Research on image processing technology of dynamic face video recognition based on computer vision monitoring

Abstract

At present, face recognition methods are mainly used to identify static images. In surveillance video, faces in different video frames are correlated and only part of the face can effectively reflect face information; according to the characteristics of face image changes in surveillance videos, A face recognition method based on surveillance video is proposed; firstly, a video face sequence is obtained by combining face detection and tracking technology, and then a partial face image recognition result in the video face sequence is taken as a guide to select all face sequence images The representative face image in the image is recognized, and finally the face information is comprehensively reflected according to the recognition results of all the selected face images; the experimental results show that this method can effectively improve the face in the surveillance video while ensuring the recognition rate and the false recognition rate. Real-time recognition.