Smoke fire detection method and system based on vision

Abstract

The invention relates to the technical field of fire fighting, in particular to a smoke and fire behavior detection method and system based on vision, and the method comprises the following steps: an image obtaining step: collecting and extracting a smoke and fire behavior image and a negative sample from the image; an image preprocessing step: preprocessing the acquired smoke and fire behavior image and the negative sample to become training data; a model training step: inputting the training data to the involute residual deep neural network to set network parameters for several times of training, and storing a trained detection model; and a fire behavior identification step: acquiring an image, converting the image into a model input format, inputting the image into the trained detection model, and judging whether a smoke fire behavior occurs in the monitoring area. According to the method, the position information of the smog fire behavior of different scales can be obtained through the involute residual deep neural network, and the effective extraction of the varied smog fire behavior features is satisfied, so that the detection rate of the network is higher, and the generalization performance of the model is better.

Description

technical field [0001] The invention relates to the technical field of fire protection, in particular to a method and system for detecting smoke and fire based on vision. Background technique [0002] With the continuous development of society, while the social wealth is increasing, the risk of smog fires is also increasing. In recent years, smog fires have become more frequent in residential buildings, vehicles, industrial production, forests, etc. , causing huge loss of life and property to people, so how to detect smoke and fire in time and issue an alarm has become the focus of research. [0003] Early smoke and fire detection was mainly judged based on smoke, temperature and light, which was not conducive to the early detection of smoke and fire, and could not be applied to large spaces and outdoor environments; in recent years, image-based fire detection has also With continuous development, the traditional smoke and fire detection method is a machine learning method ...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to provide a method and system for detecting smoke and fire based on vision, so as to solve the problem of low detection accuracy proposed in the above-mentioned background technology

Images