Method and device of self-adaptive video diagnosis

Abstract

The invention provides a method of self-adaptive video diagnosis, which is applied to a video diagnosis server. The method comprises the following steps of: step 1, figuring out a result A of a diagnosis item of a currently-output video image of a front-end camera according to a diagnosis item algorithm; step 2, correcting the result A of the diagnosis time through a correction value, which is obtained by figuring up a standard result S of the diagnosis item and a result B of the diagnosis item of n previous video images obtained by current statistics, and outputting the corrected result of the diagnosis item. According to a scheme of the invention, a feedback mechanism is introduced in a video diagnosis system, so that the video diagnosis system is capable of correcting an output result as an input factor in a real-time manner, self-adaption is carried out on the environment of video diagnosis, and the false detection, which is caused by the change of environment, of the video diagnosis is reduced.

Description

technical field [0001] The present invention relates to the technical field of video monitoring, in particular to an adaptive video diagnosis method and device Background technique [0002] With the continuous expansion of the monitoring network scale, there are more and more video acquisition devices in the video monitoring network. How to effectively diagnose the abnormality of the front-end equipment through the video diagnosis system in a timely and accurate manner is becoming more and more important. At present, many manufacturers have invested in the field of video diagnosis, but the video diagnosis systems of each manufacturer have the problem of poor environmental adaptability. When the video surveillance network to be diagnosed is relatively large, serious no-alarms may occur. For example, a video surveillance network needs to diagnose continuously for 12 hours. Due to the large time span, the surrounding environment will change greatly. In this way, the diagnostic...

AI Technical Summary

Problems solved by technology

[0004] The flexibility of this solution is not enough. For example, if the daytime diagnosis solution encounters a cloudy or rainy day, it will have a serious impact on the video diagnosis alarm.

In the environment of some large-scale video surveillance networks (tens of thousands of front-end devices), the time span is relatively large, and the diagnosis results during the day and night are quite different, resulting in a relatively high probability of misdiagnosis in actual use.

Images