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

Human face identification method and device based on matching pursuit algorithm

A face recognition and matching tracking technology, applied in the field of image recognition, can solve the problems of large number of iterations and calculation, easy to be affected by light, and low recognition efficiency.

Active Publication Date: 2017-01-04
LUDONG UNIVERSITY
View PDF1 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Existing face recognition methods such as using the sparsity adaptive matching and pursuit algorithm for face recognition, the recognition rate of this method is not much different, and the number of iterations and the amount of calculation are very large when reconstructing
In addition, it is also easily affected by light, and its recognition efficiency is not high

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
  • Human face identification method and device based on matching pursuit algorithm
  • Human face identification method and device based on matching pursuit algorithm
  • Human face identification method and device based on matching pursuit algorithm

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0053] The specific embodiments of the present invention will be described in further detail below in conjunction with the drawings and embodiments. The following examples are used to illustrate the present invention, but not to limit the scope of the present invention.

[0054] In practical applications, there is often noise or occlusion in the face image. Considering this situation, the model is established as:

[0055] y=y 0 +e 0 =Ax 0 +e 0 (1-1)

[0056] Where e 0 ∈R M Is an error vector, R M Is the error matrix, its dimension is the same as Ax 0 Consistent, y 0 Represents the face image in the absence of noise or occlusion, x 0 Table sparsity factor.

[0057] In the face image, the noise and occlusion are only a small part, therefore, it can be considered that e 0 The non-zero elements are only a small part. Let the proportion of non-zero elements be P, that is, the proportion of noise or occlusion in an image. Since the position of the occluded part changes with the test image...

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 human face identification method and device based on a matching pursuit algorithm. The method comprises the following steps: S1, normalizing all columns of a preliminarily established dictionary matrix to L<1> norms of a unit; S2, carrying out characteristic extraction on a to-be-detected image, and establishing a to-be-detected image model as shown in the specification; S3, solving the to-be-detected image model as shown in the specification by adopting a KRSAMP method; S4, according to a value of a w0, calculating a residual ri(y) as shown in the specification between the to-be-detected image and a sample of the dictionary matrix; and S5, finding a category corresponding to a minimum value of the residual as the category of the to-be-detected image. According to the human face identification method and device based on the improved matching pursuit algorithm disclosed by the invention, the screening of atoms is carried out via an incremental gradient ascent method, so the calculating speed is accelerated.

Description

Technical field [0001] The present invention relates to the technical field of image recognition, in particular to a face recognition method and device based on a matching tracking algorithm. Background technique [0002] The security doors in the home now adopt face recognition technology. At present, a large number of face recognition algorithms are focused on how to effectively extract facial image features. Therefore, feature extraction has become a hot spot in the research of face recognition, and the number of feature selections Reconciliation directly affects the recognition effect and becomes a crucial issue in face recognition. The emergence and development of compressed sensing theory has brought new inspiration to face recognition: if the face image can be sparsely represented on a certain basis, then feature selection will no longer be a difficult point, and a large number of feature values ​​will become Advantages available in the algorithm. [0003] In the practical...

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
IPC IPC(8): G06K9/00
CPCG06V40/161G06V40/168G06V40/172
Inventor 岳峻朱华牟梦媛李振波李长青卞大鹏
Owner LUDONG UNIVERSITY
Features
  • Generate Ideas
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More