Smoke and fire detection method based on video

Abstract

The invention relates to the technical field of image processing, in particular to a smoke and fire detection method based on videos. The method includes the steps of image acquisition, image preprocessing, image combination, smoke and fire target detection, deep learning and the like. According to the design of the invention, through combination of smoke and fire motion information and the advantages of deep learning, three-channel color image input of deep learning target detection is modified into a multi-channel image formed by combining images of the same camera at different time points; then, the moving smoke and fire targets possibly existing in the image are detected and tracked through multiple target detection algorithms or two or more combined algorithms, so that the probability of misinformation can be effectively reduced, the requirement of the method for image quality is reduced, the detection accuracy is improved, and the application scene of the smoke and fire detection method is expanded; in conclusion, the method can be effectively applied to safety monitoring of the environment.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a video-based smoke detection method. Background technique [0002] At present, the deep learning method of fireworks detection basically adopts the method of sample labeling and model training. This method is applicable in most scenarios, but it requires relatively high image quality. For example, there should be no objects similar to smoke or fire in the picture. , or something with a similar color, otherwise the detection accuracy will be low, and a stationary object or an object that moves too slowly may be misjudged as a target. Contents of the invention [0003] The purpose of the present invention is to provide a video-based smoke detection method to solve the problems raised in the above-mentioned background technology. [0004] In order to solve the above-mentioned technical problems, one of the objects of the present invention is to provide a method for dete...

AI Technical Summary

Problems solved by technology

[0002] At present, the deep learning method of fireworks detection basically adopts the method of sample labeling and model training. This method is applicable in most scenarios, but it requires relatively high image quality. For example, there should be no objects similar to smoke or fire in the picture. , or something with a similar color, otherwise the detection accuracy will be low, and a stationary object or an object that moves too slowly may be misjudged as a target

Images