CN105868716A - Method for human face recognition based on face geometrical features - Google Patents
Method for human face recognition based on face geometrical features Download PDFInfo
- Publication number
- CN105868716A CN105868716A CN201610187618.2A CN201610187618A CN105868716A CN 105868716 A CN105868716 A CN 105868716A CN 201610187618 A CN201610187618 A CN 201610187618A CN 105868716 A CN105868716 A CN 105868716A
- Authority
- CN
- China
- Prior art keywords
- face
- feature
- geometric
- facial
- human face
- 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
-
- 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
- G06V40/168—Feature extraction; Face representation
- G06V40/171—Local features and components; Facial parts ; Occluding parts, e.g. glasses; Geometrical relationships
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Collating Specific Patterns (AREA)
Abstract
The invention provides a method for human face recognition based on face geometrical features. The method includes human face feature points detection, geometrical feature construction, human face database classification and human face recognition. According to the invention, the method, based on human face detection and human face recognition technology, improves geometrical features used in the human face feature point detection method and human face recognition, and guarantees in-time and accuracy of a human face recognition system based on feature point detection which combines an active shape model and the elastic bunch graph matching algorithm and also simple and accurate face geometrical feature technology. According to the invention, the method decomposes data in a large database into a plurality of sub-databases with labels which are determined by selected classification geometrical features. As for the input image to be tested, matching data can be reduced by simply comparing classification geometrical features of the image to be tested with each sub-database label in order to greatly increase efficiency of human face recognition.
Description
Technical field
The present invention relates to a kind of face recognition technology, particularly relate to a kind of face identification method based on facial geometric feature.
Background technology
Face recognition technology is by the human face target object in detection image, utilizes the facial image characteristic information in destination object
Carry out the mode identification technology of identity authentication, recently as government bodies, enterprise and to the concern of personal information security and attention,
The application of face recognition technology has the most gradually been come in the life of ordinary people, and increasing scientific and technical innovation enterprise has also put in a large number
Resource participates in the research and development main forces of recognition of face, and the recognition of face that such as Alibaba releases completes payment verification.Compare biography
The mode identification technology of system, recognition of face has the advantages such as not contact, disguise, significantly improves the efficiency of authentication,
But meanwhile image imaging process inevitably adds illumination, the impact of the factors such as attitude, cause the accuracy rate identified
Decline, so the precision of recognition of face still has much room for improvement.
On the basis of ASM (active shape model) is built upon PDM (points distribution models), by training image sample acquisition
The statistical information of the characteristic point distribution of training image sample, and obtain the change direction that characteristic point allows to exist, it is achieved in target
The position of characteristic of correspondence point is found on image.Training sample needs the position of the manual all of characteristic point of labelling, recording feature
The coordinate of point, and the characteristic vector that the local gray level model calculating each Feature point correspondence adjusts as local feature region.
The model trained is being placed on target image, the when of finding the next position of each characteristic point, is using local gray level
The model searching characteristic point that local gray level model mahalanobis distance is minimum on current signature point assigned direction will as current signature point
The position moved to, referred to as suggested point, finds all of suggested points to be obtained with a search
Suggested shape, then makes current similar being adjusted to of model most probable by current model by adjusting parameter
Suggest shape, iteration is until realizing convergence.
Elastic graph bundle matching algorithm uses labeled graph to represent facial image, and the node of labeled graph describes face local feature with one group
Two-Dimensional Gabor Wavelets conversion coefficient represent；The limit of labeled graph describes adjacent two nodes to the metric of correspondence position
Represent.Realized by the coupling between the labeled graph of different facial images contacting between the local feature of face corresponding position,
It is thus possible to facial image solution is compared and Classification and Identification, and then each node location in figure is most preferably mated.
For affecting the external condition of discrimination in natural environment, existing face recognition algorithms is concentrated mainly on side based on template
Method and Statistics-Based Method.Algorithm based on template is to care, and rotationally-varying comparison is sensitive, between template and detection object
Ratio, when rotating consistent with illumination, could obtain preferable effect, and in larger data base, efficiency also can be obvious simultaneously
Reduce.And based on the recognizer added up when the angle of destination object changes, accuracy of identification also can drastically decline.Simultaneously
Both of which is to carry out feature extraction with face entirety, and data volume is big, calculates speed slow, causes the real-time effect of system not
Enough ideals.
In view of the above, it is provided that a kind of can solve the problem that the authentication of user in extensive face database, it is ensured that recognition of face
The real-time of system and the face identification method of accuracy requirement are necessary.
Summary of the invention
The shortcoming of prior art in view of the above, it is an object of the invention to provide a kind of face based on facial geometric feature and knows
Other method, for solving masterplate matching process and statistical learning method in prior art, all to there is accuracy of identification and real-time not enough
Problem.
For achieving the above object and other relevant purposes, the present invention provides a kind of face identification method based on facial geometric feature,
Including step: step 1), the method utilizing active shape model to be combined with elastic graph bundle matching algorithm is instructed in face database
Practice Face datection model；Step 2), choose specific characteristic point, build face geometry for every face in facial image database
Characteristic set；Step 3), Face geometric eigenvector set is chosen geometric properties of specifically classifying, facial image database is decomposed
For some sub-facial image databases；Step 4), utilize step 1) in Face datection model, mate the pass in facial image to be measured
Key characteristic point, builds the Face geometric eigenvector in testing image, simultaneously according to classification geometric properties, looks in sub-facial image database
The sub-facial image database of arest neighbors to testing image place；Step 5), the weights of computational geometry characteristic component；Step 6),
By similarity function, calculate in testing image the face characteristic of Face geometric eigenvector vector and the facial image word bank at its place to
The similarity degree of amount, if similarity degree is higher than threshold value, is then judged as same person, if similarity degree is less than threshold value, then judges
For different people.
As a kind of preferred version of the face identification method based on facial geometric feature of the present invention, step 1) in, actively shape
The method that shape model is combined with elastic graph bundle matching algorithm includes: the feature of described active shape model uses one group to describe face office
The Two-Dimensional Gabor Wavelets conversion coefficient of portion's feature represents, i.e. pixel at key feature points and different directions and the Gabor of frequency
Coefficient sets after core convolution, it describes the fritter gray value that image gives around pixel；By described active shape model with
The method coupling testing image that elastic graph bundle matching algorithm combines obtains face face organ's profile key feature points.
As a kind of preferred version of the face identification method based on facial geometric feature of the present invention, step 2) in, described spy
Fixed characteristic point includes: eye eminence and two boundary points of mouth eminence face profile, each two key points of left and right eyebrow, left and right eyes
Two canthus points and peak, minimum point, each wing of nose two boundary point, nose minimum point, three, upper lip coboundary key
Point three key points of lower boundary, one key point of lower lip lower boundary, bottom two key points of the corners of the mouth and cheek at 1/3rd
Two key points.
As a kind of preferred version of the face identification method based on facial geometric feature of the present invention, step 2) in, face is several
The structure of what feature includes: distances based on two eyebrow midpoints to chin set up the reference range of vertical direction, based on the left and right tail of the eye
The distance of horizontal eminence sets up the reference range of horizontal direction, utilizes reference range, calculates ratio characteristic component, constructs with this several
What characteristic vector, and the normalized of the geometric properties vector to structure.
As a kind of preferred version of the face identification method based on facial geometric feature of the present invention, step 3) in, according to not
Same face, changes big classification geometric properties under varying environment, the data in facial image database be decomposed into some with label
Sub-facial image database, described label is determined by the classification geometric properties chosen, and for the testing image of input, only need to be classified
Geometric properties compares with the label of each sub-facial image database.
As a kind of preferred version of the face identification method based on facial geometric feature of the present invention, step 5) in, use 20
The image composition training set of everyone 3 different attitudes of individual carries out Weight Training, first each feature provides 3-5 reasonably
Weights, then calculate similarity to the image of everyone 3 different attitudes, take so that same person difference pose presentation is similar
Spend the final weights that the highest value is geometric properties component.
As it has been described above, the face identification method based on facial geometric feature of the present invention, have the advantages that base of the present invention
In Face datection and face recognition technology, the geometric properties used in facial feature points detection method and recognition of face is improved,
The feature point detection combined based on active shape model and elastic graph bundle matching algorithm, and simple accurate facial geometric feature
, it is ensured that the real-time of face identification system and accuracy rate.Large-scale face database is classified by the present invention, according to not
Same face, changes big classification geometric properties under varying environment, the data in large database be decomposed into some with label
Subdata base, this label is determined by the classification geometric properties chosen.For the testing image of input, geometry of only need to being classified is special
Levying and compare with each subdata base label, it is possible to reduce data scale to be mated, the efficiency of the recognition of face made is greatly improved.
Accompanying drawing explanation
Fig. 1 is shown as the schematic flow sheet of the face identification method based on facial geometric feature of the present invention.
Fig. 2 is shown as in the face identification method based on facial geometric feature of the present invention, face critical point detection and geometric properties
The schematic flow sheet that point builds.
Element numbers explanation
S11～S19 step
S121～S128 step
Detailed description of the invention
Below by way of specific instantiation, embodiments of the present invention being described, those skilled in the art can be by disclosed by this specification
Content understand other advantages and effect of the present invention easily.The present invention can also be added by the most different detailed description of the invention
To implement or application, the every details in this specification can also be based on different viewpoints and application, in the essence without departing from the present invention
Various modification or change is carried out under god.
Refer to Fig. 1～Fig. 2.It should be noted that the diagram provided in the present embodiment illustrates the present invention's the most in a schematic way
Basic conception, component count when only display with relevant assembly in the present invention rather than is implemented according to reality in diagram then, shape and
Size is drawn, and during its actual enforcement, the kenel of each assembly, quantity and ratio can be a kind of random change, and its assembly layout type
State is likely to increasingly complex.
As shown in Figures 1 and 2, the present embodiment provides a kind of face identification method based on facial geometric feature, including step:
First step 1 is carried out), utilize the method that active shape model is combined with elastic graph bundle matching algorithm in face database
Training face detection model.
As example, step 1) in, the method that active shape model is combined with elastic graph bundle matching algorithm includes: described active
The feature of shape uses one group of Two-Dimensional Gabor Wavelets conversion coefficient describing face local feature to represent, i.e. key feature
Coefficient sets after the Gabor core convolution of the pixel at Dian and different directions and frequency, it describes image and gives around pixel
Fritter gray value；The method coupling testing image being combined with elastic graph bundle matching algorithm by described active shape model obtains face
Face organ's profile key feature points.
As it is shown in figure 1, then carry out step 2), choose specific characteristic point, build for every face in facial image database
Face geometric eigenvector set；
As example, step 2) including:
Step S11, first carries out facial contour feature point extraction, needs to choose the characteristic point detected, removes
The characteristic information the least to identifying contribution.Choosing of characteristic point requires to include: ensure enough information with the fewest characteristic point
Reflect most important feature in recognition of face；The geometric properties that selected characteristic point is constituted should be the simplest；To illumination
Dependency is little；Less sensitive to the expression shape change of face；Bigger in different people amplitude of variation on the face.According to these requirements from extraction
Characteristic point in choose 28 characteristic points being applicable to recognition of face, described specific characteristic point includes: eye eminence and mouth eminence face
Two boundary points of profile, each two key points of left and right eyebrow, two canthus points of left and right eyes and peak, minimum point, respectively
The individual wing of nose two boundary point, nose minimum point, three, upper lip coboundary, three key points of key point lower boundary, lower lip lower boundary
One key point, two key points at 1/3rd bottom two key points of the corners of the mouth and cheek.
Step S12, structure Face geometric eigenvector vector: distances based on two eyebrow midpoints to chin set up the stand-off of vertical direction
From, distance based on the horizontal eminence of the left and right tail of the eye sets up the reference range of horizontal direction, utilizes reference range, calculates ratio special
Levy component, with this constructive geometry characteristic vector, and the normalized of the geometric properties vector to structure.
As in figure 2 it is shown, more specifically, face critical point detection includes with geometric properties point construction step:
Step S121, uncalibrated image storehouse septum reset profile point；
Step S122, the method utilizing active shape model to be combined with elastic graph bundle matching algorithm trains face in face database
Detection model；
Step S123, inputs image to be detected；
Step S124, carries out the detection of face face contour point based on Face datection model；
Step S125, filters out the key point being suitable for identification from face face contour point；
Step S126, selected characteristic point builds reference range, and distances based on two eyebrow midpoints to chin set up the benchmark of vertical direction
Distance, distance based on the horizontal eminence of the left and right tail of the eye sets up the reference range of horizontal direction；
Step S127, utilizes reference range, calculates ratio characteristic component, constructs ratio geometric properties vector with this, and to structure
Ratio geometric properties vector normalized；
Step S128, calculates and determines the weights of each component in characteristic vector.
As it is shown in figure 1, then carry out step S13, Face geometric eigenvector set is chosen geometric properties of specifically classifying, will
Facial image database is decomposed into some sub-facial image databases.
Specifically, in this step, according to different faces, big classification geometric properties is changed under varying environment, by facial image
Data in storehouse are decomposed into some sub-facial image databases with label, and described label is determined by the classification geometric properties chosen, right
In the testing image of input, geometric properties of only need to being classified compares with the label of each sub-facial image database.This step can solve
In extensive face database, feature there will be discrimination and the problem of matching speed decline in the matching process.
As in figure 2 it is shown, then carry out step S14～S18, utilize step 1) in Face datection model, input face figure to be measured
As S14, mate the key feature points S15 in facial image to be measured, build the Face geometric eigenvector S16 in testing image, simultaneously
According to classification geometric properties, sub-facial image database finds the arest neighbors facial image database S17 at testing image place, and
This sub-facial image database finds the classification S18 of the arest neighbors of testing image.
Then step 5 is carried out), the weights of computational geometry characteristic component；
Specifically, in this step, the image composition training set of 20 everyone 3 different attitudes of people is used to carry out Weight Training,
First each feature is given 3-5 reasonably weights, then the image of everyone 3 different attitudes is calculated similarity,
Take the final weights that value be geometric properties component the highest so that same person difference pose presentation similarity.
As it is shown in figure 1, finally carry out step S19, by similarity function, calculate Face geometric eigenvector vector in testing image
With the similarity degree of the face feature vector of the facial image word bank at its place, if similarity degree is higher than threshold value, then it is judged as same
Individual, if similarity degree is less than threshold value, is then judged as different people.
As it has been described above, the face identification method based on facial geometric feature of the present invention, have the advantages that base of the present invention
In Face datection and face recognition technology, the geometric properties used in facial feature points detection method and recognition of face is improved,
The feature point detection combined based on active shape model and elastic graph bundle matching algorithm, and simple accurate facial geometric feature
, it is ensured that the real-time of face identification system and accuracy rate.Large-scale face database is classified by the present invention, according to not
Same face, changes big classification geometric properties under varying environment, the data in large database be decomposed into some with label
Subdata base, this label is determined by the classification geometric properties chosen.For the testing image of input, geometry of only need to being classified is special
Levying and compare with each subdata base label, it is possible to reduce data scale to be mated, the efficiency of the recognition of face made is greatly improved.
So, the present invention effectively overcomes various shortcoming of the prior art and has high industrial utilization.
The principle of above-described embodiment only illustrative present invention and effect thereof, not for limiting the present invention.Any it is familiar with this skill
Above-described embodiment all can be modified under the spirit and the scope of the present invention or change by the personage of art.Therefore, such as
All that in art, tool usually intellectual is completed under without departing from disclosed spirit and technological thought etc.
Effect is modified or changes, and must be contained by the claim of the present invention.
Claims (6)
1. a face identification method based on facial geometric feature, it is characterised in that include step:
Step 1), utilize method training of human in face database that active shape model is combined with elastic graph bundle matching algorithm
Face detection model；
Step 2), choose specific characteristic point, build Face geometric eigenvector set for every face in facial image database；
Step 3), Face geometric eigenvector set is chosen geometric properties of specifically classifying, if being decomposed into by facial image database
Dry sub-facial image database；
Step 4), utilize step 1) in Face datection model, mate the key feature points in facial image to be measured, structure
Build the Face geometric eigenvector in testing image, simultaneously according to classification geometric properties, in sub-facial image database, find testing image
The sub-facial image database of arest neighbors at place；
Step 5), the weights of computational geometry characteristic component；
Step 6), by similarity function, calculate Face geometric eigenvector vector and the facial image at its place in testing image
The similarity degree of the face feature vector of word bank, if similarity degree is higher than threshold value, is then judged as same person, if similarity journey
Degree less than threshold value, is then judged as different people.
Face identification method based on facial geometric feature the most according to claim 1, it is characterised in that: step 1) in, main
The method that dynamic shape is combined with elastic graph bundle matching algorithm includes: the feature of described active shape model uses one group of description
The Two-Dimensional Gabor Wavelets conversion coefficient of face local feature represents, i.e. pixel at key feature points and different directions and frequency
Coefficient sets after the Gabor core convolution of rate, it describes the fritter gray value that image gives around pixel；By described master
It is special that the method coupling testing image that dynamic shape is combined with elastic graph bundle matching algorithm obtains face face organ's profile key
Levy a little.
Face identification method based on facial geometric feature the most according to claim 1, it is characterised in that: step 2) in, institute
State specific characteristic point to include: eye eminence and two boundary points of mouth eminence face profile, each two key points of left and right eyebrow, left
Two canthus points of right eye eyeball and peak, minimum point, each wing of nose two boundary point, nose minimum point, upper lip coboundary
Three key points of three key point lower boundaries, one key point of lower lip lower boundary, bottom two key points of the corners of the mouth and cheek
Two key points at 1/3rd.
Face identification method based on facial geometric feature the most according to claim 1, it is characterised in that: step 2) in, people
The structure of face geometric properties includes: distances based on two eyebrow midpoints to chin set up the reference range of vertical direction, based on left and right
The distance of the horizontal eminence of the tail of the eye sets up the reference range of horizontal direction, utilizes reference range, calculates ratio characteristic component, with
This constructive geometry characteristic vector, and the normalized of the geometric properties vector to structure.
Face identification method based on facial geometric feature the most according to claim 1, it is characterised in that: step 3) in, root
According to different faces, under varying environment, change big classification geometric properties, the data in facial image database are decomposed into some with
The sub-facial image database of label, described label is determined by the classification geometric properties chosen, and for the testing image of input, only needs
Geometric properties of being classified compares with the label of each sub-facial image database.
Face identification method based on facial geometric feature the most according to claim 1, it is characterised in that: step 5) in, adopt
Form training set with the image of 20 everyone 3 different attitudes of people and carry out Weight Training, first each feature is provided 3-5
Individual rational weights, then calculate similarity to the image of everyone 3 different attitudes, take so that same person difference appearance
The value that state image similarity is the highest is the final weights of geometric properties component.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610187618.2A CN105868716B (en) | 2016-03-29 | 2016-03-29 | A kind of face identification method based on facial geometric feature |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610187618.2A CN105868716B (en) | 2016-03-29 | 2016-03-29 | A kind of face identification method based on facial geometric feature |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105868716A true CN105868716A (en) | 2016-08-17 |
| CN105868716B CN105868716B (en) | 2019-08-13 |
Family
ID=56625331
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610187618.2A Active CN105868716B (en) | 2016-03-29 | 2016-03-29 | A kind of face identification method based on facial geometric feature |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105868716B (en) |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106503686A (en) * | 2016-10-28 | 2017-03-15 | 广州炒米信息科技有限公司 | The method and system of retrieval facial image |
| CN107146275A (en) * | 2017-03-31 | 2017-09-08 | 北京奇艺世纪科技有限公司 | A kind of method and device of setting virtual image |
| CN107358163A (en) * | 2017-06-14 | 2017-11-17 | 广东数相智能科技有限公司 | Visitor's line trace statistical method, electronic equipment and storage medium based on recognition of face |
| CN107563336A (en) * | 2017-09-07 | 2018-01-09 | 廖海斌 | Human face similarity degree analysis method, the device and system of game are matched for famous person |
| CN107767358A (en) * | 2016-08-23 | 2018-03-06 | 阿里巴巴集团控股有限公司 | A kind of objects in images fuzziness determines method and apparatus |
| CN108108760A (en) * | 2017-12-19 | 2018-06-01 | 山东大学 | A kind of fast human face recognition |
| CN109063628A (en) * | 2018-07-27 | 2018-12-21 | 平安科技（深圳）有限公司 | Face identification method, device, computer equipment and storage medium |
| CN109726625A (en) * | 2018-08-13 | 2019-05-07 | 平安科技（深圳）有限公司 | Anti- face identification method, device, terminal device and storage medium |
| CN109801462A (en) * | 2019-01-21 | 2019-05-24 | 江西阳光安全设备集团有限公司 | The burglar alarm of intelligent mobile rack |
| CN109993042A (en) * | 2017-12-29 | 2019-07-09 | 国民技术股份有限公司 | A kind of face identification method and its device |
| CN110097586A (en) * | 2019-04-30 | 2019-08-06 | 青岛海信网络科技股份有限公司 | A kind of Face datection method for tracing and device |
| CN112308084A (en) * | 2020-11-09 | 2021-02-02 | 西安工程大学 | Clothing contour classification method based on multi-scale Gabor |
| CN113537173A (en) * | 2021-09-16 | 2021-10-22 | 中国人民解放军国防科技大学 | Face image authenticity identification method based on face patch mapping |
| CN113657851A (en) * | 2021-07-30 | 2021-11-16 | 绿漫科技有限公司 | Visitor access management method |
| CN114612983A (en) * | 2022-03-15 | 2022-06-10 | 北京拙河科技有限公司 | Face recognition method and system based on elastic graph and SVM |
| CN116895093A (en) * | 2023-09-08 | 2023-10-17 | 苏州浪潮智能科技有限公司 | Face recognition method, device, equipment and computer readable storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1529278A (en) * | 2003-10-09 | 2004-09-15 | 重庆大学 | Multi-category-based human face classifying and identifying method |
| CN101308571A (en) * | 2007-05-15 | 2008-11-19 | 上海中科计算技术研究所 | Method for generating novel human face by combining active grid and human face recognition |
| US20100232657A1 (en) * | 2009-03-12 | 2010-09-16 | Jie Wang | Automatic Face Recognition |
| CN104239856A (en) * | 2014-09-04 | 2014-12-24 | 电子科技大学 | Face recognition method based on Gabor characteristics and self-adaptive linear regression |
-
2016
- 2016-03-29 CN CN201610187618.2A patent/CN105868716B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1529278A (en) * | 2003-10-09 | 2004-09-15 | 重庆大学 | Multi-category-based human face classifying and identifying method |
| CN101308571A (en) * | 2007-05-15 | 2008-11-19 | 上海中科计算技术研究所 | Method for generating novel human face by combining active grid and human face recognition |
| US20100232657A1 (en) * | 2009-03-12 | 2010-09-16 | Jie Wang | Automatic Face Recognition |
| CN104239856A (en) * | 2014-09-04 | 2014-12-24 | 电子科技大学 | Face recognition method based on Gabor characteristics and self-adaptive linear regression |
Non-Patent Citations (1)
| Title |
|---|
| 李梅: "《物联网科技导论》", 31 August 2015, 北京邮电大学出版社 * |
Cited By (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107767358A (en) * | 2016-08-23 | 2018-03-06 | 阿里巴巴集团控股有限公司 | A kind of objects in images fuzziness determines method and apparatus |
| CN107767358B (en) * | 2016-08-23 | 2021-08-13 | 斑马智行网络(香港)有限公司 | Method and device for determining ambiguity of object in image |
| CN106503686A (en) * | 2016-10-28 | 2017-03-15 | 广州炒米信息科技有限公司 | The method and system of retrieval facial image |
| CN107146275A (en) * | 2017-03-31 | 2017-09-08 | 北京奇艺世纪科技有限公司 | A kind of method and device of setting virtual image |
| CN107146275B (en) * | 2017-03-31 | 2020-10-27 | 北京奇艺世纪科技有限公司 | Method and device for setting virtual image |
| CN107358163A (en) * | 2017-06-14 | 2017-11-17 | 广东数相智能科技有限公司 | Visitor's line trace statistical method, electronic equipment and storage medium based on recognition of face |
| CN107563336A (en) * | 2017-09-07 | 2018-01-09 | 廖海斌 | Human face similarity degree analysis method, the device and system of game are matched for famous person |
| CN108108760A (en) * | 2017-12-19 | 2018-06-01 | 山东大学 | A kind of fast human face recognition |
| CN109993042A (en) * | 2017-12-29 | 2019-07-09 | 国民技术股份有限公司 | A kind of face identification method and its device |
| CN109063628B (en) * | 2018-07-27 | 2023-04-21 | 平安科技（深圳）有限公司 | Face recognition method, device, computer equipment and storage medium |
| CN109063628A (en) * | 2018-07-27 | 2018-12-21 | 平安科技（深圳）有限公司 | Face identification method, device, computer equipment and storage medium |
| CN109726625A (en) * | 2018-08-13 | 2019-05-07 | 平安科技（深圳）有限公司 | Anti- face identification method, device, terminal device and storage medium |
| CN109726625B (en) * | 2018-08-13 | 2023-06-23 | 平安科技（深圳）有限公司 | Face recognition method, device, terminal equipment and storage medium |
| CN109801462A (en) * | 2019-01-21 | 2019-05-24 | 江西阳光安全设备集团有限公司 | The burglar alarm of intelligent mobile rack |
| CN110097586A (en) * | 2019-04-30 | 2019-08-06 | 青岛海信网络科技股份有限公司 | A kind of Face datection method for tracing and device |
| CN112308084A (en) * | 2020-11-09 | 2021-02-02 | 西安工程大学 | Clothing contour classification method based on multi-scale Gabor |
| CN112308084B (en) * | 2020-11-09 | 2024-04-23 | 西安工程大学 | Clothing profile classification method based on multi-scale Gabor |
| CN113657851A (en) * | 2021-07-30 | 2021-11-16 | 绿漫科技有限公司 | Visitor access management method |
| CN113537173B (en) * | 2021-09-16 | 2022-03-18 | 中国人民解放军国防科技大学 | Face image authenticity identification method based on face patch mapping |
| CN113537173A (en) * | 2021-09-16 | 2021-10-22 | 中国人民解放军国防科技大学 | Face image authenticity identification method based on face patch mapping |
| CN114612983A (en) * | 2022-03-15 | 2022-06-10 | 北京拙河科技有限公司 | Face recognition method and system based on elastic graph and SVM |
| CN116895093A (en) * | 2023-09-08 | 2023-10-17 | 苏州浪潮智能科技有限公司 | Face recognition method, device, equipment and computer readable storage medium |
| CN116895093B (en) * | 2023-09-08 | 2024-01-23 | 苏州浪潮智能科技有限公司 | Face recognition method, device, equipment and computer readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105868716B (en) | 2019-08-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105868716A (en) | Method for human face recognition based on face geometrical features | |
| CN106682598B (en) | Multi-pose face feature point detection method based on cascade regression | |
| CN107194341B (en) | Face recognition method and system based on fusion of Maxout multi-convolution neural network | |
| US10049262B2 (en) | Method and system for extracting characteristic of three-dimensional face image | |
| Papazov et al. | Real-time 3D head pose and facial landmark estimation from depth images using triangular surface patch features | |
| JP6754619B2 (en) | Face recognition method and device | |
| Gilani et al. | Deep, dense and accurate 3D face correspondence for generating population specific deformable models | |
| Kalogerakis et al. | Learning 3D mesh segmentation and labeling | |
| CN107169455B (en) | Face attribute recognition method based on depth local features | |
| CN102682302B (en) | Human body posture identification method based on multi-characteristic fusion of key frame | |
| CN100472556C (en) | Testing apparatus and method for special object | |
| Qian et al. | People identification using floor pressure sensing and analysis | |
| CN106897675A (en) | The human face in-vivo detection method that binocular vision depth characteristic is combined with appearance features | |
| Li et al. | Efficient 3D face recognition handling facial expression and hair occlusion | |
| CN108182397B (en) | Multi-pose multi-scale human face verification method | |
| CN104517097A (en) | Kinect-based moving human body posture recognition method | |
| CN110084211B (en) | Action recognition method | |
| CN109460734A (en) | The video behavior recognition methods and system shown based on level dynamic depth projection difference image table | |
| CN104899578B (en) | A kind of method and device of recognition of face | |
| CN106096517A (en) | A kind of face identification method based on low-rank matrix Yu eigenface | |
| KR101558547B1 (en) | Age Cognition Method that is powerful to change of Face Pose and System thereof | |
| JP6417664B2 (en) | Person attribute estimation device, person attribute estimation method and program | |
| CN113591763B (en) | Classification recognition method and device for face shapes, storage medium and computer equipment | |
| Beksi et al. | Object classification using dictionary learning and rgb-d covariance descriptors | |
| CN105917356A (en) | Contour-based classification of objects |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |