【摘要】 图像边缘检测可以作为危化品堆垛火灾检测的技术手段,传统Canny和部分改进Canny算法以及常规的像素多特征检测算法在有效检测火焰区域边缘信息的同时,存在去除噪声干扰和背景干扰能力较差的问题,不利于后续的基于火焰特征信息判断危化品堆垛火情状态的准确性。该文采用多种算法融合的火灾检测算法,将像素RGB&HIS颜色信息、相与去噪、otsu前景提取、二值图边缘提取方法等算法有机结合,提取火灾火焰边缘,减少干扰。实验表明,该文算法的火焰检测正确率达到93.5%。更多还原

【Abstract】 Image edge detection can be used as a technical means of dangerous chemicals stacking fire detection.Traditional Canny algorithm,partially improved Canny algorithm and conventional pixel multi feature detection algorithm can effectively detect the edge information of the flame area,but at the same time,the ability of removing noise and background interference is poor,which is not conducive to the subsequent accurate judgment of dangerous chemicals stacking fire state based on the flame feature information accuracy. In this paper,a fire detection algorithm based on multi algorithm fusion is adopted,which combines pixel RGB & HIS color information,phase and denoising,otsu foreground extraction,binary image edge extraction and other algorithms to extract fire flame edge and reduce interference. The experimental results show that the correct rate of flame detection is 93.5%.更多还原