A face recognition method based on sparse representation
A face recognition and sparse representation technology, applied in the field of face recognition based on sparse representation, can solve the problems of low recognition efficiency, low recognition accuracy, and inability to adapt to non-rigid visual changes of facial images, and achieves high accuracy and efficiency. High recognition accuracy and computational efficiency, and the effect of improving recognition accuracy and computational efficiency
Active Publication Date: 2021-11-16
NANJING UNIV OF POSTS & TELECOMM
View PDF3 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
For practical applications, Wagner et al. proposed an improved face recognition system. Although it is not constrained by the linear correlation of test samples, it still has disadvantages such as low recognition accuracy, low recognition efficiency, and inability to adapt to non-rigid visual changes in facial images.
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0036] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.
[0037] Such as figure 1 Shown, described a kind of face recognition method based on sparse representation, comprises the following steps:
[0038] Step 1: first input the training sample matrix, concatenate all the training samples in the training sample matrix with all object categories, and then input a test sample;
[0039] Among them, the training sample matrix is A=[X 1,1 ,X 1，2 ,...,X K,N ], the test sample is represented by Y∈Rm;
[0040] Among them, for the training samples, A i =[X i，1 , X i，2 ,...,X i，ni ]∈R m Represents the set of training samples in the i-th layer, where X i，j ∈ R m Represents the vector of all pixels m in the facial image I; the goal of face recognition is to test any test image Y∈R m Identify the class it is in; at the beginning we do not know all the members i of the test image Y, so we define a new...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a face recognition method based on sparse representation, comprising the following steps: Step 1: first input a training sample matrix, concatenate all training samples with all object categories, and then input a test sample; Step 2: Calibrate each training sample in the training sample matrix with the test sample to obtain the calibration training sample matrix; Step 3: Standardize the calibration training sample matrix; Step 4: Calculate each training sample in the standardized calibration training sample matrix The error between the sample and the corresponding standard map to obtain the minimum error; Step 5: Calculate the residual error between the test sample and the corresponding standard map; Step 6: Set the identity for all object categories; Step 7: Output the minimum error The object category corresponding to the training sample corresponding to the value is used as the object category of the test sample. The invention has the advantages that it can better adapt to non-rigid visual changes of facial images, and has high recognition accuracy and calculation efficiency.
Description
technical field [0001] The invention relates to the technical field of image data processing, in particular to a face recognition method based on sparse representation. Background technique [0002] With the rapid development of big data technology, more and more facial images have been uploaded to the Internet. Face recognition, as the most important vision task, recognizes specific identities from unknown objects with facial image features, has been extensively studied in computer vision, such as facial emotion recognition, video surveillance, and biometrics, etc. [0003] Recently, Sparse Representation Model (SRM) was proposed for the task of face recognition. The main idea is to reconstruct the test samples on a full dictionary, the basic element being the training face images themselves. Once a test image can be represented linearly across space by all training samples, sparse reconstruction can be used to identify relevant classes. SRM achieves impressive results a...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/00G06K9/62
CPCG06V40/172G06F18/22G06F18/214
Inventor 周全从德春杨文斌卢竞男王雨
Owner NANJING UNIV OF POSTS & TELECOMM