A Method for Detecting Lines on Soccer Field by Color of Grass Variation

摘要

The line detection algorithm in a soccer field image is an important step in order to detect an offside situation. In this paper, a method of line detection in a soccer field is proposed by using color of grass variation. This method consists of three steps. Firstly, the soccer ground is segmented by a constraint color in RGB channel. Second, the edge in the image is detected. Finally, a line is detected by a line hough transform technique. To evaluate the performance of the proposed method, 50,000 image form 22 matches are used to detecting lines. Results is compared with ground truth data. The proposed method has shown a satisfactory result.