Image-based smoke detection using feature mapping and discrimination
- PDF / 924,266 Bytes
- 10 Pages / 595.276 x 790.866 pts Page_size
- 28 Downloads / 196 Views
(0123456789().,-volV)(0123456789(). ,- volV)
METHODOLOGIES AND APPLICATION
Image-based smoke detection using feature mapping and discrimination Norah Asiri1 • Ouiem Bchir1 • Mohamed Maher Ben Ismail1 • Mohammed Zakariah1 • Yousef A. Alotaibi1
Ó Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract Typically, image-based smoke detection is formulated as a frame classification task that aims to automatically assign the captured frames to the predefined ‘‘smoke’’ or ‘‘smoke-free’’ classes. This classification is based on the visual content of the images. In other words, the keystone of such a solution is the choice of the visual descriptor(s) used to encode the visual characteristics of the smoke into numerical vectors. In this paper, we propose to learn a new feature space to represent the visual descriptors extracted from the video frames in an unsupervised manner. This mapping is intended to yield better discrimination between smoke-free images and those showing smoke patterns. The proposed approach is inspired by the linear hyperspectral unmixing techniques. It defines the axes of the new feature space as the vertices of a minimum-volume simplex enclosing all image pixels in the frame. The obtained empirical results prove that the proposed feature mapping approach reinforces the discrimination power of the visual descriptors and produces better smoke detection performance. In addition, the proposed approach exhibits the valuable ability to automatically determine the most relevant visual descriptors. Keywords Pattern recognition Image understanding Smoke detection
1 Introduction Fire detection is an important alarm technique that aims to prevent losses caused by fire disaster. Image processing has been adopted by various fire alarm methods (Chen et al. 2004) and proved to be effective by achieving low false alarm rate. Such image-based solutions rely on the assumption that flame patterns appear in the captured scene, and can be encoded using typical image descriptors or features. This assumption delays the detection time and may yield unrecoverable losses, especially if the fire occurs outdoors, such as in forests or open areas. In particular, waiting until flames appear in the fire regions may be a risky strategy. Infrared cameras have been proposed as an alternative (Chen et al. 2004) to overcome this limitation.
Communicated by V. Loia. & Mohamed Maher Ben Ismail [email protected] 1
Computer Science Department, College of Computer and Information Sciences, King Saud University, Riyadh, Saudi Arabia
However, their performance is considerably constrained by the potential presence of objects that occlude the flames. Given that the smoke usually precedes the flame appearance, automatic smoke detection emerged as a natural solution for fire mishaps. Conventional smoke detectors use photoelectric and ionization devices that sense the reduction in the amounts of ionized air molecules (Toreyin et al. 2005). The widespread use of these sensors can be attributed to their usability and afford
Data Loading...
