Flame detection method based on video analysis

Abstract

The invention relates to the field of video monitoring and particularly relates to a flame detection method based on video analysis. The flame detection method comprises the following steps: extracting a binary image of a foreground according to a background in a scene image and getting a plurality of foreground object images according to the binary image of the foreground; and screening according to the color feature of a red, green and blue (RGB) histogram of the plurality of the foreground object images and the scale invariant feature transformation (SIFT) of the plurality of the foreground object images or the speeded up robust feature (SURF) so as to screen out the foreground object images which are in line with the flame feature. Through the flame detection method based on the video analysis, which is provided by the embodiment of the invention, the flame situation can be known more efficiently and more accurately.

Description

technical field [0001] The invention relates to the field of video monitoring, in particular to a flame detection method based on video analysis. Background technique [0002] Fire is a disaster that brings huge losses to the people, so flame detection is an important part of all security systems. Traditional flame detection is mainly based on sensors such as temperature, infrared, smoke, etc. Although these methods are cheap and easy to implement, they cannot be applied to all fire situations because their detection range is limited and they are not smart enough. In recent years, with the popularity of surveillance cameras and the development of computer vision, people have begun to propose some flame detection solutions based on video surveillance. But these methods are too simple and not mature enough to be put into practical application. [0003] Color characteristics are a distinctive feature of flames and an important basis for judging flames. Therefore, color featu...

AI Technical Summary

Problems solved by technology

However, there are many problems with the use of color features in these algorithms: First, the color model is too simple, mostly generated by artificial statistics, which only depends on experience and cannot cover all situations, such as bright flames under the overexposure of the camera; secondly , the color feature is simply judged pixel by pixel, and does not combine the motion information of the target in the specific scene; finally, many algorithms only use the color feature to judge the flame, while ignoring other important features, but even if the color feature is robust, Relying on it alone is unreliable

However, in fact, the beating of the flame is not completely a random event. In practical applications, airflow often affects the movement of the flame, and the occlusion of people can also give the computer an illusion.

Therefore, in these cases, the judgment method based on HMM will fail

Images