A gait feature fusion method based on nonlinear coupling metric learning

A technology of nonlinear coupling and metric learning, which is applied in the field of gait feature fusion based on nonlinear coupling metric learning, can solve the problems of low recognition rate and degradation of recognition performance, and achieve the effect of high recognition rate

Inactive Publication Date: 2019-04-16
SICHUAN NORMAL UNIVERSITY
View PDF3 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0013] The purpose of the present invention is to solve the gait characterization method in the prior art because there are few useful gait features for recognition, resulting in low recognition rate; especially when the walkin...

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
  • A gait feature fusion method based on nonlinear coupling metric learning
  • A gait feature fusion method based on nonlinear coupling metric learning
  • A gait feature fusion method based on nonlinear coupling metric learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0091] 1.1 Test sample: Use the USF HumanID gait dataset (the USF HumanID gait dataset is an outdoor gait video taken under a long-distance complex background, and the quality of the extracted side profile is poor). Collect video data of 122 people under different conditions: according to the collection angle of view (R / L) (the angle between the optical axes of the left and right cameras is about 30°), walking on cement / grass (C / G), carrying / not backpacking ( BF / NB), wearing different shoes (A / B) and different acquisition time (T), divided into multiple different groups, the specific division is shown in Table 1.

[0092] Table 1 USF database situation

[0093]

[0094]

[0095] 1.2 Test method: The experiment uses the support of the feet as the starting point of the gait cycle, obtains several energy-like maps of a gait sequence, and then averages them into a single-cycle GEI and AEI for analysis, and uses corresponding feature extraction on this basis method (specific...

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 feature fusion method based on nonlinear coupling metric learning, and belongs to the technical field of pattern recognition. The technical problem that a gait characterization method in the prior art is low in recognition performance is solved. The gait feature fusion method disclosed by the invention comprises the following steps of: 1, acquiring a binary contour sequence of a person from a gait video stream by adopting a codebook detection method, and standardizing and centralizing each frame of image; 2, detecting a non-front gait period according to the periodicity of the separation degree of the two legs of the person in walking, and extracting gait energy diagram characteristics and active energy diagram characteristics in one period; Step 3, performing nonlinear coupling metric learning on the gait energy diagram features and the active energy diagram features, and projecting the features to a nuclear coupling space to obtain two new features; Andstep 4, carrying out weighted fusion on the two groups of obtained new feature vectors to obtain new gait features of the nuclear coupling space. The gait feature fusion method is effective for non-frontal periodic gait sequences and high in recognition rate.

Description

technical field [0001] The invention belongs to the technical field of pattern recognition, and in particular relates to a gait feature fusion method based on nonlinear coupling metric learning. Background technique [0002] Gait recognition is a research direction that has attracted much attention in the field of computer vision and biometric recognition in recent years. Compared with other biometric recognition technologies, gait recognition is the only long-distance recognition method in biometric recognition [1] . The advantages of non-contact gait and difficult camouflage have great application prospects in intelligent video surveillance. [0003] However, pedestrians will be affected by the external environment and their own factors during walking, such as different walking surfaces, different viewing angles, different clothing and other factors. In the presence of the above influencing factors, the difference in gait representation brings difficulties to gait recogn...

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): G06K9/00G06K9/62
CPCG06V40/103G06F18/2411
Inventor 吕卓纹谢瑞强冯娟宰文姣王一斌
Owner SICHUAN NORMAL UNIVERSITY
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