Gait-based personnel identity recognition method and device and electronic equipment

Abstract

The embodiment of the invention provides a gait-based personnel identity recognition method and device and electronic equipment; after personnel key point information in each frame is extracted, gait scoring is performed on personnel in each frame; if the gait score of the personnel in a certain frame meets a preset threshold value, the personnel in the frame is considered to be in a normal walking gait; if the gait score of the person in a certain frame does not meet the preset threshold value, it is indicated that the person in the frame is in an abnormal walking gait; all the frames with the gait scores meeting the preset threshold value form a gait sequence, all the frames with the normal walking gait of the person are screened out, and then gait features are extracted based on the gait sequence. According to the invention, the gait feature extraction is only performed on each frame of the normal walking gait of the personnel, the workload of extracting the gait features is reduced, and the personnel identity recognition is performed by using the extracted gait features, so that the personnel identity recognition efficiency is improved.

Description

technical field [0001] The invention relates to the technical field of video monitoring, in particular to a gait-based personnel identification method, device and electronic equipment. Background technique [0002] With the continuous development of society and technology, areas such as community personnel management, employee attendance management, public transportation fee management, and public safety management are becoming more and more intelligent. The realization of intelligentization in the above fields requires automatic identification of personnel identities. In the current personnel identification, identification is usually based on various human biometrics, which are characteristics unique to each person, such as iris, fingerprints, face, gait, etc. Compared with other biological features, gait has the advantages of being difficult to steal and imitate due to its unique non-contact characteristics, and is especially suitable for long-distance person identificatio...

AI Technical Summary

Problems solved by technology

[0004] When performing personnel identification, the feature that can really be distinguished from other personnel is the normal walking gait of the personnel. In the above personnel identification method, it is necessary to extract the gait features of each frame of personnel, and in most scenarios, the The monitoring personnel are not always in a normal walking gait. If the gait features are extracted in each frame, the above-mentioned personnel identification method will run very slowly, which will greatly affect the efficiency of personnel identification.

Images