Real-time detection of specific shaped target in infrared video

The technology of target tracking in infrared video has a quite extensive future of application. In this paper, we introduce a real-time algorithm of specific shape target detection in infrared video, using single circle target as an example. The primary idea of this algorithm is to set several scan lines in each frame of the video. Then we use the intersections of these lines and the edge of the target to calculate the parameters of the target. We also improve the randomized Hough transform algorithm and use it to calculate the parameters of the target via the parameters of the intersections. The experiments indicate that the algorithm has favorable veracity and high velocity and is capable of meeting the requirement of practical application.