Optimal Fire Detection Using Image Processing

Abstract

Flame detection and classification of flames is challenging work in the field of image processing. Use of image processing instead of sensors helps detection of fire in more accurate and quick manner. Thus this paper introduces a optimal approach for detection of fire using image processing technology which enables fire detection at the earliest in order to reduce life and property loss. The proposed system uses color segmentation model for detecting fire with reference to HSV and YCbCr color models. Also, density of fire growth is detected by the frame difference technique.