Research on Automatic Fire Monitoring and Recognition Technology Based on Digital Image Processing Technology

摘要

With the rapid development of China's industrialization and urbanization, various large-scale buildings are also emerging, and fire hazards are also increasing. Compared with the traditional fire detection technology, the visual fire detection technology based on digital image processing has the unparalleled superiority of quick response, rich information, and no constraints from large space environment, which has important social and economic value. Aiming at the main problems faced by digital image processing fire detection technology, this paper proposes an idea and method for automatic detection and recognition of flame surveillance video images. The original surveillance video image collected on-site by the visual monitoring system is the main processing object, and digital image processing technology is the main technical means. The pre-processing of surveillance video images, the segmentation of surveillance video images and the detection of suspicious flame target regions, and the analysis and extraction of feature parameters of flame image segmentation regions enable automatic detection and recognition of flame visual images.