Method and system for detecting video occlusion in video monitoring

Abstract

The invention discloses a method and a system for detecting video occlusion in video monitoring. The method comprises the specific processing steps of obtaining a video image, preprocessing the image,then extracting regions of interest, calculating the image frequency of the previous n frames; establishing an initial Gaussian model, then calculating the frequency of the new frame image, judging whether the new frame hits the Gaussian model or not, setting the number of times of abnormity as 0 if the new frame hits the Gaussian model, updating the model by [alpha]1, adding 1 if the new frame hits the Gaussian model, updating the model by [alpha]2, judging that the video is shielded when the number of times of abnormity is greater than a threshold value, and giving an alarm. According to the method, the detection speed is high, the influence of image noise and illumination change is solved, the detection accuracy is improved, and the false alarm condition is avoided.

Description

technical field [0001] The invention relates to the technical field of video surveillance image processing, in particular to a method and system for detecting video occlusion in video surveillance. Background technique [0002] With the continuous development of social economy and the continuous improvement of people's living standards, security monitoring is becoming more and more popular, and surveillance cameras can be seen almost everywhere in work and life. Whether on the road, in an elevator, or in a shopping mall, security cameras play an important role. [0003] However, if the video camera is blocked, it will have a great impact on the security system. Video occlusion, that is, the viewing of the content of the screen is affected by the interference of the surveillance camera. In many monitoring scenarios, many monitored areas are located in relatively remote places, and few people patrol to check, and some occlusions make it difficult to judge the working conditi...

AI Technical Summary

Problems solved by technology

In many monitoring scenarios, many monitored areas are located in relatively remote places, and few people patrol to check, and some occlusions make it difficult to judge the working condition of the camera with the naked eye.

Therefore, it often happens that the video information in a certain area is blocked or dusty and cannot be used, which seriously hinders the investigation time of the accident and also causes a waste of monitoring equipment resources.

[0004] At present, some existing video occlusion detection methods are greatly affected by image noise and light, not only have low detection efficiency, but also low accuracy, and often cause false positives, which brings many unnecessary problems to the later maintenance work. Trouble

Images