Gait recognition method based on deep transfer learning

A technology of transfer learning and gait recognition, which is applied in the field of pattern recognition, can solve the problems of identity authentication passivity, etc., and achieve the effect of improving robustness and good recognition performance

Pending Publication Date: 2021-04-06
HANGZHOU DIANZI UNIV
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Biometric recognition technologies such as fingerprint recognition, iris recognition, and face recognition, which are widely used at present, mostly require the cooperation of the detected object, and sometimes require the detected object to complete specific actions to be recognized, which will inevitably lead to passive identity authentication. sex

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
  • Gait recognition method based on deep transfer learning
  • Gait recognition method based on deep transfer learning
  • Gait recognition method based on deep transfer learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0044] Such as Figures 1 to 8 As shown, the first major step of the present invention is gait data preprocessing, and its specific implementation steps are as follows:

[0045] The collected gait video is extracted frame by frame, and the color image is grayscaled to obtain the grayscale gait image H n (x, y) for further identification; delete frames that do not have a complete human body contour map (considered as useless frames) in the picture, and retain frames that have a complete human body contour map (considered a useful frame); RGB background image Convert to a grayscale background image T n (x, y), and average the multiple frames to obtain the mean background image Q(x, y);

[0046]

[0047] Carry out the background subtraction operation, use the formula (2) to remove the background image Q(x,y) and extract the foreground moving image P n (x,y), retain the complete human body contour image;

[0048] P n (x,y)=|H n (x,y)-Q(x,y)| (2)

[0049] When only extrac...

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 gait recognition method based on deep transfer learning. The method comprises the following steps of 1, performing frame sequence picture extraction on a gait video, performing morphological processing on a gait sequence picture subjected to background subtraction, removing useless backgrounds, and only reserving a human body gait contour picture, then performing up-down and left-right cutting on the human body, and performing normalization processing on the left half body, the right half body, the upper half body, the lower half body and the whole body of the human body side image contour map; 2, inputting the five human body side-shadow contour maps under the same visual angle into a deep migration learning network, adjusting migration network parameters, training to obtain five deep migration models, and carrying out model integration on a probability matrix of an output layer; and 3, carrying out model integration on the human gait side-shadow contour maps under the n visual angles to obtain a final integration model, and obtaining the identity category of the testee according to the obtained final probability. The method has good recognition performance, can reach a good classification effect, and has good model transportability.

Description

technical field [0001] The invention belongs to the field of pattern recognition, and in particular relates to a gait recognition method based on deep transfer learning. Background technique [0002] Biometric technology is a method of identifying an individual. It uses high-tech information detection technology and uses the inherent physiological or behavioral characteristics of the human body to identify individuals, including face recognition, fingerprint recognition, iris recognition, and gait recognition. Since each person's biometrics are unique and universal, it is not easy to forge and counterfeit, so the use of biometric technology for identity authentication has the advantages of safety, reliability, and accuracy. Biometric recognition technologies such as fingerprint recognition, iris recognition, and face recognition, which are widely used at present, mostly require the cooperation of the detected object, and sometimes require the detected object to complete spe...

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 Applications(China)
IPC IPC(8): G06K9/00G06K9/46G06K9/62G06N3/04G06N3/08G06T7/215G06T7/246G06T7/254G06T7/11
CPCG06T7/215G06T7/246G06T7/254G06N3/08G06T7/11G06T2207/10016G06T2207/20081G06T2207/20084G06T2207/20224G06T2207/30196G06V40/25G06V10/267G06V10/44G06N3/045G06F18/2415
Inventor 于雪东林鹏曹九稳王建中
Owner HANGZHOU DIANZI 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