Title:	RECOGNITION OF FIRE/SMOKE BASED ON COMPUTER-VISION IMAGE SEGMENTATION
Author(s):	Marie Providence Umugwaneza, Jean Paul Dukuzumuremyi, Bei-Ji Zou, Lei Wang, Yixiong Liang
ISBN:	978-972-8939-48-9
Editors:	Yingcai Xiao
Year:	2011
Edition:	Single
Keywords:	Computer vision, fire recognition, neural network, k-means, wavelet transform, discreet cosine transform.
Type:	Full Paper
First Page:	60
Last Page:	66
Language:	English
Cover:
Full Contents:	click to dowload
Paper Abstract:	Dangers of fire/smoke are recently serious whereas scientific research to fight them seems to be left behind. The newest innovations use cameras and computer algorithms to analyze the visible effects of fire and its motion in their applications. Their approaches present some drawbacks when working in spatial domain. The main difficulty is to identify objects if they do not occur at the expected position. In this paper, we present a fast and robust algorithm for recognition fire/smoke in a cluttered scene from a pair of cameras. The input images are first segmented according to a pre-determined disparity threshold map and the real-time disparities of the fire. Binary image processing techniques are used to reject noise introduced into the segmented images through low-resolution disparity calculations which consequently can lead to the gain of clearer results. In order to reduce the false alarms, a new segmentation method used in this approach shows that stereo vision segmentation is more accurate than the color-based method based on k-means segmentation for the overall recognition; more specifically for images taken in night. The wavelet and discrete cosine transforms are then used for image feature extraction for a neural network classifier hence the system could generate a warning in case the fire/smoke is recognized.
	Go Back