A fire identification device based on multi-feature fusion

Abstract

The invention relates to a fire identification device based on multi-feature fusion, which comprises an image acquisition device used for acquiring and collecting video images; an energy spectrum extraction module used for extracting energy spectrum information of the video images; a moving object detection module for detecting a moving object in a video image; a feature extraction module configured to extract an area change feature, a perimeter change feature and a flame stroboscopic feature of the moving target; a fusion discrimination module, which, according to the energy spectrum information, area change characteristics, perimeter change characteristics and flame stroboscopic characteristics, establishes a discriminant model, and, according to the discriminant model, determines whether a fire occurs. The invention can effectively improve the accuracy of flame identification.

Description

technical field [0001] The invention relates to the technical field of fire recognition, in particular to a fire recognition device based on multi-feature fusion. Background technique [0002] Fire is a common catastrophic problem faced by people all over the world. It has caused many serious losses of life and property to human society. In order to avoid major accidents caused by fire, people have adopted many fire early detection methods, and most of these methods use sensors to detect air humidity, temperature or smoke to issue an alarm. The biggest disadvantage of these automatic fire monitoring methods, such as smoke and temperature detectors, is the distance limitation. The effective detection distance is relatively small. It is easily affected by factors such as space altitude temperature and humidity and cannot function effectively. In order to overcome the shortcomings of traditional detectors, many researchers have proposed a video-based flame detection method. ...

AI Technical Summary

Problems solved by technology

[0003] At present, video-based fire recognition algorithms usually combine static features with dynamic features to make a comprehensive judgment on multiple interrelated features in the fire, so as to more accurately reflect the comprehensive characteristics of real fire phenomena, improve the recognition rate, and make the detection method It is more intelligent. Although there are many studies on fusion algorithms, most of them use static features as the criterion for whether a fire occurs, and there are too few dynamic features, which makes the commercialized fire detection system on the market have a high false alarm rate.

Images