Method and system for identifying facial features based on a security video

Abstract

The invention provides an indoor positioning method and system for a security and protection monitoring project. The method comprises the following steps: step 1, carrying out real-time image acquisition through an erected security monitoring camera, and returning a monitoring image of the monitoring camera in real time; step 2, preprocessing the monitoring image and filtering and calculating a face shielding object; step 3, restoring the facial features of the to-be-identified target after the shielding object is removed; 4, performing face comparison on the restored facial features; and 5, carrying out decision command according to a face comparison result. According to the invention, a monitoring camera and an image automatic identification technology which are arranged in a public security and protection manner, and an optimized facial feature extraction filtering method and an artificial intelligence deep learning algorithm are utilized; the problems of low face recognition accuracy and long recognition period in the current security and protection field, especially in the face shielding condition, are solved, a reliable basis is provided for security and protection syntheticcommand decision making, the decision making efficiency of security and protection synthetic command is improved, and public potential safety hazards caused by inaccurate recognition are reduced.

Description

technical field [0001] The invention relates to a video recognition technology for public security governance, in particular to a method and system for automatically identifying facial features and performing face comparison and matching in smart public security videos under facial occlusion. Background technique [0002] As a hot research issue in the field of pattern recognition, face recognition has received extensive attention. Face recognition technology has broad application prospects in identity verification in many fields. In the actual process of face image processing, occlusion of the face image often occurs, such as hair, mask, scarf, beard, etc., and occlusion has a great impact on face recognition. Therefore, how to detect face occlusion regions accurately and automatically becomes one of the key issues in occluded face recognition processing. Especially in the field of security, some extremists have masked burqas and long beards, causing great difficulties in ...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to solve the problems of inaccurate recognition and low recognition efficiency in the existing actual situation, and provide a method and system for recognizing facial features based on security video. It is stored in the security video recognition system of the prevention and control center, and the received video is calculated in real time to perform image preprocessing. If the target to be recognized is blocked, the image is handed over to the blocking object filtering processing module to filter and restore facial features, and then The restored facial features are compared and identified with various public security personnel information databases, and the identified identity information is finally transmitted to various command application systems such as public security command, anti-terrorism command, traffic control command, and major security command. The system conducts decision-making and action command based on information results

Images