Fast Pitch Detection Based on Improved FCM Algorithm and Hough Transform

Football videos always appeals to large audiences in recent years. The pitch plays the fundamental role in automatically analyzing the factors in the game. In this paper, a fast pitch detection method based on an improved FCM algorithm and Hough transform is proposed. Firstly, use WFCM clustering algorithm for image processing. Secondly, determine the boundary by Hough transform, and then fill the region within the boundaries to get the mask. Hough transform is rarely used before, since it is difficult to detect the boundary from the many lines obtained by Hough transform. There are two reasons why we choose this method: the first one is that Hough transform is mature and fast, secondly, a good segmentation result can be obtained after WFCM and corrosion processing which can reduce the number of the detected lines. Experimental results show that the proposed method is robust to various football videos.