图像处理技术在船舶火灾探测系统中的运用

摘要

研究基于图像处理的船舶火灾探测系统，其主要用颜色模型算法对采集到的实时图像进行分割，提取目标区域的图像特征，把图像特征作为分类器的输入，输出为1则表示有火灾发生，发出报警信号，输出为2则表示无火灾，继续监控。通过该系统可以对初期火灾进行探测。%A ship fire detection system based on image processing is studied. In the system,the color model algorithm is used to segment the acquired real⁃time image,and extract the image features of target area. And then the image features are taken as the inputs of the classifier. If the output of the classifier is 1,it means there is a fire incident,and the system will send the alarming signal. If the output of the classifier is 2,it means there is no fire incident,and the system will monitor unceasingly. The system can detect the early fire.