Method of Information Extraction Based on Infrared Target Recognition
a number of infrared target indicators are set according to industrial control field, use the sensor information as the drive signal of infrared indicators after the strength or frequency modulation, get on-site monitoring images that contained infrared targets by using infrared cameras, and propose a new method of using surveillance image to extract the sensor information from background. After local image analysis on the changes of the average gray value and the local gray mean contrast, the method is an infrared target enhancement method using differential treatment image. On this basis, the maximum possible target locations is determined as seed point by searching the local gray level maxima in the enhanced image, The seeded region growing algorithm is utilized to realize the separation of the target and the background . The sensor information is extracted from the background finally by the frequency and intensity of target appears in the image sequence. Experimental results on power system equipment running environment show its effectiveness.
Image segmentation Infrared target recognition Average gray value Gray mean contrast Seeded region growing
Chen Jianming Xu Shiqing
North China Institute of Water Conservancy and Hydroelectric Power Zhengzhou, China Nanyang Power Supply Company Nanyang, China
国际会议
成都
英文
1766-1769
2010-12-17（万方平台首次上网日期，不代表论文的发表时间）