Background edge model-based video camera anomaly detection method and system

Abstract

The invention discloses a background edge model-based video camera anomaly detection method and a background edge model-based video camera anomaly detection system. A PBAS foreground detection method is used for extract a foreground picture, and a background sample picture is obtained by updating via weighted random, so the method and the system of the invention could perform correct detection under a scene where the light condition is complex and large crowds move, and is not sensitive to noise, and is strong in anti-jamming capability. By extracting a edge feature via edge detection and combining self-adapting edge threshold value to preliminarily judge abnormal condition of a video camera, the method and the system of the invention could be used for fast and preliminarily judging abnormal condition of the video camera, thereby consuming less CPU and less memory resource. By using surf angle characteristic to eliminate the phenomenon that the normal condition that a group of people appeared in a monitoring picture is falsely alarmed as the abnormal condition of the video camera, thereby improving accuracy of anomaly detection. The method and the system of the invention could be widely applied in the field of video monitoring.

Description

technical field [0001] The invention relates to the field of video monitoring, in particular to a camera anomaly detection method and system based on a background edge model. Background technique [0002] In recent years, with the rapid development of economy and society and the continuous advancement of science and technology, video surveillance has penetrated into various fields of people's daily life. Video surveillance is of great significance to building a harmonious society. In the field of security, with the rapid growth of urban population and the increasingly complex urban environment, various crimes, mass incidents, riots, terrorist attacks, etc. are increasing. Video surveillance systems play an important role in effectively controlling various incidents. The Ministry of Public Security of our country organized the "3111" project to promote the construction of urban alarm and monitoring nationwide. Building a video surveillance network is currently one of the m...

AI Technical Summary

Problems solved by technology

[0004] (1) It is mainly analyzed in static scenes with few people, and cannot be correctly detected in scenes with complex lighting conditions and a large number of crowd activities. It is sensitive to noise and has weak anti-interference ability;

[0005] (2) It will take up a lot of memory and consume a lot of CPU resources of the server when the analysis is running;

[0006] (3) It is easy to misreport the normal situation of a group of people in the monitoring screen as abnormality of the camera, which is not accurate enough

Images