Gait recognition method, device and system based on artificial intelligence, storage medium and server

Abstract

The invention discloses a gait recognition method, device and system based on artificial intelligence, a storage medium and a server. The gait recognition method comprises the following steps: A, recognizing a human body in a video; B, identifying and tracking the main joints of the human body, and storing the position information of the main joints in the video frames to form a video sequence; C,learning and extracting gait features in the video sequence; and D, comparing the gait features with gait features preset in a storage module to obtain a recognition result. The gait recognition device comprises a moving human body recognition module, a human body joint tracking module, a self-encoding learning module, a gait feature recognition module and a storage module. The gait recognition system comprises a video acquisition terminal, a use terminal and a server. Based on the artificial intelligence technology, invalid search of the optimal gait feature in a high-dimensional space caused by the manual selection of the gait features is avoided, and robust and complete extraction of the optimal gait feature is achieved to the maximum degree.

Description

technical field [0001] The invention belongs to the technical field of biological feature recognition, and in particular relates to a gait recognition method, device, system, storage medium and server based on artificial intelligence. Background technique [0002] Gait refers to the posture shown by the human body when walking and all the movements of walking. Gait information includes but not limited to information such as step frequency, step length, lower limb swing angle and acceleration when walking. Generally speaking, human gait can be composed of seven time-continuous stages, which are: heel strike, sole strike, heel lift, toe off the ground, early swing, middle swing, and tail swing (heel strike). These stages form a basic gait cycle. The gait cycle of each person's feet represents their unique biological characteristics. Therefore, gait, like fingerprints, irises, and human faces, can be used as a biological characteristic of the human body for natural person iden...

AI Technical Summary

Problems solved by technology

The feature space of gait features is generally a high-dimensional space, and its features include both the spatial dimension and the time dimension. The gait features are determined artificially, and it is impossible to accurately determine the feature vector that best represents the human gait feature. When selecting gait features in the feature space of the gait sequence, the probability of missing the most valuable and most accurate features of human gait is very high, and the possibility of obtaining the optimal gait feature is relatively small

At the same time, these shortcomings are also one of the important reasons why gait recognition technology cannot be used and deployed on a large scale.

Images