Multi-target tracking and facial feature information identification method

Abstract

The invention discloses a multi-target tracking and facial feature information identification method. The method comprises the following steps: firstly, inputting a facial video and converting the facial video into a human face key frame set SF; obtaining a data set SFD through classification and data enhancement SF; extracting SFD features by using an improved miniXaction model to obtain a self-adaptive optimization facial feature recognition model FFsmodel; extracting a key frame sequence in the face data stream FEV, and adaptively aggregating face tracking features to obtain a multi-targetface key frame position set FEC; loading an FFsmodel model and inputting the FFsmodel model into FEC to generate a multi-target facial feature classification result set FECR; and finally, opening an adaptive identification interface to process a terminal request, and storing an obtained processing result set of sign-in and target feature tracking state identification in a multi-target scene in a Web server. The improved multi-target tracking and self-adaptive facial feature recognition technology is combined, the picture feature result label with the highest accuracy can be effectively obtained, and the accuracy and use value of face tracking and facial feature recognition in a multi-target scene are improved.

Description

technical field [0001] The invention belongs to the field of image classification and feature recognition, in particular to a method for multi-target tracking and facial feature information recognition. Background technique [0002] With the rapid development of computer technology and AI in recent years, people have gradually begun to pay attention to the use of computer means to track and identify human facial feature information. Detecting and tracking the facial feature information of multiple targets in real-time video is of great practical significance in schools and railway stations. It can not only provide schools with effective feature information result sets in classrooms, but also avoid certain security risks. For this reason, the present invention proposes a multi-target tracking and facial feature information recognition method based on the improved mini-Xception model and self-adaptive optimization. Face feature tracking and recognition results are more accur...

AI Technical Summary

Problems solved by technology

[0012] In terms of multi-target feature information tracking and detection, most of the current research is mainly oriented to the unilateral processing of multi-target tracking or feature recognition models, and there is a lack of research on multi-target feature adaptive classification methods in key frame sequences with timing attributes. The fusion is relatively simple, and the tracking and analysis efficiency of multi-target feature information under data with time series attributes is also limited

Images