A video monitoring system and its real-time gait recognition method

Abstract

The invention provides a video monitoring system and its real-time gait recognition method. In the image preprocessing process, the method cuts the extracted binary contour image according to its minimum circumscribed rectangle, and keeps the aspect ratio unchanged. Next, the image is scaled according to the maximum value of the image height of all image frames. Therefore, all preprocessed image frames not only have the same height, but also maintain the same aspect ratio before and after image processing, making it easier to extract gait features. Moreover, the gait recognition method comprehensively considers the static and dynamic characteristics of human gait, and extracts the centroid height, stride length, and the rate of change of angle and distance of lower limb contour as gait features, which comprehensively reflect the gait differences of different individuals. Through the above method, the video surveillance system effectively shortens the recognition time on the basis of ensuring a high recognition rate, and at the same time has good robustness to the user's clothing change, and is especially suitable for real-time user identification in a video surveillance environment.

Description

technical field [0001] The invention relates to the field of pattern recognition of video surveillance, in particular to a video surveillance system and a real-time gait recognition method thereof. Background technique [0002] With the development of computer science and technology and the progress of human society, people's awareness of safety precautions in life, work, travel, etc. has gradually increased. How to ensure social safety has become the most serious problem facing people today. Today, as an effective way to protect people's lives and property safety, biometric recognition technologies such as face recognition, voice recognition, and iris recognition are more and more widely used. Nevertheless, since criminals can use various camouflage methods to hinder the collection of these biometrics, so that the recognition accuracy of these technologies is greatly reduced. In addition, these technologies also have shortcomings such as tampering, high requirements for th...

AI Technical Summary

Problems solved by technology

[0004] At present, although the existing gait recognition methods have achieved good recognition results, they are still unsatisfactory in practical applications; many methods are obtained by sacrificing the recognition time, so that the recognition takes a long time. There is still a certain gap with the real-time identification required by the video surveillance system

Images