基于图像处理的舰船火灾烟雾检测技术研究

摘要

Warship fire detection and alarm system is the key component of damage control monitoring system, its performance advantages and disadvantages will directly affect the safety of warship and crew. In order to improve the computational efficiency and accuracy of warship fire detection technology, and reduce the rates of false positives and false negatives, etc. A video based fire smoke detection technology is proposed in this paper, which is based on the color features and the dynamics features of smoke. The technology detects the regions of moving targets in video images, and extracts the features associated with the color, the irregularity of the shape and the diffusion property of smoke. After analysis and judgment, a target area that satisfies the smoke feature criteria can be judged as a smoke area. Test results show that the technology can accurately and quickly detect fire smoke with a low false alarm rate and high reliability.%舰船火灾探测报警系统作为损管监测体系的关键构成部分，其性能的优劣将会直接影响到舰船和舰员的安全问题。为提高舰船火灾探测技术的实时性和准确性，减少误报、漏报等情况的发生，提出一种结合烟雾颜色特征与运动特征进行研究的视频火灾烟雾检测技术。该技术通过运动目标区域检测和对烟雾颜色特征、形状不规则性和扩散性特征的提取，经过分析与判断，将满足烟雾特征判据的目标区域判别为烟雾区域。检测结果表明该技术能够准确且迅速地探测火灾烟雾，误报率低，可靠性好。