Character identity recognition method combining face recognition and human body recognition

Abstract

The invention relates to the technical field of video monitoring, in particular to a character identity recognition method combining face recognition and human body recognition, which comprises the following steps: acquiring clear face images and human body images of passengers at a security check place, then receiving video frames in monitoring equipment in a waiting room, recognizing the face images in the video frames, and if the face recognition is successful, carrying out human body recognition, if a human body feature comparison score obtained through human body recognition is smaller than a clothes changing threshold value, determining that the passenger has a clothes changing behavior, and replacing the human body image corresponding to current successful face recognition with the old human body image of the current passenger; if the face recognition fails, carrying out human body recognition, sending out an alarm signal only when the human body recognition fails, and otherwise, not sending out the alarm signal. According to the method, the face recognition technology is combined with the pedestrian recognition technology, and the false alarm rate of character identity recognition in a non-cooperative scene is reduced. The method has the advantages that the false alarm rate can be reduced, and meanwhile the clothes changing behavior of the passenger can be recognized.

Description

technical field [0001] The invention relates to the technical field of video monitoring, in particular to a person identification method combining face recognition and human body recognition. Background technique [0002] At present, face recognition technology has been widely used in the automatic security inspection system of high-speed rail, and it is mainly used in the comparison of face images and ID card face images on the high-speed rail scene. With the continuous development of biometric technology and the continuous expansion of the high-speed rail station's demand for automatic security inspection system efficiency improvement, more and more high-speed rail stations now need to perform identity verification at the security checkpoints such as high-speed rail In other non-user cooperation scenarios such as carriages and waiting platforms, it is also necessary to determine whether the characters appearing in the video image are the corresponding passengers of the cur...

AI Technical Summary

Problems solved by technology

[0003] The recognition rate of face recognition technology in the prior art in this non-cooperative scene will also be greatly affected by various conditions such as illumination and face angle, resulting in a high possibility of false alarms, and high The false alarm rate requires frequent manual review and verification, which reduces user experience and increases the labor cost of the station

Images