and calculates face location scoring with video original image central point;Face picture overall score is calculated according to facial angle scoring, face location scoring and the scoring of face area;It determines to capture and is spaced the highest face picture of interior overall score, and face picture is carried out to scratch figure.The present invention can filter out face than calibration, and the picture that face area is larger and face location is more placed in the middle is able to ascend stingy figure effect.
Description
Technical field
The present invention relates to preferred method, devices, end that technical field of image processing more particularly to a kind of face scratch figure picture
End and readable storage medium storing program for executing.
Background technique
It is monitoring or is capturing in video, generalling use the stingy diagram technology of face and pluck out face figure from the image of sequence of frames of video
Piece.Since the state of pedestrian in video constantly changes, the position of face also can constantly change in video, carry out under scene on daytime
When face snap scratches figure, the picture that face scratches figure is easy to appear the not positive problem of face.In addition, since pedestrian is moved to different location
When, the size and location of pedestrian's face are also different, have by the face picture that face scratches figure smaller, and some comparisons are leaned on
The frame of nearly monitored picture, the effect for causing face to scratch figure are poor.
In view of this, it is necessary to propose that scratching diagram technology to current face is further improved.
Summary of the invention
To solve an above-mentioned at least technical problem, the main object of the present invention is to provide a kind of face and scratches the preferred of figure picture
Method, apparatus, terminal and readable storage medium storing program for executing.
To achieve the above object, first technical solution that the present invention uses are as follows: a kind of face is provided and scratches the excellent of figure picture
Choosing method, comprising:
Face datection is carried out to the video original image of sequence of frames of video, obtains multiple face frame coordinates;
Human face region is detected according to default face position, obtains multiple face key points;
Facial angle scoring is calculated according to the face key point and the positive face of standard that detect, according to the face frame detected
Coordinate calculates face area and calculates the scoring of face area according to face area and video artwork area;According to what is detected
Face frame coordinate position calculates face center and calculates face location scoring with video original image central point;
Face picture overall score is calculated according to facial angle scoring, face location scoring and the scoring of face area;
It determines to capture and is spaced the highest face picture of interior overall score, and face picture is carried out to scratch figure.
Wherein, the basis is preset face position and is detected to human face region, obtains multiple face key points, comprising:
After detecting multiple face frame coordinates, face picture is plucked out from video original image;
Face picture is filled into designated shape in the form of placed in the middle and narrows down to setting resolution ratio;
It is handled using face picture of the nnie neural network accelerating engine to diminution, and according to deep learning nerve net
Network algorithm detects multiple face key points.
It is wherein, described that facial angle scoring is calculated according to face key point, comprising:
In the multiple face key points detected, a face key point is selected as starting point, remaining key point conduct
Multiple groups vector is obtained after terminal, connection source and multiple terminals, and calculates the angle between two adjacent groups vector;
Multiple standard vectors are obtained according to multiple key points of the positive face of standard, and are calculated between two adjacent groups standard vector
Angle;
Go out face with the angle calcu-lation between corresponding two adjacent groups standard vector according to the angle between two adjacent groups vector
Angle scoring.
Wherein, the face frame coordinate that the basis detects calculates face area and according to face area and video original image
Areal calculation goes out the scoring of face area, comprising:
The Width x Height of face is calculated according to face frame coordinate;
Face area is calculated according to face height and width;And
By face area divided by video artwork area, the scoring of face area is calculated.
Wherein, the face frame coordinate position that the basis detects calculate face center and with video original image central point
Calculate face location scoring, comprising:
The center point coordinate and face center coordinate of video original image are calculated according to the face frame coordinate of multiple faces;
The face center of each face is calculated to the distance of video original image central point according to center point coordinate;
To the distance of each face center to video original image central point divided by the cornerwise length of video original image, people is obtained
Face location score.
Wherein, described that face picture is calculated according to facial angle scoring, face location scoring and the scoring of face area
Overall score, specifically:
The face picture general comment is divided into the scoring of a* facial angle, the scoring of b* face area and c* face location scoring three
The sum of person,
Wherein, a, b, c are scoring coefficient, a=10000, b=100, c=1.
To achieve the above object, second technical solution that the present invention uses are as follows: a kind of face is provided and scratches the excellent of figure picture
Screening device, comprising:
First detection module carries out Face datection for the video original image to sequence of frames of video, obtains multiple face frames and sits
Mark;
It is crucial to obtain multiple faces for detecting according to default face position to human face region for second detection module
Point;
Computing module calculates facial angle scoring according to the face key point and the positive face of standard that detect, according to detection
Face frame coordinate out calculates face area and calculates the scoring of face area according to face area and video artwork area;Root
Face center is calculated according to the face frame coordinate position detected and calculates face location scoring with video original image central point;
Statistical module, for being scored according to facial angle, face location scores and the scoring of face area calculates face
Picture overall score;
Confirmation module is spaced the highest face picture of interior overall score for determining to capture, and carries out scratching figure to face picture.
To achieve the above object, the third technical solution that the present invention uses are as follows: a kind of terminal is provided, including memory,
Processor and storage on a memory and the program that can run on a processor, when the processor executes described program in realization
The step of stating method.
To achieve the above object, the 4th technical solution that the present invention uses are as follows: provide a kind of readable storage medium storing program for executing, thereon
It is stored with program, which is characterized in that the step of above-mentioned method is realized when described program is executed by processor.
Technical solution of the present invention mainly using first Face datection is carried out to the video original image of sequence of frames of video, obtains multiple
Then face frame coordinate detects human face region according to default face position, obtains multiple face key points;And according to
The face key point and the positive face of standard detected calculates facial angle scoring, calculates people according to the face frame coordinate detected
Face product simultaneously calculates the scoring of face area according to face area and video artwork area, according to the face frame coordinate bit detected
It sets and calculates face center and calculate face location scoring with video original image central point, then scored according to facial angle,
Face location scoring and the scoring of face area calculate face picture overall score, finally determine and capture overall score highest in interval
Face picture, and face picture is carried out to scratch figure, face can be filtered out than calibration, face area is larger and face location
Compare picture placed in the middle, is able to ascend stingy figure effect.
Detailed description of the invention
Fig. 1 is the method flow diagram for the preferred method that one embodiment of the invention face scratches figure picture;
Fig. 2 is the block diagram for the preferred embodiment that one embodiment of the invention face scratches figure picture;
Fig. 3 is the internal structure chart of one embodiment of the invention terminal.
The embodiments will be further described with reference to the accompanying drawings for the realization, the function and the advantages of the object of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiment is only a part of the embodiments of the present invention, instead of all the embodiments.Base
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts it is all its
His embodiment, shall fall within the protection scope of the present invention.
It is to be appreciated that the description of " first ", " second " involved in the present invention etc. is used for description purposes only, and should not be understood as
Its relative importance of indication or suggestion or the quantity for implicitly indicating indicated technical characteristic.Define as a result, " first ",
The feature of " second " can explicitly or implicitly include at least one of the features.In addition, the technical side between each embodiment
Case can be combined with each other, but must be based on can be realized by those of ordinary skill in the art, when the combination of technical solution
Conflicting or cannot achieve when occur will be understood that the combination of this technical solution is not present, also not the present invention claims guarantor
Within the scope of shield.
Fig. 1 is please referred to, Fig. 1 is the method flow diagram for the preferred method that one embodiment of the invention face scratches figure picture.At this
In inventive embodiments, which scratches the preferred method of figure picture, comprising:
Step S10, Face datection is carried out to the video original image of sequence of frames of video, obtains multiple face frame coordinates;
Step S20, human face region is detected according to default face position, obtains multiple face key points;
Step S30, facial angle scoring is calculated according to the face key point that detects and the positive face of standard, according to detecting
Face frame coordinate calculate face area and the scoring of face area calculated according to face area and video artwork area;According to
The face frame coordinate position detected calculates face center and calculates face location scoring with video original image central point;
Step S40, being scored according to facial angle scoring, face location scoring and face area, it is total to calculate face picture
Scoring;
Step S50, it determines to capture and is spaced the highest face picture of interior overall score, and face picture is carried out to scratch figure.
Sequence of frames of video in the present embodiment is to think camera on hi3559 platform based on sea to be captured, the lattice of video frame
Formula is NV21, video resolution 960*540.Have plurality of pictures in one sequence of frames of video, each video original image be both needed into
Row Face datection obtains multiple face frame coordinates.Later further according to face frame coordinate from the 960*540 original image of the NV21 of video
Deduct face picture, data format NV21.Human face region is detected according to default face position, multiple faces is obtained and closes
Key point.Multiple face key points can be left eye center, right eye center, nose, mouth left comer and mouth right corner etc..Then according to detection
Face key point and the positive face of standard out calculates facial angle scoring, calculates face face according to the face frame coordinate detected
Product simultaneously calculates the scoring of face area according to face area and video artwork area.According to the face frame coordinate position meter detected
It calculates face center and calculates face location scoring with video original image central point, further, according to the sum of the weighting of three
Overall score is calculated, most preferred picture scheme is obtained.It finally determines to capture and is spaced the highest face picture of interior overall score, and is right
Face picture carries out scratching figure.
Technical solution of the present invention mainly using first Face datection is carried out to the video original image of sequence of frames of video, obtains multiple
Then face frame coordinate detects human face region according to default face position, obtains multiple face key points;And according to
The face key point and the positive face of standard detected calculates facial angle scoring, calculates people according to the face frame coordinate detected
Face product simultaneously calculates the scoring of face area according to face area and video artwork area, according to the face frame coordinate bit detected
It sets and calculates face center and calculate face location scoring with video original image central point, then scored according to facial angle,
Face location scoring and the scoring of face area calculate face picture overall score, finally determine and capture overall score highest in interval
Face picture, and face picture is carried out to scratch figure, face can be filtered out than calibration, face area is larger and face location
Compare picture placed in the middle, is able to ascend stingy figure effect.
In a specific embodiment, the basis is preset face position and is detected to human face region, obtains multiple
Face key point, comprising:
After detecting multiple face frame coordinates, face picture is plucked out from video original image;
Face picture is filled into designated shape in the form of placed in the middle and narrows down to setting resolution ratio;
It is handled using face picture of the nnie neural network accelerating engine to diminution, and according to deep learning nerve net
Network algorithm detects multiple face key points.
In the present embodiment, the quantity of each face frame coordinate is specially 4, can be selected by 4 people's face frame coordinates
Face frame coordinate.The face frame that 4 face frame coordinates are surrounded is rectangle, and face figure then can be plucked out from video original image
Piece.Face picture is filled into designated shape in the form of placed in the middle and narrows down to setting resolution ratio.Designated shape is preferably square,
Face picture is located at the centre of square.Filled face picture can narrow down to the resolution ratio of critical point detection model.
It is described that facial angle scoring is calculated according to face key point in a specific embodiment, comprising:
In the multiple face key points detected, a face key point is selected as starting point, remaining key point conduct
Multiple groups vector is obtained after terminal, connection source and multiple terminals, and calculates the angle between two adjacent groups vector;
Multiple standard vectors are obtained according to multiple key points of the positive face of standard, and are calculated between two adjacent groups standard vector
Angle;
Go out face with the angle calcu-lation between corresponding two adjacent groups standard vector according to the angle between two adjacent groups vector
Angle scoring.
In the present embodiment, multiple key points have 5, and respectively left eye center, right eye center, nose, mouth left comer and mouth is right
Angle., can be using nose key point as starting point in actual calculating, other key points are terminal, and obtain 4 vectors: nose closes
Key point is to left eye center key point vector, nose key point to right eye center key point vector, and nose key point to mouth left comer is closed
Key point vector, angle of the nose key point to mouth the right key point vector, between then calculating these vectors two-by-two.The positive face of standard
It is handled in the manner described above, obtains the angle two-by-two of the positive face of standard.According between two adjacent groups vector angle with it is corresponding
Angle calcu-lation between two adjacent groups standard vector goes out facial angle scoring.Face location can also be normalized,
The ratio of the two is closer to 1, then it represents that face is more just.
In a specific embodiment, the face frame coordinate that the basis detects calculates face area and according to people
Face is long-pending to calculate the scoring of face area with video artwork area, comprising:
The Width x Height of face is calculated according to face frame coordinate;
Face area is calculated according to face height and width;And
By face area divided by video artwork area, the scoring of face area is calculated.
In the present embodiment, face frame coordinate has 4, and 4 face frame coordinates surround rectangle, can be with by face frame coordinate
The width and length of rectangle are calculated, to obtain face area, by face area divided by video artwork area, calculates face face
Product scoring.
In a specific embodiment, the face frame coordinate position that the basis detects calculates face center simultaneously
Face location scoring is calculated with video original image central point, comprising:
The center point coordinate and face center coordinate of video original image are calculated according to the face frame coordinate of multiple faces;
According to center point coordinate calculate separately out face center to video original image central point distance;
The distance of face center to video original image central point divided by the cornerwise length of video original image, face position is obtained
It comments on point.
In the present embodiment, the distance of face center to video original image central point divided by the cornerwise length of video original image
Degree further includes after obtaining the step of face location scores, and face according to the algorithm for expanding Cheng great Tu, obtains expanding Cheng great Tu's
Coordinate (for example expanding according to 1.5 times, face figure Width x Height is expanded into 1.5 times), if the coordinate for expanding Cheng great Tu has been more than view
The range of frequency original image, then directly allowing face location scoring is 0.
It is described to be commented according to facial angle scoring, face location scoring and face area in a specific embodiment
Divide and calculate face picture overall score, specifically:
The face picture general comment is divided into the scoring of a* facial angle, the scoring of b* face area and c* face location scoring three
The sum of person,
Wherein, a, b, c are scoring coefficient, a=10000, b=100, c=1.
A calculation formula of the present embodiment is provided below, specific as follows:
Face picture overall score=facial angle * 10000+ face area that finally scores finally scores * 100+ face location
Final scoring.
Referring to figure 2., Fig. 2 is the block diagram for the preferred embodiment that one embodiment of the invention face scratches figure picture.This hair
In bright embodiment, which scratches the preferred embodiment of figure picture, comprising:
First detection module 10 carries out Face datection for the video original image to sequence of frames of video, obtains multiple face frames
Coordinate;
Second detection module 20 obtains multiple faces and closes for being detected according to default face position to human face region
Key point;
Computing module 30 calculates facial angle scoring according to the face key point and the positive face of standard that detect, according to inspection
The face frame coordinate measured calculates face area and calculates the scoring of face area according to face area and video artwork area;
It calculates face center according to the face frame coordinate position detected and calculates face location with video original image central point and comment
Point;
Statistical module 40, for being scored according to facial angle, face location scores and the scoring of face area calculates people
Face picture overall score;
Confirmation module 50 is spaced the highest face picture of interior overall score for determining to capture, and carries out to face picture stingy
Figure.
In the present embodiment, sequence of frames of video is to think camera on hi3559 platform based on sea to be captured, the lattice of video frame
Formula is NV21, video resolution 960*540.There is plurality of pictures in one sequence of frames of video, each video original image passes through
First detection module 10 carries out Face datection, obtains multiple face frame coordinates.Later again by the second detection module 20, according to people
Face frame coordinate deducts face picture, data format NV21 from the 960*540 original image of the NV21 of video.According to default face
Human face region is detected in position, obtains multiple face key points.Multiple face key points can be in left eye center, right eye
The heart, nose, mouth left comer and mouth right corner etc..Then by computing module 30, according to the face key point and the positive face of standard detected
Facial angle scoring is calculated, face area is calculated according to the face frame coordinate detected and according to face area and video original
The area of pictural surface calculates the scoring of face area.According to the face frame coordinate position detected calculate face center and with video original
Figure central point calculates face location scoring, further, calculates general comment according to the sum of the weighting of three by statistical module 40
Point, obtain most preferred picture scheme.Finally, determine by determining module and capture the highest face picture of overall score in interval,
And face picture is carried out to scratch figure.
Referring to figure 3., Fig. 3 is the internal structure chart of one embodiment of the invention terminal.In one embodiment, which includes
Processor, memory and the network interface connected by system bus.Wherein, the processor of the terminal is calculated and is controlled for providing
Ability processed.The memory of the terminal includes non-volatile memory medium, built-in storage.The non-volatile memory medium is stored with behaviour
Make system, program and database.The built-in storage provides for the operation of operating system and program in non-volatile memory medium
Environment.The network interface of the terminal is used to communicate with external terminal by network connection.When the program is executed by processor with
Realize that a kind of face scratches the preferred method of figure picture.
It will be understood by those skilled in the art that structure shown in Fig. 3, only part relevant to application scheme is tied
The block diagram of structure, does not constitute the restriction for the terminal being applied thereon to application scheme, and specific terminal may include than figure
Shown in more or fewer components, perhaps combine certain components or with different component layouts.
In one embodiment, a kind of terminal is provided, including memory, processor and storage are on a memory and can be
The program run on processor, processor realize the step in above each embodiment of the method when executing program.
In one embodiment, a kind of readable storage medium storing program for executing is provided, program is stored thereon with, program is executed by processor
Step in each embodiment of the method for Shi Shixian or more.
Those of ordinary skill in the art will appreciate that realizing all or part of the process in above-described embodiment method, being can be with
Relevant hardware is instructed to complete by program, the program can be stored in a non-volatile readable and take in storage medium,
The program is when being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein, in each embodiment provided herein
Used any reference to memory, storage, database or other media, may each comprise non-volatile and/or volatibility
Memory.Nonvolatile memory may include read-only memory (ROM), programming ROM (PROM), electrically programmable ROM
(EPROM), electrically erasable ROM (EEPROM) or flash memory.Volatile memory may include random access memory (RAM)
Or external cache.By way of illustration and not limitation, RAM is available in many forms, such as static state RAM (SRAM),
It is dynamic ram (DRAM), synchronous dram (SDRAM), double data rate sdram (DDRSDRAM), enhanced SDRAM (ESDRAM), same
Walk link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) directly RAM (RDRAM), direct memory bus
Dynamic ram (DRDRAM) and memory bus dynamic ram (RDRAM) etc..
The above description is only a preferred embodiment of the present invention, is not intended to limit the scope of the invention, all at this
Under the inventive concept of invention, using equivalent structure transformation made by description of the invention and accompanying drawing content, or directly/use indirectly
It is included in other related technical areas in scope of patent protection of the invention.
Claims (9)
1. the preferred method that a kind of face scratches figure picture, which is characterized in that the face scratches the preferred method of figure picture, comprising:
Face datection is carried out to the video original image of sequence of frames of video, obtains multiple face frame coordinates;
Human face region is detected according to default face position, obtains multiple face key points;
Facial angle scoring is calculated according to the face key point and the positive face of standard that detect, according to the face frame coordinate detected
It calculates face area and the scoring of face area is calculated according to face area and video artwork area;According to the face detected
Frame coordinate position calculates face center and calculates face location scoring with video original image central point;
Face picture overall score is calculated according to facial angle scoring, face location scoring and the scoring of face area;
It determines to capture and is spaced the highest face picture of interior overall score, and face picture is carried out to scratch figure.
2. the preferred method that face as described in claim 1 scratches figure picture, which is characterized in that the basis presets face position
Human face region is detected, multiple face key points are obtained, comprising:
After detecting multiple face frame coordinates, face picture is plucked out from video original image;
Face picture is filled into designated shape in the form of placed in the middle and narrows down to setting resolution ratio;
It is handled using face picture of the nnie neural network accelerating engine to diminution, and is calculated according to deep learning neural network
Method detects multiple face key points.
3. the preferred method that face as claimed in claim 2 scratches figure picture, which is characterized in that described according to face key point meter
Calculate facial angle scoring, comprising:
In the multiple face key points detected, select a face key point as starting point, remaining key point as terminal,
Multiple groups vector is obtained after connection source and multiple terminals, and calculates the angle between two adjacent groups vector;
Multiple standard vectors are obtained according to multiple key points of the positive face of standard, and calculate the folder between two adjacent groups standard vector
Angle;
Go out facial angle with the angle calcu-lation between corresponding two adjacent groups standard vector according to the angle between two adjacent groups vector
Scoring.
4. the preferred method that face as claimed in claim 3 scratches figure picture, which is characterized in that the face that the basis detects
Frame coordinate calculates face area and calculates the scoring of face area according to face area and video artwork area, comprising:
The Width x Height of face is calculated according to face frame coordinate;
Face area is calculated according to face height and width;And
By face area divided by video artwork area, the scoring of face area is calculated.
5. face as claimed in claim 4 scratches the preferred method of figure picture, which is characterized in that the face that the basis detects
Frame coordinate position calculates face center and calculates face location scoring with video original image central point, comprising:
The center point coordinate and face center coordinate of video original image are calculated according to the face frame coordinate of multiple faces;
The face center of each face is calculated to the distance of video original image central point according to center point coordinate;
To the distance of each face center to video original image central point divided by the cornerwise length of video original image, face position is obtained
It comments on point.
6. the preferred method that face as claimed in claim 5 scratches figure picture, which is characterized in that described to be commented according to facial angle
Point, face location scoring and face area scoring calculate face picture overall score, specifically:
The face picture general comment be divided into a* facial angle scoring, b* face area scoring and c* face location scoring three it
With,
Wherein, a, b, c are scoring coefficient, a=10000, b=100, c=1.
7. the preferred embodiment that a kind of face scratches figure picture, which is characterized in that the face scratches the preferred embodiment of figure picture, comprising:
First detection module carries out Face datection for the video original image to sequence of frames of video, obtains multiple face frame coordinates;
Second detection module obtains multiple face key points for detecting according to default face position to human face region;
Computing module calculates facial angle scoring according to the face key point and the positive face of standard that detect, according to what is detected
Face frame coordinate calculates face area and calculates the scoring of face area according to face area and video artwork area;According to inspection
The face frame coordinate position measured calculates face center and calculates face location scoring with video original image central point;
Statistical module, for being scored according to facial angle, face location scores and the scoring of face area calculates face picture
Overall score;
Confirmation module is spaced the highest face picture of interior overall score for determining to capture, and carries out scratching figure to face picture.
8. a kind of terminal including memory, processor and stores the program that can be run on a memory and on a processor, special
The step of sign is, the processor realizes any one of claims 1 to 6 the method when executing described program.
9. a kind of readable storage medium storing program for executing, is stored thereon with program, which is characterized in that realize power when described program is executed by processor
Benefit require any one of 1 to 6 described in method the step of.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910043304.9A CN109711390A (en) | 2019-01-17 | 2019-01-17 | Face scratches the preferred method and device of figure picture |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910043304.9A CN109711390A (en) | 2019-01-17 | 2019-01-17 | Face scratches the preferred method and device of figure picture |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109711390A true CN109711390A (en) | 2019-05-03 |
Family
ID=66262290
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910043304.9A Pending CN109711390A (en) | 2019-01-17 | 2019-01-17 | Face scratches the preferred method and device of figure picture |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109711390A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110765913A (en) * | 2019-10-15 | 2020-02-07 | 浙江大华技术股份有限公司 | Human body target optimization method and device based on multiple evaluation indexes and storage medium |
CN111325157A (en) * | 2020-02-24 | 2020-06-23 | 高新兴科技集团股份有限公司 | Face snapshot method, computer storage medium and electronic device |
CN111741220A (en) * | 2020-07-06 | 2020-10-02 | 江苏云从曦和人工智能有限公司 | Image acquisition method, system, device and medium |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106446851A (en) * | 2016-09-30 | 2017-02-22 | 厦门大图智能科技有限公司 | Visible light based human face optimal selection method and system |
CN108491784A (en) * | 2018-03-16 | 2018-09-04 | 南京邮电大学 | The identification in real time of single feature towards large-scale live scene and automatic screenshot method |
CN108765351A (en) * | 2018-05-31 | 2018-11-06 | Oppo广东移动通信有限公司 | Image processing method, device, electronic equipment and storage medium |
CN108960087A (en) * | 2018-06-20 | 2018-12-07 | 中国科学院重庆绿色智能技术研究院 | A kind of quality of human face image appraisal procedure and system based on various dimensions evaluation criteria |
-
2019
- 2019-01-17 CN CN201910043304.9A patent/CN109711390A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106446851A (en) * | 2016-09-30 | 2017-02-22 | 厦门大图智能科技有限公司 | Visible light based human face optimal selection method and system |
CN108491784A (en) * | 2018-03-16 | 2018-09-04 | 南京邮电大学 | The identification in real time of single feature towards large-scale live scene and automatic screenshot method |
CN108765351A (en) * | 2018-05-31 | 2018-11-06 | Oppo广东移动通信有限公司 | Image processing method, device, electronic equipment and storage medium |
CN108960087A (en) * | 2018-06-20 | 2018-12-07 | 中国科学院重庆绿色智能技术研究院 | A kind of quality of human face image appraisal procedure and system based on various dimensions evaluation criteria |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110765913A (en) * | 2019-10-15 | 2020-02-07 | 浙江大华技术股份有限公司 | Human body target optimization method and device based on multiple evaluation indexes and storage medium |
CN111325157A (en) * | 2020-02-24 | 2020-06-23 | 高新兴科技集团股份有限公司 | Face snapshot method, computer storage medium and electronic device |
CN111741220A (en) * | 2020-07-06 | 2020-10-02 | 江苏云从曦和人工智能有限公司 | Image acquisition method, system, device and medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10783354B2 (en) | Facial image processing method and apparatus, and storage medium | |
JP6688277B2 (en) | Program, learning processing method, learning model, data structure, learning device, and object recognition device | |
US8131013B2 (en) | Method and detecting system for positions of facial parts | |
CN105740780B (en) | Method and device for detecting living human face | |
CN110431404A (en) | Surface defect inspection method and surface defect inspection apparatus | |
CN110309876A (en) | Object detection method, device, computer readable storage medium and computer equipment | |
CN109711390A (en) | Face scratches the preferred method and device of figure picture | |
US9639950B2 (en) | Site estimation device, site estimation method, and site estimation program | |
US10861128B2 (en) | Method of cropping an image, an apparatus for cropping an image, a program and a storage medium | |
CN111091075A (en) | Face recognition method and device, electronic equipment and storage medium | |
EP3441907A1 (en) | Face similarity evaluation method and electronic device | |
WO2022002262A1 (en) | Character sequence recognition method and apparatus based on computer vision, and device and medium | |
CN106033539A (en) | Meeting guiding method and system based on video face recognition | |
CN110796709A (en) | Method and device for acquiring size of frame number, computer equipment and storage medium | |
TWI498857B (en) | Dozing warning device | |
CN112017231A (en) | Human body weight identification method and device based on monocular camera and storage medium | |
CN107194364B (en) | Huffman-L BP multi-pose face recognition method based on divide and conquer strategy | |
CN109785343A (en) | Face based on clarity scratches the preferred method and device of figure picture | |
US9959635B2 (en) | State determination device, eye closure determination device, state determination method, and storage medium | |
CN109815772A (en) | Fingerprint enhancement, recognition methods, device and Fingerprint enhancement identifying system | |
CN110852220A (en) | Intelligent recognition method of facial expression, terminal and computer readable storage medium | |
CN112507766A (en) | Face image extraction method, storage medium and terminal equipment | |
CN113837020B (en) | Cosmetic progress detection method, device, equipment and storage medium | |
JP2008084109A (en) | Eye opening/closing determination device and eye opening/closing determination method | |
JP7152651B2 (en) | Program, information processing device, and information processing 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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190503 |
|
RJ01 | Rejection of invention patent application after publication |