An Improved Compressed Sensing Face Recognition Method

A face recognition and compressive sensing technology, applied in the field of improved compressive sensing face recognition, can solve the problems of complex calculation, slow calculation speed and huge calculation amount of the minimum norm solution, so as to achieve the preservation of local texture characteristics and increase the calculation speed. , the effect of good statistical similarity

Inactive Publication Date: 2019-07-23
SHANDONG NORMAL UNIV
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

but The calculation of the norm minimum solution is complex and the amount of calculation is huge, and as the number of dictionary atoms increases, the calculation speed drops sharply, resulting in slow recognition speed, which is not suitable for real-time applications

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • An Improved Compressed Sensing Face Recognition Method
  • An Improved Compressed Sensing Face Recognition Method
  • An Improved Compressed Sensing Face Recognition Method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0037] The present invention will be described in detail below in conjunction with the drawings:

[0038] 1. Local Binary Pattern (LBP) coding principle

[0039] Usually a pair of values ​​(P, R) is used to describe (x c ,y c ) Is the center and R is the radius of the local circular neighborhood, such as figure 1 Shown. Use g c Indicates the gray value of the center pixel, g p (p=0,...,P-1) represents the gray value of P adjacent pixels distributed at equal intervals on the circumference. Assuming that the coordinates of the central pixel are (0,0), the coordinates of the neighboring pixels (x p ,y p ) Is (-R sin(2πp / P), Rcos(2πp / P)). If (x p ,y p ) Does not fall on the pixel point, use bilinear interpolation to calculate the gray value g of the point p . Different combinations of P and R values ​​describe different LBP descriptors.

[0040] The LBP algorithm uses structured thinking to analyze the texture features of the image in a fixed window. The gray value of the center pix...

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

No PUM Login to view more

Abstract

The invention discloses an improved compressed sensing face recognition method, which includes step 1 image preprocessing; adopting a cumulative distribution function to perform histogram equalization on a gray-scale face image, and then adopting an LBP rotation-invariant uniform mode to extract a gray-scale face The face image feature vector constructs the LBP feature space; Step 2 finds the neighbor samples of the test image in the low-dimensional LBP feature space; Step 3 uses the neighbor samples to adaptively construct a redundant dictionary to complete the perceptual recognition of the test image. This method looks for the neighbor samples of the test sample in the low-dimensional LBP feature space, and the complete redundant dictionary is formed by the neighbor samples, the number of dictionary atoms is greatly reduced, and the dictionary atoms and the test samples have higher structural similarity, so the algorithm improves Recognition speed, and improve the correct recognition rate.

Description

Technical field [0001] The invention discloses an improved compressed sensing face recognition method. Background technique [0002] Human facial features have become a new identity authentication medium because of their uniqueness, stability, and non-theft. Compared with other biometric recognition, the method of face recognition is more in line with the habit of humans to identify their own identity, the recognition is more natural and intuitive, and the facial image collection equipment is simple. It can be completed by ordinary household cameras in natural light, and it is not required for collection. The user's cooperation is even concealed, thus reducing the probability of pretending to deceive. Human faces will not easily leave marks on the medium, and the security is higher. Therefore, the application prospects of face recognition technology in military security, public security, civil and economic fields are very broad. Because human faces are easily affected by facto...

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
no application Login to view more
Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/00G06K9/46G06K9/62
CPCG06V40/161G06V40/172G06V10/507G06F18/24147
Inventor 魏冬梅周茂霞马娜
Owner SHANDONG NORMAL UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap