Application of neural network to analyses of CCD colour TV-camera image for the detection of car fires in expressway tunnels

This study aims to investigate the effectiveness of the early detection of a car fire occurring in an expressway tunnel using surveillance cameras. Road and rail tunnels need to be built in mountainous areas, in order to provide the effective distribution of goods. A fire in a tunnel can endanger human life as the heat flow and smoke increase rapidly in such a long enclosed space. In addition, it can significantly affect social economies owing to a stagnation of the distribution and the breakdown of the distribution network.