Vision Based Flame Detection System For Surveillance Camera

Flame detection systems are useful for the development of fire early detection systems. In general, conventional fire detection system which is usually used heat detector cannot detect the presence of fire since the detector requires much time to detect temperature changes. The use of a camera as an image scanner sensor has the potential to be used as an identifier of a flame shape or color. This study aims to build an image processing flame detection system using image enhancement, segmentation and filtering methods which is applied the Vission Assistant available in LabView.The system implemented in a frame rate of 30 fps with a resolution of 1024 x 768, in which the system accuracy reach up to 98%.