CN110097001A - Generate method, system, equipment and the storage medium of best plurality of human faces image - Google Patents
Generate method, system, equipment and the storage medium of best plurality of human faces image Download PDFInfo
- Publication number
- CN110097001A CN110097001A CN201910358663.3A CN201910358663A CN110097001A CN 110097001 A CN110097001 A CN 110097001A CN 201910358663 A CN201910358663 A CN 201910358663A CN 110097001 A CN110097001 A CN 110097001A
- Authority
- CN
- China
- Prior art keywords
- face
- image
- human faces
- quality
- human
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/217—Validation; Performance evaluation; Active pattern learning techniques
- G06F18/2193—Validation; Performance evaluation; Active pattern learning techniques based on specific statistical tests
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/172—Classification, e.g. identification
Abstract
The present invention provides a kind of method, system, equipment and storage medium for generating best plurality of human faces image, this method comprises: obtaining multiple the plurality of human faces images being continuously shot, and carries out global quality multiple described plurality of human faces images and analyzes to obtain global optimized image;Face datection and face quality analysis are carried out multiple described plurality of human faces images, identify the facial image of the best face quality corresponding to everyone;The face image synthesis of best face quality corresponding to everyone is obtained into best plurality of human faces image into global optimized image.For using PS means, global image optimal quality can be obtained, is not necessarily to artificial subjective judgement and later period PS, substantially increases the effect and efficiency of shooting group photo.
Description
Technical field
The present invention relates to technical field of image processing, more particularly to a kind of method for generating best plurality of human faces image, are
System, equipment and storage medium.
Background technique
During more people group photo, generally require that all faces is allowed all to receive uniform and direction according to lighting conditions
Consistent illumination, and owner can be allowed to be arranged in specific formation, however can allow owner or smiling or select other
Suitable moulding and expression.Also, occur closing one's eyes in order to prevent, expression is strange, head rotation when the motion blur that occurs of face
The problems such as, it can also require that " the pendulum that the personnel that are taken in the moulding and expression being kept fixed in former seconds of tripping, that is, are commonly called as
pose”。
However, e.g., group picture is just more difficult to ensure that everyone is tripping as the personnel amount being taken increases
Moment all at " pendulum pose ", just often there is groups of people's eye closing, the strange, motion blur of expression etc. in plurality of human faces group photo and ask
The situations such as topic, existing conventional treatment mode by photographic device continuous shooting multiple, therefrom the preferable photo of screening effect carries out PS.
Summary of the invention
In view of the foregoing deficiencies of prior art, best plurality of human faces image is generated the purpose of the present invention is to provide a kind of
Method, system, equipment and storage medium, using artificial PS work when for solving to generate best plurality of human faces image in the prior art
Measure big problem.
In order to achieve the above objects and other related objects, the application's in a first aspect, the present invention provide it is a kind of generate it is best
The method of plurality of human faces image, comprising:
Multiple the plurality of human faces images being continuously shot are obtained, and global quality is carried out multiple described plurality of human faces images and is analyzed
To global optimized image;
Face datections and face quality analysis are carried out multiple described plurality of human faces images, are identified corresponding to everyone most
The facial image of beautiful woman's face quality;
The face image synthesis of best face quality corresponding to everyone is obtained into best more people into global optimized image
Face image.
The second aspect of the application provides a kind of system for generating best plurality of human faces image, comprising:
Image overall quality analysis module, for obtaining multiple the plurality of human faces images being continuously shot, and it is described more multiple
Facial image carries out global quality and analyzes to obtain global optimized image;
Face quality analysis module, for carrying out Face datection and face quality analysis multiple described plurality of human faces images,
Identify the facial image of the best face quality corresponding to everyone;
Human face segmentation module, for most preferably scheming the face image synthesis of best face quality corresponding to everyone to global
Best plurality of human faces image is obtained as in.
The third aspect of the application, provides a kind of electronic equipment, comprising:
One or more processors;
Memory;And
One or more programs, wherein one or more of programs be stored in the memory and be configured as by
One or more of processors execute instruction, and execute instruction described in one or more of processors execution so that the electronics
The method that equipment executes the above-mentioned best plurality of human faces image of generation.
The fourth aspect of the application provides a kind of storage medium, comprising:
The storage medium includes the program of storage, wherein described program realizes above-mentioned generation in called execute
The method of best plurality of human faces image.
As described above, method, system, equipment and the storage medium of the best plurality of human faces image of generation of the invention, have with
It is lower the utility model has the advantages that
By using Face datection and face recognition algorithms, come be accurately positioned with all faces in tag image, and lead to
It crosses using face analysis algorithm and detects the face quality point of each face in each image, realize automatically in image
Quick, effective detection of face quality;Using human face segmentation algorithm by everyone the highest people of face quality in multiple images
Face ROI image is synthesized to global optimized image and obtains best plurality of human faces image, for using PS means, can obtain complete
Office's picture quality is optimal, is not necessarily to artificial subjective judgement and later period PS, substantially increases the effect and efficiency of shooting group photo.
Detailed description of the invention
Fig. 1 is shown as a kind of method flow diagram for generating best plurality of human faces image provided by the invention;
Fig. 2 is shown as step S1 flow chart in a kind of method for generating best plurality of human faces image provided by the invention;
Fig. 3 is shown as step S2 flow chart in a kind of method for generating best plurality of human faces image provided by the invention;
Fig. 4 is shown as step S3 flow chart in a kind of method for generating best plurality of human faces image provided by the invention;
Fig. 5 is shown as a kind of system structure diagram for generating best plurality of human faces image provided by the invention;
Fig. 6 is shown as a kind of system complete structure block diagram for generating best plurality of human faces image provided by the invention;
Fig. 7 is shown as a kind of electronic devices structure block diagram for generating best plurality of human faces image provided by the invention.
Specific embodiment
Presently filed embodiment is illustrated by particular specific embodiment below, those skilled in the art can be by this explanation
Content disclosed by book understands other advantages and effect of the application easily.
In described below, with reference to attached drawing, attached drawing describes several embodiments of the application.It should be appreciated that also can be used
Other embodiments, and can be carried out without departing substantially from spirit and scope of the present disclosure mechanical composition, structure, electrically with
And the operational detailed description changed below should not be considered limiting, and the range of embodiments herein
Only the limited of claims of the patent by announcing term used herein is merely to describe specific embodiment, and be not
It is intended to limit the application.The term of space correlation, for example, "upper", "lower", "left", "right", " following ", " lower section ", " lower part ",
" top ", " top " etc. can be used in the text in order to an elements or features and another element or spy shown in explanatory diagram
The relationship of sign.
Although term first, second etc. are used to describe various elements herein in some instances, these elements
It should not be limited by these terms.These terms are only used to distinguish an element with another element.For example, first is pre-
If threshold value can be referred to as the second preset threshold, and similarly, the second preset threshold can be referred to as the first preset threshold, and
The range of various described embodiments is not departed from.First preset threshold and preset threshold are to describe a threshold value, still
Unless context otherwise explicitly points out, otherwise they are not the same preset thresholds.Similar situation further includes first
Volume and the second volume.
Furthermore as used in herein, singular " one ", "one" and "the" are intended to also include plural number shape
Formula, unless having opposite instruction in context it will be further understood that term "comprising", " comprising " show that there are the spies
Sign, step, operation, element, component, project, type, and/or group, but it is not excluded for one or more other features, step, behaviour
Work, element, component, project, the presence of type, and/or group, appearance or addition term "or" used herein and "and/or" quilt
Be construed to inclusive, or mean any one or any combination therefore, " A, B or C " or " A, B and/or C " mean " with
Descend any one: A;B;C;A and B;A and C;B and C;A, B and C " is only when element, function, step or the combination of operation are in certain modes
Under it is inherently mutually exclusive when, just will appear the exception of this definition.
Referring to Fig. 1, being a kind of method flow diagram for generating best plurality of human faces image provided by the invention, comprising:
Step S1 obtains multiple the plurality of human faces images being continuously shot, and carries out global matter multiple described plurality of human faces images
Amount analysis obtains global optimized image;
Step S2 carries out Face datection and face quality analysis multiple described plurality of human faces images, identifies everyone institute
The facial image of corresponding best face quality;
Step S3 obtains the face image synthesis of best face quality corresponding to everyone most into global optimized image
Good plurality of human faces image.
In the present embodiment, relative to traditional PS mode, on the one hand, the present invention design have evaluation criterion, be not necessarily to based on
See judgement, select global optimized image, best face quality facial image when, accuracy is high, on the other hand, without artificial
It is synthesized using PS, regardless of number limitation in plurality of human faces image, generates the ultrahigh in efficiency of best plurality of human faces image.
Referring to Fig. 2, being step S1 flow chart in a kind of method for generating best plurality of human faces image provided by the invention, in detail
It states as follows:
Multiple plurality of human faces images are continuously shot using photographic device;
For example, photographic device can be smart phone, camera or digital camera etc., carried out using the camera of smart phone
Continuous shooting obtains the plurality of human faces image that multiple include multiple faces;
Using quality score algorithm according to the fuzziness of image overall, illumination consistency and formation symmetry to every more people
Face image quality score;
For example, quality score algorithm is image overall analysis algorithm;
Global quality scoring in all plurality of human faces images highest is considered as global optimized image.
Specifically, step S201 chooses first image;
First figure is arbitrarily selected from multiple plurality of human faces images of input.
Step S202 calculates image overall quality point;
Using image overall analysis algorithm, quality information in analysis picture other than face quality, for example (,) it is fuzzy
Degree, light source consistency, formation symmetry obtain image overall quality point.
Image overall analysis algorithm can select one or more algorithm combinations according to quality information, such as: fuzziness
The way of realization of parser includes but is not limited to gradient function, Laplace function;Light source consistency analysis algorithm and formation pair
The way of realization of title property algorithm, for example, whether having face in analysis picture, and obtain institute firstly, user's face detection algorithm
There are position, size and the angle of face, using face quality score algorithm, calculates the face quality point of all faces;It is comprehensive
Size, position and the face quality of face point, calculates the weight of all faces;Comprehensive all face quality point and face power
Weight, calculates face comprehensive score, secondly, calculating the scoring of light source consistency according to light source consistency analysis algorithm, being come with this
Whether the light-source angle judged on every face is consistent;Again, according to formation parser, formation scoring is calculated, is come with this
Judge whether all faces have certain symmetric arrays mode;Finally, score in conjunction with face comprehensive score, light source consistency,
Formation scoring, calculates the overall score of plurality of human faces image, that is, it is highest for global optimized image to obtain overall score.
Step S203, if the judgement that image procossing is completed;
For the image sequence of processing is completed, global optimized image is therefrom chosen;For the image sequence of untreated completion
Column select next image to carry out new analysis.
Step S204 chooses next image;
For the image sequence of untreated completion, next image is selected to carry out new analysis.
Step S205 selects global optimized image;
From all images of input, divides that image tagged of highest for global optimized image image overall quality, use
Make the primary image used when subsequent composograph.
In the present embodiment, it is by the way that global quality scoring is highest in multiple plurality of human faces images of quality score algorithm evaluation
Global optimized image enables the primary image of subsequent synthesis to be optimal, and compares PS mode, there is a set of evaluation criterion, nothing
Need artificial subjective judgement.
Referring to Fig. 3, being step S2 flow chart in a kind of method for generating best plurality of human faces image provided by the invention, in detail
It states as follows:
The human face region that the plurality of human faces image is detected using Face datection algorithm obtains owning in every plurality of human faces image
The image coordinate and facial size of face;
The face quality that the plurality of human faces image is detected using face analysis algorithm is obtained in every plurality of human faces image
The face quality of each face point;
Divide associated storage to face with face quality the human face region in multiple described plurality of human faces images according to face ID
Library;
Wherein, face database saves in multiple plurality of human faces images about the multiple human face regions and face of same face ID
Quality point, or, face database, which saves same face ID by update mode, only has the highest human face region of face quality point, it is subsequent logical
Face database is called directly after crossing recognition of face.
It identifies each face ID in the plurality of human faces image, the highest face of face quality point is matched according to the face ID
Region is considered as the facial image of best face quality.
Specifically, step S301 chooses first image;
First figure is arbitrarily selected from multiple plurality of human faces images of input.
Step S302, Face datection;
User's face detection algorithm detects all human face regions occurred in picture to be verified, obtains each in present image
The image coordinate and facial size of face.Wherein, Face datection algorithm includes but is not limited to AdaBoost, deep learning calculation
Method.
Step S303, face quality analysis;
The quality that all faces occurred in picture to be verified are analyzed using face analysis algorithm, obtains present image
In each face face quality point.
Face analysis algorithm includes deep learning algorithm, further include whether closed one's eyes using face, facial contour and five
Official's clarity calculates comprehensive point of rule-based algorithm.For example, firstly, being pre-processed (Face datection, face figure to facial image
As quality testing, image segmentation), secondly, being scored facial image (uses Expression analysis algorithm, face in analysis picture
It is whether pleasant, and pleasant degree, obtain human face expression score item;The folding situation of right and left eyes is identified respectively, so
After be converted to eyes folding score item;Colour of skin abnormal case (including bloom, yin-yang are identified using colour of skin Outlier Detection Algorithm
The situations such as face, overexposure, colour of skin dark), it is then converted into colour of skin score item), face is obtained by counting the sum of each score item
Quality point.
Step S304, if be the judgement of first image;
For first image, system establishes face database, realizes the purpose of face database data initialization;Non- first is schemed
Picture has the data in the image before of each face, therefore need to carry out people in this step since face database is initialized
Face identification, and update the data of corresponding face ID.
Step S305, recognition of face;
For the face images obtained in the S304, system is looked into face database using face recognition algorithms
It looks for, obtains each face corresponding face ID in face database.
Face recognition algorithms include but is not limited to deep neural network algorithm, template matching algorithm.
Step S306 updates face database;
All faces ROI image obtained in S302 and corresponding face quality point are stored in face database by system, and
And face ROI image is corresponding with the face ID that step S308 is obtained.
Step S307, establishes face database;
Face images obtained in S302 are saved and numbered, as face ROI (Region Of
Interest, region of interest) image and face ID (IDentity, identity number).In addition, to each obtained in S305
Face quality point and corresponding picture number are saved in face database.
Step S308, if the judgement that image procossing is completed;
For the image sequence of processing is completed, the facial image of best face quality is therefrom chosen;For untreated complete
At image sequence, select next image carry out new analysis.
Step S309 chooses next image;
For the image sequence of untreated completion, next image is selected to carry out new analysis.
In the present embodiment, quality analysis is carried out by faces all from multiple plurality of human faces images, screens each face
The facial image of the corresponding best face quality of ID, so that the human face data source of subsequent synthesis is most preferably, largely to save PS
It time, not only ensure that the quality of face, but also improved the efficiency of subsequent composograph.
Referring to Fig. 4, being step S3 flow chart in a kind of method for generating best plurality of human faces image provided by the invention, in detail
It states as follows:
Identify face ID in global optimized image, and by the face ID match in face database face quality divide it is highest
Face ROI image;
Every face in global optimized image is substituted for by face ID by face quality point highest using human face segmentation algorithm
Face ROI image obtain best plurality of human faces image.
Specifically, step S401 chooses global optimized image;
Wherein, select overall situation optimized image obtained in S205 as base image to be synthesized.
Step S402, Face datection;
Identical with step S302, user's face detection algorithm detects all human face regions occurred in picture to be verified, obtains
The coordinate and facial size of each facial image into present image.Wherein, Face datection algorithm includes but is not limited to
AdaBoost, deep learning algorithm.
Step S404, recognition of face;
The face images obtained in the S402, are searched in face database using face recognition algorithms, are obtained every
A face corresponding face ID in face database.Face recognition algorithms include but is not limited to deep neural network algorithm, template
With algorithm.
Step S405 obtains the optimal face of quality point;
For the face ID obtained in the S404, the corresponding face quality point highest of each face ID is searched in face database
Face ROI image.
Step S406, human face segmentation;
The coordinate and facial size of each facial image according to obtained in from S402, and based on global optimized image
Image will be synthesized to from the highest face ROI image of everyone quality obtained in step S405 point using human face segmentation algorithm
In base image, the effect changed face is realized.Wherein, human face segmentation algorithm includes but is not limited to graph cut algorithm.
In the present embodiment, innovative by human face segmentation algorithm, automatically by everyone in multiple image graphics people
The highest face ROI image of face quality is selected and synthesizes new images.In new images, global image quality not only ensure that most
It is excellent, and also assure everyone face ROI image be it is optimal, this is that tradition is difficult to realize by PS technology.
Referring to Fig. 5, being a kind of system structure diagram for generating best plurality of human faces image provided by the invention, comprising:
Image overall quality analysis module 1, for obtaining multiple the plurality of human faces images being continuously shot, and it is described more multiple
Facial image carries out global quality and analyzes to obtain global optimized image;
Specifically, image overall quality analysis module 1 further comprises: being continuously shot multiple plurality of human faces using photographic device
Image;
Using quality score algorithm according to the fuzziness of image overall, illumination consistency and formation symmetry to every more people
Face image quality score;
Global quality scoring in all plurality of human faces images highest is considered as global optimized image.
Face quality analysis module 2, for carrying out Face datection and face quality analysis multiple described plurality of human faces images,
Identify the facial image of the best face quality corresponding to everyone;
Human face segmentation module 3, for by the face image synthesis of best face quality corresponding to everyone to it is global most preferably
Best plurality of human faces image is obtained in image.
Human face segmentation module 3 further comprises:
Identify face ID in global optimized image, and by the face ID match in face database face quality divide it is highest
Face ROI image;
Every face in global optimized image is substituted for by face ID by face quality point highest using human face segmentation algorithm
Face ROI image obtain best plurality of human faces image.
Referring to Fig. 6, being a kind of system complete structure block diagram for generating best plurality of human faces image provided by the invention, it is described in detail
It is as follows:
Image overall quality analysis module 1, for obtaining multiple the plurality of human faces images being continuously shot, and it is described more multiple
Facial image carries out global quality and analyzes to obtain global optimized image;
Face quality analysis module 2, for carrying out Face datection and face quality analysis multiple described plurality of human faces images,
Identify the facial image of the best face quality corresponding to everyone;
Human face segmentation module 3, for by the face image synthesis of best face quality corresponding to everyone to it is global most preferably
Best plurality of human faces image is obtained in image;
Wherein, face quality analysis module 2 further comprises:
Face datection unit 21 is detected the human face region of the plurality of human faces image using Face datection algorithm, obtains every
The image coordinate and facial size of all faces in plurality of human faces image;
Face mass analysis cell 22 also detects the face matter of the plurality of human faces image using face analysis algorithm
Amount obtains the face quality point of each face in every plurality of human faces image;
Face database 23, for closing the human face region in multiple described plurality of human faces images with face quality point according to face ID
Connection storage;
Face identification unit 24, each face ID in the plurality of human faces image, matches according to the face ID for identification
The highest human face region of face quality point is considered as the facial image of best face quality.
Since the system for generating best plurality of human faces image and the method for generating best plurality of human faces image are one-to-one relationship,
The technical detail that it is related to is identical as effect, will not repeat them here.
Referring to Fig. 7, providing a kind of electronic equipment for the present invention, comprising:
One or more processors 41;
Memory 42;And
One or more programs, wherein one or more of programs are stored in the memory 42 and are configured as
It is executed instruction by one or more of processors 41, is executed instruction described in one or more of processors execution so that described
Electronic equipment executes the method such as the above-mentioned best plurality of human faces image of generation.
The processor 41 is operationally coupled with memory and/or non-volatile memory device.More specifically, processor
41 can be performed the instruction stored in memory and/or non-volatile memory device to execute operation in calculating equipment, such as
It generates image data and/or image data is transferred to electronic console.In this way, processor may include one or more general micro-
Processor, one or more application specific processor (ASIC), one or more Field Programmable Logic Array (FPGA) or they
Any combination.
The application provides a kind of storage medium, and the storage medium includes the program of storage, wherein is run in described program
When control the storage medium where equipment execute the method that best plurality of human faces image is generated described in upper item.
It, can be with if the function is realized in the form of SFU software functional unit and when sold or used as an independent product
It is stored in a computer readable storage medium.Based on this understanding, the technical solution of the application is substantially in other words
The part of the part that contributes to existing technology or the technical solution can be embodied in the form of software products, the meter
Calculation machine software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be a
People's computer, server or network equipment etc.) execute each embodiment the method for the application all or part of the steps.
In embodiment provided by the present application, the computer-readable storage medium of writing may include read-only memory
(ROM, Read-OnlyMemory), random access memory (RAM, RandomAccessMemory), EEPROM, CD-ROM or
Other optical disk storage apparatus, disk storage device or other magnetic storage apparatus, flash memory, USB flash disk, mobile hard disk or it can be used in
Store any other Jie that there is the desired program code of instruction or data structure form and can be accessed by computer
Matter.In addition, any connection can be properly termed as computer-readable medium.For example, if instruction is using coaxial cable, light
The wireless technology of fine optical cable, twisted pair, digital subscriber line (DSL) or such as infrared ray, radio and microwave etc, from net
Stand, server or other remote sources send, then the coaxial cable, optical fiber cable, twisted pair, DSL or such as infrared ray,
The wireless technology of radio and microwave etc includes in the definition of the medium.
In conclusion the present invention is by using Face datection and face recognition algorithms, come be accurately positioned in tag image
All faces, and detect by using face analysis algorithm the face quality point of each face in each image, it is real
Quick, effective detection automatically to face quality in image is showed;Using human face segmentation algorithm by everyone in multiple images
The highest face ROI image of face quality is synthesized to global optimized image and obtains best plurality of human faces image, compared to using PS hand
Duan Eryan can obtain global image optimal quality, be not necessarily to artificial subjective judgement and later period PS, substantially increase shooting group photo
Effect and efficiency.The present invention effectively overcomes various shortcoming in the prior art and has high industrial utilization value.
The above-described embodiments merely illustrate the principles and effects of the present invention, and is not intended to limit the present invention.It is any ripe
The personage for knowing this technology all without departing from the spirit and scope of the present invention, carries out modifications and changes to above-described embodiment.Cause
This, institute is complete without departing from the spirit and technical ideas disclosed in the present invention by those of ordinary skill in the art such as
At all equivalent modifications or change, should be covered by the claims of the present invention.
Claims (10)
1. a kind of method for generating best plurality of human faces image characterized by comprising
Multiple the plurality of human faces images being continuously shot are obtained, and global quality is carried out multiple described plurality of human faces images and analyzes to obtain entirely
Office's optimized image;
Face datection and face quality analysis are carried out multiple described plurality of human faces images, identify the most beautiful woman corresponding to everyone
The facial image of face quality;
The face image synthesis of best face quality corresponding to everyone is obtained into best plurality of human faces figure into global optimized image
Picture.
2. the method according to claim 1 for generating best plurality of human faces image, which is characterized in that the acquisition is continuously shot
Multiple plurality of human faces images, and global qualities are carried out multiple described plurality of human faces images and analyze to obtain the step of global optimized image
Suddenly, comprising:
Multiple plurality of human faces images are continuously shot using photographic device;
Using quality score algorithm according to the fuzziness of image overall, illumination consistency and formation symmetry to every plurality of human faces figure
As quality score;
Global quality scoring in all plurality of human faces images highest is considered as global optimized image.
3. the method according to claim 1 for generating best plurality of human faces image, which is characterized in that described described more multiple
Facial image carries out Face datection and face quality analysis, identifies the facial image of the best face quality corresponding to everyone
The step of, comprising:
The human face region that the plurality of human faces image is detected using Face datection algorithm obtains all faces in every plurality of human faces image
Image coordinate and facial size;
The face quality that the plurality of human faces image is detected using face analysis algorithm is obtained each in every plurality of human faces image
The face quality of face point;
Divide associated storage to face database with face quality the human face region in multiple described plurality of human faces images according to face ID;
It identifies each face ID in the plurality of human faces image, the highest human face region of face quality point is matched according to the face ID
It is considered as the facial image of best face quality.
4. the method according to claim 1 for generating best plurality of human faces image, which is characterized in that by corresponding to everyone most
The step of face image synthesis of beautiful woman's face quality obtains best plurality of human faces image into global optimized image, comprising:
It identifies face ID in global optimized image, and matches the highest face of face quality point in face database by the face ID
ROI image;
Every face in global optimized image is substituted for by face ID by the highest people of face quality point using human face segmentation algorithm
Face ROI image obtains best plurality of human faces image.
5. a kind of system for generating best plurality of human faces image characterized by comprising
Image overall quality analysis module, for obtaining multiple the plurality of human faces images being continuously shot, and multiple plurality of human faces
Image carries out global quality and analyzes to obtain global optimized image;
Face quality analysis module, for carrying out Face datection and face quality analysis, identification multiple described plurality of human faces images
The facial image of best face quality corresponding to everyone out;
Human face segmentation module, for by the face image synthesis of best face quality corresponding to everyone into global optimized image
Obtain best plurality of human faces image.
6. the system according to claim 5 for generating best plurality of human faces image, which is characterized in that described image global quality
Analysis module further comprises:
Multiple plurality of human faces images are continuously shot using photographic device;
Using quality score algorithm according to the fuzziness of image overall, illumination consistency and formation symmetry to every plurality of human faces figure
As quality score;
Global quality scoring in all plurality of human faces images highest is considered as global optimized image.
7. the system according to claim 5 for generating best plurality of human faces image, which is characterized in that the face quality analysis
Module includes:
Face datection unit is detected the human face region of the plurality of human faces image using Face datection algorithm, obtains every plurality of human faces
The image coordinate and facial size of all faces in image;
Face mass analysis cell is also detected the face quality of the plurality of human faces image using face analysis algorithm, obtained
The face quality of each face point in every plurality of human faces image;
Face database is deposited for being associated with the human face region in multiple described plurality of human faces images with face quality point according to face ID
Storage;
Face identification unit, each face ID in the plurality of human faces image, matches face matter according to the face ID for identification
The highest human face region of amount point is considered as the facial image of best face quality.
8. the system according to claim 5 for generating best plurality of human faces image, which is characterized in that the human face segmentation module
Include:
It identifies face ID in global optimized image, and matches the highest face of face quality point in face database by the face ID
ROI image;
Every face in global optimized image is substituted for by face ID by the highest people of face quality point using human face segmentation algorithm
Face ROI image obtains best plurality of human faces image.
9. a kind of electronic equipment, which is characterized in that the electronic equipment includes:
One or more processors;
Memory;And
One or more programs, wherein one or more of programs are stored in the memory and are configured as by described
One or more processors execute instruction, and execute instruction described in one or more of processors execution so that the electronic equipment
Execute the method for generating best plurality of human faces image as Claims 1 to 4 is described in any item.
10. a kind of storage medium, which is characterized in that the storage medium includes the program of storage, wherein described program is being adjusted
With the method for realizing the generation best plurality of human faces image as described in any in claim 1-4 when executing.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910358663.3A CN110097001A (en) | 2019-04-30 | 2019-04-30 | Generate method, system, equipment and the storage medium of best plurality of human faces image |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910358663.3A CN110097001A (en) | 2019-04-30 | 2019-04-30 | Generate method, system, equipment and the storage medium of best plurality of human faces image |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110097001A true CN110097001A (en) | 2019-08-06 |
Family
ID=67446571
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910358663.3A Pending CN110097001A (en) | 2019-04-30 | 2019-04-30 | Generate method, system, equipment and the storage medium of best plurality of human faces image |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110097001A (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111212223A (en) * | 2020-01-10 | 2020-05-29 | 深圳奥比中光科技有限公司 | Method, system and computer readable storage medium for setting imaging parameters |
CN111401206A (en) * | 2020-03-11 | 2020-07-10 | 恒睿(重庆)人工智能技术研究院有限公司 | Panorama sharing method, system, device and medium |
CN111444770A (en) * | 2020-02-26 | 2020-07-24 | 北京大米未来科技有限公司 | Image processing method, image processing apparatus, electronic device, and medium |
CN111507889A (en) * | 2020-04-13 | 2020-08-07 | 北京字节跳动网络技术有限公司 | Image processing method, image processing device, electronic equipment and computer readable storage medium |
CN111814840A (en) * | 2020-06-17 | 2020-10-23 | 恒睿(重庆)人工智能技术研究院有限公司 | Method, system, equipment and medium for evaluating quality of face image |
CN112085688A (en) * | 2020-09-16 | 2020-12-15 | 蒋芳 | Method and system for removing pedestrian shielding during photographing |
CN113033344A (en) * | 2021-03-10 | 2021-06-25 | 咪咕文化科技有限公司 | Image processing method and device and electronic equipment |
CN113630552A (en) * | 2021-07-16 | 2021-11-09 | 深圳全王科技有限公司 | Collective photographing system |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105303161A (en) * | 2015-09-21 | 2016-02-03 | 广东欧珀移动通信有限公司 | Method and device for shooting multiple people |
US20170223265A1 (en) * | 2014-10-10 | 2017-08-03 | Alibaba Group Holding Limited | Methods and devices for establishing photographing template database and providing photographing recommendation information |
CN107734253A (en) * | 2017-10-13 | 2018-02-23 | 广东欧珀移动通信有限公司 | Image processing method, device, mobile terminal and computer-readable recording 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 |
CN109151325A (en) * | 2018-10-26 | 2019-01-04 | 昆山亿趣信息技术研究院有限公司 | A kind of processing method and processing unit synthesizing smiling face |
-
2019
- 2019-04-30 CN CN201910358663.3A patent/CN110097001A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20170223265A1 (en) * | 2014-10-10 | 2017-08-03 | Alibaba Group Holding Limited | Methods and devices for establishing photographing template database and providing photographing recommendation information |
CN105303161A (en) * | 2015-09-21 | 2016-02-03 | 广东欧珀移动通信有限公司 | Method and device for shooting multiple people |
CN107734253A (en) * | 2017-10-13 | 2018-02-23 | 广东欧珀移动通信有限公司 | Image processing method, device, mobile terminal and computer-readable recording 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 |
CN109151325A (en) * | 2018-10-26 | 2019-01-04 | 昆山亿趣信息技术研究院有限公司 | A kind of processing method and processing unit synthesizing smiling face |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111212223A (en) * | 2020-01-10 | 2020-05-29 | 深圳奥比中光科技有限公司 | Method, system and computer readable storage medium for setting imaging parameters |
CN111212223B (en) * | 2020-01-10 | 2021-01-22 | 奥比中光科技集团股份有限公司 | Method, system and computer readable storage medium for setting imaging parameters |
CN111444770A (en) * | 2020-02-26 | 2020-07-24 | 北京大米未来科技有限公司 | Image processing method, image processing apparatus, electronic device, and medium |
CN111401206A (en) * | 2020-03-11 | 2020-07-10 | 恒睿(重庆)人工智能技术研究院有限公司 | Panorama sharing method, system, device and medium |
CN111507889A (en) * | 2020-04-13 | 2020-08-07 | 北京字节跳动网络技术有限公司 | Image processing method, image processing device, electronic equipment and computer readable storage medium |
CN111814840A (en) * | 2020-06-17 | 2020-10-23 | 恒睿(重庆)人工智能技术研究院有限公司 | Method, system, equipment and medium for evaluating quality of face image |
CN112085688A (en) * | 2020-09-16 | 2020-12-15 | 蒋芳 | Method and system for removing pedestrian shielding during photographing |
CN113033344A (en) * | 2021-03-10 | 2021-06-25 | 咪咕文化科技有限公司 | Image processing method and device and electronic equipment |
CN113033344B (en) * | 2021-03-10 | 2024-04-12 | 咪咕文化科技有限公司 | Image processing method and device and electronic equipment |
CN113630552A (en) * | 2021-07-16 | 2021-11-09 | 深圳全王科技有限公司 | Collective photographing system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110097001A (en) | Generate method, system, equipment and the storage medium of best plurality of human faces image | |
US10540757B1 (en) | Method and system for generating combined images utilizing image processing of multiple images | |
CN106951867B (en) | Face identification method, device, system and equipment based on convolutional neural networks | |
RU2617557C1 (en) | Method of exposure to virtual objects of additional reality | |
KR102443026B1 (en) | A method and apparatus for generating a virtual avatar, and a storage medium | |
CN108805047A (en) | A kind of biopsy method, device, electronic equipment and computer-readable medium | |
Yampolskiy et al. | Face recognition in the virtual world: recognizing avatar faces | |
Eckert et al. | Facial cosmetics database and impact analysis on automatic face recognition | |
CN108875540A (en) | Image processing method, device and system and storage medium | |
KR20120005587A (en) | Method and apparatus for generating face animation in computer system | |
Dellana et al. | Data augmentation in CNN-based periocular authentication | |
CN110348358A (en) | A kind of Face Detection system, method, medium and calculate equipment | |
Mal et al. | Facial expression detection using facial expression model | |
CN116091667B (en) | Character artistic image generation system based on AIGC technology | |
CN113392769A (en) | Face image synthesis method and device, electronic equipment and storage medium | |
Heo et al. | Performance evaluation of face recognition using visual and thermal imagery with advanced correlation filters | |
Yuan et al. | Ear detection based on CenterNet | |
CN108875445A (en) | A kind of pedestrian recognition methods and device again | |
JP2021026744A (en) | Information processing device, image recognition method, and learning model generation method | |
Chu et al. | Semi-supervised 3d human pose estimation by jointly considering temporal and multiview information | |
CN110287765A (en) | Baby's appearance prediction technique, device and storage medium based on recognition of face | |
Qi et al. | Fusing visual saliency for material recognition | |
CN113793409A (en) | Virtual prop display method and device | |
Brown | Faster upper body pose recognition and estimation using Compute Unified Device Architecture | |
Komalasari et al. | Shape analysis using generalized Procrustes analysis on active appearance model for facial expression 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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190806 |