Unsupervised Segmentation of Fire and Smoke from Infra-Red Videos

TL;DR

This paper presents an unsupervised vision-based system for fire and smoke segmentation in infrared videos, utilizing spatial, temporal, and motion features with multiple clustering algorithms, notably Markov Random Fields, achieving high detection accuracy.

Contribution

It introduces a novel unsupervised segmentation approach combining multiple features and clustering algorithms, with Markov Random Fields outperforming others in fire detection tasks.

Findings

Achieved 95.39% frame-wise fire detection rate.

Markov Random Field outperforms other clustering algorithms.

Method suitable for real-time fire detection in surveillance systems.

Abstract

This paper proposes a vision-based fire and smoke segmentation system which use spatial, temporal and motion information to extract the desired regions from the video frames. The fusion of information is done using multiple features such as optical flow, divergence and intensity values. These features extracted from the images are used to segment the pixels into different classes in an unsupervised way. A comparative analysis is done by using multiple clustering algorithms for segmentation. Here the Markov Random Field performs more accurately than other segmentation algorithms since it characterizes the spatial interactions of pixels using a finite number of parameters. It builds a probabilistic image model that selects the most likely labeling using the maximum a posteriori (MAP) estimation. This unsupervised approach is tested on various images and achieves a frame-wise fire detection rate of 95.39%. Hence this method can be used for early detection of fire in real-time and it can be incorporated into an indoor or outdoor surveillance system.