Gait recognition method and system based on optimal axial projection features

Abstract

The invention discloses a gait recognition method based on optimal axial projection features, the background modeling and the gait region detection are respectively carried out on the sequence of testpictures and the sequence of reference pictures of each test object, thereby obtaining a gait region image of each picture; the gait feature of each test object is extracted based on the optimal axial projection on the basis of the obtained gait region images; the similarity measurement is carried out on the extracted gait features; and the test object in the sequence of the test pictures is identified according to the similarity between the gait features of the sequence of the test pictures and the gait features of the sequence of the reference pictures of each test object. The invention simultaneously discloses a gait recognition system based on the optimal axial projection features. The use of the invention can retain more original gait information, have higher recognition rate, reducethe computational complexity and improve the calculation speed.

Description

technical field [0001] The invention relates to gait recognition technology, in particular to a gait recognition method based on optimal axis projection features. Background technique [0002] Gait recognition is a new biometric-based identification technology, and its main content is to identify individuals based on their walking posture. Compared with biometric technologies such as face recognition, fingerprint recognition and iris recognition, it has the following advantages: non-invasive, can be recognized based on low-resolution images, it is difficult to hide the recognized object, and can be recognized at a long distance. Early medical research shows that: if the 24 components of human gait are considered, each person's gait is unique. That is to say, gait recognition can uniquely identify a person's identity. [0003] At present, researchers in the field of gait recognition have proposed a large number of algorithms, which can be mainly divided into two types: mode...

AI Technical Summary

Problems solved by technology

However, both methods have their own limitations:

[0009]The second method is the gait recognition method of contour unwrapping. Although this method can retain more gait information and have a high recognition rate after reducing the data dimension, However, in the process of extracting gait features, this method needs to calculate the points on the contour (xi, yi) to the center of mass (xc, the distance between yc) needs to perform square and root operations, which is complex and slow

Images