Vision-based ball recognition for soccer robots without color classification

Recognizing the FIFA ball without color classification is a significant ability for robocup middle size league soccer robots to play competition without the constraint of current color-coded environment. This paper describes a novel method to recognize the ball based on omni-directional vision system and perspective camera without color classification. Firstly the imaging character of the omni-directional vision system is analyzed and the conclusion that the ball on the field can be imaged to be ellipse approximately is derived. Then the arbitrary FIFA ball can be recognized by detecting the ellipse imaged by the ball according to the derivation, and an estimating algorithm for ball speed is integrated to track the ball. For making up the deficiency of omni-directional vision, a perspective camera system is also added to recognize the arbitrary ball near and in front of robot by Sobel filter and Hough transform algorithm. The experiment results show that the method can recognize the arbitrary FIFA ball effectively and in real-time even in cluttered environments.