CN111126169A - Face recognition method and system based on orthogonalization graph regular nonnegative matrix decomposition - Google Patents

Face recognition method and system based on orthogonalization graph regular nonnegative matrix decomposition Download PDF

Info

Publication number
CN111126169A
CN111126169A CN201911221189.6A CN201911221189A CN111126169A CN 111126169 A CN111126169 A CN 111126169A CN 201911221189 A CN201911221189 A CN 201911221189A CN 111126169 A CN111126169 A CN 111126169A
Authority
CN
China
Prior art keywords
matrix
face
face recognition
image
orthogonalized
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
Application number
CN201911221189.6A
Other languages
Chinese (zh)
Other versions
CN111126169B (en
Inventor
王国胤
喻君武
夏书银
彭道万
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Chongqing University of Post and Telecommunications
Original Assignee
Chongqing University of Post and Telecommunications
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Chongqing University of Post and Telecommunications filed Critical Chongqing University of Post and Telecommunications
Priority to CN201911221189.6A priority Critical patent/CN111126169B/en
Publication of CN111126169A publication Critical patent/CN111126169A/en
Application granted granted Critical
Publication of CN111126169B publication Critical patent/CN111126169B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

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
    • G06V40/161—Detection; Localisation; Normalisation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
    • G06F18/24147—Distances to closest patterns, e.g. nearest neighbour classification
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T3/00—Geometric image transformation in the plane of the image
    • G06T3/40—Scaling the whole image or part thereof
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/40—Image enhancement or restoration by the use of histogram techniques
    • G06T5/70—
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/42—Global feature extraction by analysis of the whole pattern, e.g. using frequency domain transformations or autocorrelation
    • 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
    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10004—Still image; Photographic image
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30196—Human being; Person
    • G06T2207/30201—Face

Abstract

The invention discloses a face recognition method and a face recognition system based on orthogonalization graph regular nonnegative matrix decomposition. The method comprises the following steps: 1) preprocessing a face image; 2) extracting the human face characteristics by utilizing an orthogonalized image regular non-negative matrix decomposition; 3) extracting the characteristics of the face image to be tested; 4) and carrying out face recognition detection by using a nearest neighbor classifier. The method makes the result of the non-negative matrix decomposition more robust by utilizing orthogonalization, and improves the accuracy of face recognition.

Description

Face recognition method and system based on orthogonalization graph regular nonnegative matrix decomposition
Technical Field
The invention belongs to the field of face recognition, and particularly relates to a face recognition method based on orthogonalization graph regular nonnegative matrix decomposition.
Background
In the field of computer vision, the face recognition technology has become a popular research direction, and a great amount of innovation is applied to the face recognition technology every year, so that the face recognition accuracy rate is continuously refreshed. The application scene of the face recognition is also very wide, and people can scan faces to get in a station, recognize faces in offices to punch cards, and use the face recognition to carry out criminal investigation and case solving and the like when riding a car. Face recognition is expected to be more promising in the future.
non-Negative Matrix Factorization (NMF) is an algorithm proposed by Lee and Seung in 2000. Under the condition that the matrix elements are not negative, the NMF algorithm decomposes the matrix X into X ═ UV, where U and V are the feature matrix and the weight matrix, respectively. The face features extracted by the NMF can be expressed as a linear combination of feature matrixes, wherein the feature matrixes can express local features of eyes, noses and the like of the face. Deng Cai et al proposed a canonical nonnegative matrix decomposition (GNMF) based on NMF, which considers that if two points in the original eigenspace are close, then in the new eigenspace after matrix decomposition, the two points should remain close, and then GNMF incorporates geometric information into NMF.
Existing methods of non-negative matrix factorization may continue to improve. Experiments on clustering show that the limitation of orthogonalization can optimize the clustering effect. Because the matrix resulting from the matrix decomposition is more sparse after applying the orthogonalizing terms. In order to enhance the sparse representation capability of the matrix, an orthogonalization term is added on the basis of GNMF. The whole OGNMF model can well keep the structure information and sparsity of an original matrix, can decompose a more robust base matrix and a more robust coefficient matrix, and can effectively improve the face recognition rate.
Disclosure of Invention
The present invention is directed to solving the above problems of the prior art. A face recognition method based on orthogonalization graph regular non-negative matrix decomposition is provided. The technical scheme of the invention is as follows:
a face recognition method based on orthogonalization graph regular nonnegative matrix decomposition comprises the following steps:
step S1, carrying out size normalization, filtering denoising and gray normalization pretreatment on the training and testing face image;
step S2, carrying out one-dimensional treatment on the images of the training set, splicing the images into a matrix, then obtaining a new base matrix by utilizing orthogonal image regular nonnegative matrix decomposition, and extracting the human face characteristics;
step S3, on the basis of step S2, projecting the face data set matrix to be tested to a base matrix to obtain a corresponding feature vector;
and step S3, face recognition detection is carried out by utilizing the nearest neighbor classifier.
Further, the step S1 includes the following steps:
step S1.1: size normalization: scaling the facial image to a uniform size;
step S1.2: filtering and denoising: removing noise in the face image by using median filtering;
step S1.3: gray level normalization: and carrying out gray level transformation on the image by utilizing histogram equalization so that the face image follows similar gray level distribution.
Further, the step S2 of extracting the face features by using an orthogonalized regular non-negative matrix decomposition of the graph includes the following steps:
step S2.1: supposing that n training samples are provided, each sample is unidimensionalized to form a training sample matrix of X ═ X1,x2,...,xn]Wherein x isiIs a single face image;
step S2.2: calculating a laplacian matrix L ═ D-W of the matrix for the sample matrix X, wherein D is a degree matrix of the sample matrix and W is an adjacency matrix of the sample matrix;
step S2.3: the maximum iteration number t, the smoothing parameter λ and the orthogonality parameter μ are set, and the weights are updated by minimizing the following objective function:
Figure RE-GDA0002407566680000021
wherein U and V are respectively a feature matrix and a weight matrix obtained by decomposing the sample matrix X, which are non-negative matrices, | · toryFIs a Frobenius paradigm, the parameter lambda is more than or equal to 0 to control the smoothness degree of matrix decomposition, the parameter mu is more than or equal to 0 to control the orthogonality of U and V decomposed by the matrix, and I is a unit matrix;
the matrices U and V are updated according to the following multiplicative iterative formula:
Figure RE-GDA0002407566680000031
Figure RE-GDA0002407566680000032
after t iterations, the U and V updates are complete.
Further, the step S3 of extracting the features of the face image to be tested includes the following steps:
s3.1: preprocessing the face image to be detected in the step S1 to obtain a matrix q;
s3.2: projecting the face image matrix q onto a feature space, wherein the obtained projected feature vector is as follows:
p＝UTq。
further, in step S4, the euclidean distance between p and the face image of the training sample is calculated by using the nearest neighbor algorithm:
di＝||p-Vi||2,i＝1,2,...,n
suppose dkIs diIf the number of the face to be detected is the k-th face in the training sample, the face to be detected belongs to the k-th face in the training sample.
A face recognition system based on orthogonalized graph regular non-negative matrix factorization, comprising:
a preprocessing module: carrying out size normalization, filtering denoising and gray normalization pretreatment on the training and testing face images;
the face feature extraction module: carrying out one-dimensional treatment on the images of the training set, splicing the images into a matrix, then decomposing by utilizing an orthogonalized image regular nonnegative matrix to obtain a new base matrix, and extracting the human face characteristics;
an extraction and identification module: projecting the face data set matrix to be tested to a base matrix to obtain a corresponding feature vector; and carrying out face recognition detection by using the nearest neighbor classifier.
The invention has the following advantages and beneficial effects:
on the basis of graph regular non-negative matrix decomposition, the invention firstly considers the constraint of orthogonality to improve the decomposition result of the non-negative matrix decomposition, and provides an OGNMF model. The OGNMF model can well keep the structural information and sparsity of an original matrix and enhance the sparse expression capacity of the matrix, so that the face recognition rate can be effectively improved.
Drawings
FIG. 1 is a flow chart of a face recognition method based on orthogonalized graph regular non-negative matrix factorization according to a preferred embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be described in detail and clearly with reference to the accompanying drawings. The described embodiments are only some of the embodiments of the present invention.
The technical scheme for solving the technical problems is as follows:
as shown in fig. 1, a face recognition method based on orthogonalization graph regular non-negative matrix decomposition includes the following steps:
1.1 scaling the face image to a uniform size;
1.2 removing noise in the face image by using median filtering;
1.3, carrying out gray level transformation on the image by utilizing histogram equalization so that the face image follows similar gray level distribution;
2.1 assume that there are n training samples, each of which is unidimensionalized to form a training sample matrix of X ═ X1,x2,...,xn]Wherein x isiIs a single face image;
2.2 calculate the laplacian matrix L ═ D-W of the matrix for the sample matrix X, where D is the degree matrix of the sample matrix and W is the adjacency matrix of the sample matrix;
2.3 set the maximum number of iterations t, the smoothing parameter λ and the orthogonality parameter μ.
The matrices U and V are updated according to the following multiplicative iterative formula:
Figure RE-GDA0002407566680000041
Figure RE-GDA0002407566680000042
after t iterations, the U and V updates are complete.
3.1, preprocessing the face image to be detected in the step 1 to obtain a matrix q;
3.2 projecting the face image matrix q to a feature space to obtain projected feature vectors as follows: p is UTq；
And 4, calculating the Euclidean distance between p and the face image of the training sample by using a nearest neighbor algorithm, wherein the minimum distance is the category.
The above examples are to be construed as merely illustrative and not limitative of the remainder of the disclosure. After reading the description of the invention, the skilled person can make various changes or modifications to the invention, and these equivalent changes and modifications also fall into the scope of the invention defined by the claims.

Claims (6)

1. A face recognition method based on orthogonalization graph regular nonnegative matrix decomposition is characterized by comprising the following steps:
step S1, preprocessing the training and testing face image including size normalization, filtering denoising and gray normalization;
s2, on the basis of S1, images of the training set are subjected to one-dimensional processing and spliced into a matrix, then a new base matrix is obtained by utilizing orthogonal image regular non-negative matrix decomposition, and the human face features are extracted;
step S3, on the basis of step S2, projecting the face data set matrix to be tested to a base matrix to obtain a corresponding feature vector;
in step S4, face recognition detection is performed using the nearest neighbor classifier in addition to step S3.
2. The face recognition method based on orthogonalized graph regular non-negative matrix factorization of claim 1, wherein the preprocessing of the step S1 comprises the following steps:
step S1.1: size normalization: scaling the facial image to a uniform size;
step S1.2: filtering and denoising: removing noise in the face image by using median filtering;
step S1.3: gray level normalization: and carrying out gray level transformation on the image by utilizing histogram equalization so that the face image follows similar gray level distribution.
3. The face recognition method based on orthogonalized graph regular non-negative matrix factorization of claim 1, wherein the step S2 of extracting face features by utilizing orthogonalized graph regular non-negative matrix factorization comprises the following steps:
step S2.1: supposing that n training samples are provided, each sample is unidimensionalized to form a training sample matrix of X ═ X1,x2,...,xn]Wherein x isiIs a single face image;
step S2.2: calculating a laplacian matrix L ═ D-W of the matrix for the sample matrix X, wherein D is a degree matrix of the sample matrix and W is an adjacency matrix of the sample matrix;
step S2.3: the maximum iteration number t, the smoothing parameter λ and the orthogonality parameter μ are set, and the weights are updated by minimizing the following objective function:
Figure FDA0002300899740000011
wherein U and V are respectively a feature matrix and a weight matrix obtained by decomposing the sample matrix X, which are non-negative matrices, | · toryFIs a Frobenius paradigm, the parameter lambda is more than or equal to 0 to control the smoothness degree of matrix decomposition, the parameter mu is more than or equal to 0 to control the orthogonality of U and V decomposed by the matrix, and I is a unit matrix;
the matrices U and V are updated according to the following multiplicative iterative formula:
Figure FDA0002300899740000021
Figure FDA0002300899740000022
after t iterations, the U and V updates are complete.
4. The face recognition method based on orthogonalized graph regular non-negative matrix factorization of claim 3, wherein the step S3 of extracting the features of the face image to be tested comprises the following steps:
s3.1: preprocessing the face image to be detected in the step S1 to obtain a matrix q;
s3.2: projecting the face image matrix q onto a feature space, wherein the obtained projected feature vector is as follows:
p＝UTq。
5. the face recognition method based on orthogonalized regular nonnegative matrix factorization of graph according to claim 4, wherein the step S4 is to calculate the Euclidean distance between p and the face image of the training sample by using the nearest neighbor algorithm:
di＝||p-Vi||2,i＝1,2,...,n
suppose dkIs diIf the number of the face to be detected is the k-th face in the training sample, the face to be detected belongs to the k-th face in the training sample.
6. A face recognition system based on orthogonalized graph regular non-Negative Matrix Factorization (NMF), comprising:
a preprocessing module: carrying out size normalization, filtering denoising and gray normalization pretreatment on the training and testing face images;
the face feature extraction module: carrying out one-dimensional treatment on the images of the training set, splicing the images into a matrix, then decomposing by utilizing an orthogonalized image regular nonnegative matrix to obtain a new base matrix, and extracting the human face characteristics;
an extraction and identification module: the face image testing device is used for extracting the features of a face image to be tested; and carrying out face recognition detection by using the nearest neighbor classifier.
CN201911221189.6A 2019-12-03 2019-12-03 Face recognition method and system based on orthogonalization graph regular nonnegative matrix factorization Active CN111126169B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911221189.6A CN111126169B (en) 2019-12-03 2019-12-03 Face recognition method and system based on orthogonalization graph regular nonnegative matrix factorization

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911221189.6A CN111126169B (en) 2019-12-03 2019-12-03 Face recognition method and system based on orthogonalization graph regular nonnegative matrix factorization

Publications (2)

Publication Number Publication Date
CN111126169A true CN111126169A (en) 2020-05-08
CN111126169B CN111126169B (en) 2022-08-30

Family

ID=70497297

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911221189.6A Active CN111126169B (en) 2019-12-03 2019-12-03 Face recognition method and system based on orthogonalization graph regular nonnegative matrix factorization

Country Status (1)

Country Link
CN (1) CN111126169B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111368943A (en) * 2020-05-27 2020-07-03 腾讯科技（深圳）有限公司 Method and device for identifying object in image, storage medium and electronic device
CN113239741A (en) * 2021-04-23 2021-08-10 中国计量大学 Face recognition method based on memory bank non-negative matrix factorization
CN114003752A (en) * 2021-11-24 2022-02-01 重庆邮电大学 Database simplification method and system based on particle ball face clustering image quality evaluation

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102592148A (en) * 2011-12-29 2012-07-18 华南师范大学 Face identification method based on non-negative matrix factorization and a plurality of distance functions
CN105930308A (en) * 2016-04-14 2016-09-07 中国科学院西安光学精密机械研究所 Nonnegative matrix factorization method based on low-rank recovery
US20180060758A1 (en) * 2016-08-30 2018-03-01 Los Alamos National Security, Llc Source identification by non-negative matrix factorization combined with semi-supervised clustering
CN108416374A (en) * 2018-02-13 2018-08-17 中国科学院西安光学精密机械研究所 Based on the non-negative matrix factorization method for differentiating orthogonal subspaces constraint
WO2018149133A1 (en) * 2017-02-17 2018-08-23 深圳大学 Method and system for face recognition by means of dictionary learning based on kernel non-negative matrix factorization, and sparse feature representation
CN109657611A (en) * 2018-12-19 2019-04-19 河南科技大学 A kind of adaptive figure regularization non-negative matrix factorization method for recognition of face
CN110334761A (en) * 2019-07-03 2019-10-15 北京林业大学 There is supervision image-recognizing method based on Orthonormality constraints increment Non-negative Matrix Factorization
CN110378262A (en) * 2019-07-08 2019-10-25 深圳大学 Core Non-negative Matrix Factorization face identification method, device, system and storage medium based on additive Gaussian core
CN110516026A (en) * 2019-07-15 2019-11-29 西安电子科技大学 Online single mode Hash search method based on figure regularization Non-negative Matrix Factorization

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102592148A (en) * 2011-12-29 2012-07-18 华南师范大学 Face identification method based on non-negative matrix factorization and a plurality of distance functions
CN105930308A (en) * 2016-04-14 2016-09-07 中国科学院西安光学精密机械研究所 Nonnegative matrix factorization method based on low-rank recovery
US20180060758A1 (en) * 2016-08-30 2018-03-01 Los Alamos National Security, Llc Source identification by non-negative matrix factorization combined with semi-supervised clustering
WO2018149133A1 (en) * 2017-02-17 2018-08-23 深圳大学 Method and system for face recognition by means of dictionary learning based on kernel non-negative matrix factorization, and sparse feature representation
CN108416374A (en) * 2018-02-13 2018-08-17 中国科学院西安光学精密机械研究所 Based on the non-negative matrix factorization method for differentiating orthogonal subspaces constraint
CN109657611A (en) * 2018-12-19 2019-04-19 河南科技大学 A kind of adaptive figure regularization non-negative matrix factorization method for recognition of face
CN110334761A (en) * 2019-07-03 2019-10-15 北京林业大学 There is supervision image-recognizing method based on Orthonormality constraints increment Non-negative Matrix Factorization
CN110378262A (en) * 2019-07-08 2019-10-25 深圳大学 Core Non-negative Matrix Factorization face identification method, device, system and storage medium based on additive Gaussian core
CN110516026A (en) * 2019-07-15 2019-11-29 西安电子科技大学 Online single mode Hash search method based on figure regularization Non-negative Matrix Factorization

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
JUNWU YU 等: "Edge Sign Prediction Based on Orthogonal Graph Regularized Nonnegative Matrix Factorization for Transfer Learning", 《2019 IEEE INTERNATIONAL CONFERENCE ON BIG KNOWLEDGE (ICBK)》 *
张欣等: "基于判别超图和非负矩阵分解的人脸识别方法", 《运筹学学报》 *
徐泰燕等: "非负矩阵分解及其应用现状分析", 《武汉工业学院学报》 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111368943A (en) * 2020-05-27 2020-07-03 腾讯科技（深圳）有限公司 Method and device for identifying object in image, storage medium and electronic device
CN113239741A (en) * 2021-04-23 2021-08-10 中国计量大学 Face recognition method based on memory bank non-negative matrix factorization
CN114003752A (en) * 2021-11-24 2022-02-01 重庆邮电大学 Database simplification method and system based on particle ball face clustering image quality evaluation

Also Published As

Publication number Publication date
CN111126169B (en) 2022-08-30

Similar Documents

Publication Publication Date Title
CN110232341B (en) Semi-supervised learning image identification method based on convolution-stacking noise reduction coding network
Montazer et al. An improved radial basis function neural network for object image retrieval
Jin et al. Low-rank matrix factorization with multiple hypergraph regularizer
CN111126169B (en) Face recognition method and system based on orthogonalization graph regular nonnegative matrix factorization
CN104077742B (en) Human face sketch synthetic method and system based on Gabor characteristic
Singh et al. Fingerprint image super-resolution via ridge orientation-based clustered coupled sparse dictionaries
CN107392107A (en) A kind of face feature extraction method based on isomery tensor resolution
CN108664911A (en) A kind of robust human face recognition methods indicated based on image sparse
Vishwakarma et al. An efficient hybrid DWT-fuzzy filter in DCT domain based illumination normalization for face recognition
Alain et al. Regularized auto-encoders estimate local statistics
US20240054760A1 (en) Image detection method and apparatus
Dong et al. Feature extraction through contourlet subband clustering for texture classification
Farhan et al. A new model for pattern recognition
CN108710836B (en) Lip detection and reading method based on cascade feature extraction
CN113673465A (en) Image detection method, device, equipment and readable storage medium
Zhao et al. Two-phase incremental kernel PCA for learning massive or online datasets
CN111127407B (en) Fourier transform-based style migration forged image detection device and method
CN111401434A (en) Image classification method based on unsupervised feature learning
CN107563287B (en) Face recognition method and device
CN109190645B (en) High-order high-dimensional image data representation and classification method
Chen et al. Edge detection and texture segmentation based on independent component analysis
CN110443255B (en) Image recognition method for image feature extraction
CN107341485B (en) Face recognition method and device
CN113887509A (en) Rapid multi-modal video face recognition method based on image set
Zou et al. An OCaNet model based on octave convolution and attention mechanism for iris 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
GR01 Patent grant
GR01 Patent grant