Fire recognition method based on flame detection in night vision scene

Abstract

The invention discloses a fire recognition method based on flame detection in a night vision scene. The fire recognition method comprises the steps of: step 1, acquiring an image sequence of a monitored scene by means of a monitoring camera with a night vision function; step 2, subjecting each frame of the acquired images to luminance segmentation, and carrying out background modeling and motion detection on the acquired image sequence; step 3, performing logic screening on an alternative flame region to find sporadic bright spots on the periphery of large-area bright spots, and filtering out interference noise of the sporadic bright spots; step 4, calculating cumulative distribution on a timeline, and forming a cumulative distribution probability image of the bright spots; step 5, carrying out binarization and clustering operation on the obtained cumulative distribution probability image in space, and filtering out clusters with probability less than 50%; step 6, extracting features of each of the rest clusters, and performing randomness detection on the timeline; step 7, and judging whether a fire occurs in the monitored scene. The fire recognition method overcomes the problem of high false alarm rate since the night fire detection method is susceptible to interference in the prior art.

Description

technical field [0001] The invention relates to the field of flame detection, in particular to a fire recognition method based on flame detection in a night vision scene. Background technique [0002] Fire is one of the most frequent and destructive disasters. The traditional fire detection method is based on smoke, heat and other detectors. It has achieved good results, but has the following shortcomings: Not applicable In open space and large space (including outdoor environment); the fire detection time is late, and the best time to put out the fire is missed. For this reason, some researchers use infrared cameras instead of those detectors and use video and image processing techniques to detect fires, but the equipment costs are too high. With the rapid development of visible light surveillance cameras and the large-scale promotion of video surveillance, fire detection based on visible light cameras has attracted the attention of more and more researchers at home and ab...

AI Technical Summary

Problems solved by technology

[0004] In view of the above-mentioned technical problems, the purpose of the present invention is to overcome the problems of high false alarm rate and low detection rate in the fire detection method in the prior art, thereby providing a method that can effectively improve the detection rate and reduce the false detection rate. Fire recognition method based on flame detection in night vision scene

Images