Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Face recognition method, system and storage medium based on kernel non-negative matrix factorization

A non-negative matrix decomposition and face recognition technology, applied in the field of face recognition, can solve problems such as slow convergence speed, inability to adjust power parameters, poor parameter controllability, etc., to ensure reliability, overcome changes in posture and illumination, The effect of fast convergence speed

Active Publication Date: 2021-04-06
SHENZHEN UNIV
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] 1. Non-negative matrix factorization algorithm (NMF) is a classic linear method in face recognition, but it often cannot effectively deal with face image data that is nonlinearly distributed due to changes in pose and illumination in face images
[0007] 2. Polynomial kernel non-negative matrix factorization (PNMF) is a nonlinear face recognition method, but it needs to converge under strong conditions, and the convergence speed is slow
In addition, the power exponent parameter of the polynomial kernel function can only be an integer, and when the power exponent parameter is a fraction, it cannot be guaranteed that it is still a kernel function
[0008] 3. Quadratic polynomial kernel non-negative matrix factorization (PKNMF) is also a nonlinear method, but it cannot theoretically prove the convergence of its iterative algorithm. In addition, its power index is fixed (d=2), and the power parameter cannot be Adjustment, that is, its parameters are less adjustable

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
  • Face recognition method, system and storage medium based on kernel non-negative matrix factorization
  • Face recognition method, system and storage medium based on kernel non-negative matrix factorization
  • Face recognition method, system and storage medium based on kernel non-negative matrix factorization

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0041] It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.

[0042] Terminology Explanation

[0043] 1. Non-negative Matrix Factorization (NMF)

[0044] The basic idea of ​​NMF is to take a non-negative sample matrix Approximate decomposition into the product of two non-negative matrices, namely:

[0045] X≈WH,

[0046] in, and are called the base image matrix and the coefficient (feature) matrix, respectively.

[0047] Kernel Function

[0048] Let χ be the input space, k(·,·) be a symmetric function defined on χ×χ, then k is a kernel function if and only if for any finite data set The following Gram matrix K is always positive semi-definite,

[0049]

[0050] 2. Kernel Non-negative Matrix Factorization (KNMF)

[0051] The basic idea of ​​KNMF is to map the non-negative sample matrix X to a high-dimensional space through a nonlinear mapping φ, so that the mapped s...

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 a face recognition method, system and storage medium based on kernel non-negative matrix decomposition. The method includes: constructing a fractional inner product kernel function, and the fractional inner product kernel function has no restriction on the power exponent parameter; Combining the fractional inner product kernel function and kernel non-negative matrix factorization, a fractional inner product kernel nonnegative matrix factorization algorithm is obtained; face recognition is performed through the fractional inner product kernel nonnegative matrix factorization algorithm. The invention overcomes the problem that the power parameter of the polynomial kernel function can only be an integer, makes the selection of the power parameter more flexible; effectively overcomes the changes in posture and illumination in face recognition; and the algorithm has a fast convergence speed and superior recognition performance.

Description

technical field [0001] The invention belongs to the technical field of face recognition, and relates to a face recognition method, system and storage medium based on kernel non-negative matrix decomposition. Background technique [0002] With the rapid development of social informatization and networking, face recognition has become one of the hottest research topics in the field of pattern recognition and image processing, and it is also one of the most successful applications of image analysis and machine vision. Face recognition technology is convenient, reliable and safe. It is a relatively widely accepted biometric method, and it plays a very important role in national security, social economy, home entertainment and other fields. [0003] In the rapid development of face recognition technology, many face recognition algorithms have been proposed one after another, the representative ones are Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA), Local ...

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): G06K9/00
CPCG06V40/16
Inventor 陈文胜刘敬敏王倩
Owner SHENZHEN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products