Video-based flame detection method

Abstract

The invention discloses a video-based flame detection method, and solves the technical problem in improving the flame detection accuracy. The video-based flame detection method comprises the followingsteps of obtaining a video image sequence; performing image preprocessing; performing color detection; performing flame feature extraction; and performing AdaBoost prediction. Compared with the priorart, the method has the advantages that all candidate flame points meeting a condition in a video image are found out according to characteristics of flame pixels in RGB and YCbCr color spaces; then,the video image is subjected to block segmentation; covariance matrixes corresponding to color and brightness attributes of flame pixel sets in video blocks are calculated; upper or lower triangularparts in the color and brightness covariance matrixes are extracted as eigenvectors; an obtained eigenvector set serves as an input of an AdaBoost classifier; the eigenvectors are input to an AdaBoostclassification model; an output is a judgment whether a fire happens or not; the fire happening situation can be detected out in real time; the false alarm rate can be reduced; and relatively high accuracy and strong robustness are achieved.

Description

technical field [0001] The invention relates to a safety monitoring method, in particular to a fire detection method. Background technique [0002] The existing automatic fire monitoring method mainly uses smoke sensors, temperature sensors, and infrared detectors for detection. Great disaster. [0003] The research methods of early flame detection are mainly reflected in the color model of flame. Chen et al. worked out rules that could distinguish flame pixels based on the rules between red, green, and blue RGB channels, and extracted the flame area. Celik et al. established a general rule-based classification of flame pixels on the color model of brightness blue chroma red chroma YCbCr (a color space, Y refers to the brightness component, Cb refers to the blue chroma component, Cr refers to the red chroma component) color model color model. Marbach et al. used the YUV (Y stands for Luminance, UV stands for Chrominance, which are two components of color) color model to ...

AI Technical Summary

Problems solved by technology

These flame detection methods only consider the local characteristics of the flame, which is easy to cause misjudgment, and is only suitable for flame detection in certain specific scenarios

[0004] There are already many video-based flame detection methods in the prior art, but most of them judge the flame too single, have low anti-interference, and have high false alarm rate and false negative rate. Higher and more adaptable approach to video flame detection

Images