Method and system for real-time detection of early fire smoke

Abstract

The invention belongs to the technical field of fire monitoring and alarming, and specifically relates to a method and system for real-time detection of early fire smoke; the technical problem to be solved is: to provide a fire detection system with simple operation, fast detection speed, and effective removal of the influence of non-smoke interference objects. The method and system for real-time detection of early fire smoke adopts the following technical scheme: a real-time detection method for early fire smoke, including: calculating the total number of frames of the video image to be tested, and converting the video image to be tested into a grayscale video image; reading the video The current frame detects the moving foreground of the grayscale video image, and uses the first frame image of the grayscale video image as the background image; performs a product operation on the background image and the moving foreground fgmask; performs a product operation on the current frame and the moving foreground; Frame difference; select the detection period and use the least square method to fit the absolute value of the change rate k, and judge the smoke alarm signal.

Description

technical field [0001] The invention belongs to the technical field of fire monitoring and alarming, and in particular relates to a real-time detection method and system for early fire smoke. Background technique [0002] The danger of fire is self-evident. Regardless of whether it is a city or a forest, once a fire breaks out, the losses caused are inestimable. In order to reduce the losses caused by fires, it is very important to detect fires in time and issue an alarm in the early control of fires. . [0003] Traditional smoke detection generally uses temperature sensing, smoke sensing, photometric sensing, odor sensing, etc. to detect special particles, unique gases, and temperature changes in the smoke; such as a smoke detector, its working process is: the smoke detector It is installed near the target object, and the detector needs to be triggered after the smoke arrives. Since the early smoke diffusion rate is usually slow, it takes a long time to reach the detector,...

AI Technical Summary

Problems solved by technology

[0005] Combining the current mainstream methods of smoke detection, it can be roughly divided into three steps: moving target detection, feature extraction and recognition. These methods are easily affected by environmental factors, have poor adaptability, and complex calculations, which are difficult to apply to real-time video fire monitoring systems. In addition, the insurmountable defect in the above-mentioned methods is that it is easily affected by non-smog disturbances such as pedestrians and automobiles in the video screen, resulting in misjudgment (higher false alarm rate), especially in complex outdoor environments. The rate of return remains high and it is difficult to be practical

Images