Multi-visual-angle gait recognition method based on optimal discrimination coupling projection

Abstract

The invention provides a multi-visual-angle gait recognition method based on optimal discrimination coupling projection. The multi-visual-angle gait recognition method comprises the following steps: extracting a plurality of gait video sequences at a known visual angle in a training set to obtain an optimal discrimination coupling projection matrix pair; extracting and storing a gait video sequence at a standard visual angle in a registration set; and extracting and estimating the gait video sequences in a test set; projecting standard visual angle gait characteristics in the registration set and gait characteristics in the test set into mutual coupling gait characteristic space which exhibits optimal discrimination capability or optimal class separability through an optimal discrimination coupling projection matrix selected by visual angle estimation, and carrying out similarity measurement to obtain a gait recognition result. The test visual angle gait characteristics and registration standard visual angle gait characteristics are projected into joint coupling space which exhibits strongest discrimination capability, so that a multi-visual-angle gait recognition still can obtain a high recognition rate when a test gait visual angle and a standard gait visual angle in the registration set exhibit obvious differences.

Description

technical field [0001] The invention relates to a gait recognition method, in particular to a multi-view gait recognition method. Background technique [0002] Gait recognition is an emerging biometric technology that aims to identify people through their walking posture. Compared with other biometric recognition technologies, gait recognition is the only method of long-distance recognition in biometric recognition, and its advantages include non-contact, non-invasive, easy to perceive, difficult to hide, difficult to camouflage, etc. Based on the above advantages, gait recognition has broad application space and high economic value in the fields of access control systems, security monitoring, human-computer interaction, and medical diagnosis. [0003] However, gait recognition also faces many difficulties in practical applications, mainly manifested in the fact that pedestrians are affected by the external environment and their own factors during walking, such as different...

AI Technical Summary

Problems solved by technology

The problem of perspective change is an unavoidable problem for gait recognition, because the walking direction of people is completely free and random, and there are differences in the positions of cameras in different areas

At present, the traditional gait recognition technology, such as the "gait-based long-distance identification method" disclosed in the patent document with the publication number CN1426020, can achieve good performance under a fixed viewing angle, but when the viewing angle changes drastically or there is an occlusion In the case of the situation, it cannot be applied or the recognition performance is significantly reduced

In theory, saving the gait information of all individuals under multiple viewing angles in the registration set can improve the recognition performance of traditional gait recognition methods when the viewing angle changes. However, storing gait information from multiple viewing angles will consume a lot of storage resources and is inconvenient. practical application

Images