Pixel-level fire smoke root node detection method based on connected particles

Abstract

The invention relates to a pixel-level fire smoke root node detection method based on connected particles. The method comprises the following steps: firstly, carrying out histogram equalization on an input image, respectively executing a dynamic feature extraction algorithm and a static feature extraction algorithm on the equalized image, and then fusing a dynamic image and a static image by using a breadth-first search algorithm to obtain a complete dynamic region contour map; detecting the number of angular points in each dynamic region of a single-frame image by using an angular point detection algorithm, and setting a threshold value to screen out a smoke region; and finally, calculating representative coordinates of smoke root nodes in each region by using a horizontal line scanning method, and screening out and determining the smoke root nodes by using a circular template.

Description

technical field [0001] The invention belongs to the fields of forest fire prevention and video target detection, and in particular relates to a pixel-level fire smoke root node detection method based on connected granules. Background technique [0002] Early forest fires relied on human patrols and regular observations from watchtowers, but the actual monitoring results were not ideal. With the rapid development of computer technology, more and more people apply computer vision technology to fire detection. In a complex forest environment, forest fires are difficult to detect at the initial stage, and the fire is uncontrollable by the time it is discovered. Smoke is the most important feature in the early stage of a fire, so it is easy to be found. However, in most scenarios, due to the slow movement of smoke and the The color is translucent and it is easy to cause missed detection. Therefore, it is of great significance to propose a method for accurately detecting smoke in...

AI Technical Summary

Problems solved by technology

In a complex forest environment, forest fires are difficult to detect at the initial stage, and the fire is uncontrollable by the time it is discovered. Smoke is the most important feature in the early stage of a fire, so it is easy to be found. However, in most scenarios, due to the slow movement of smoke and the The color is translucent and it is easy to cause missed detection. Therefore, it is of great significance to propose a method for accurately detecting smoke in slow-moving scenes.

Images