Benchmarking of wildland fire colour segmentation algorithms. Issue 12 (1st December 2015)

• Record Type: Journal Article
• Authors: Toulouse, Tom; Rossi, Lucile; Akhloufi, Moulay; Celik, Turgay; Maldague, Xavier
• Is Part Of: IET image processing. Volume 9:Issue 12(2015)
• Page Start: 1064
• Page End: 1072
• Publication Date: 2015-12-01
• Subjects: image colour analysis -- image segmentation -- fires -- image sensors -- image sequences -- probability -- object detection; wildland fire colour segmentation algorithm benchmarking -- computer vision‐based methods -- sensor‐based fire detection technologies -- visible band image sequences -- suspicious fire events -- outdoor environments -- indoor environments -- probabilistic fire segmentation algorithm -- wildland fire images; Image processing -- Periodicals; 621.36705
• DOI: 10.1049/iet-ipr.2014.0935
• Languages: English
• ISSNs: 1751-9659

Abstract

Abstract : Recently, computer vision‐based methods have started to replace conventional sensor‐based fire detection technologies. In general, visible band image sequences are used to automatically detect suspicious fire events in indoor or outdoor environments. There are several methods which aim to achieve automatic fire detection on visible band images, however, it is difficult to identify which method is the best performing as there is no fire image dataset which can be used to test the different methods. This study presents a benchmarking of state of the art wildland fire colour segmentation algorithms using a new fire dataset introduced for the first time. The dataset contains images of wildland fire in different contexts (fuel, background, luminosity, smoke etc.). All images of the dataset are characterised according to the principal colour of the fire, the luminosity, and the presence of smoke in the fire area. With this characterisation, it has been possible to determine on which kind of images each algorithm is efficient. Also a new probabilistic fire segmentation algorithm is introduced and compared to the other techniques. Benchmarking is performed in order to assess performances of 12 algorithms that can be used for the segmentation of wildland fire images.