A study of camera calibration in the vision system of soccer robots

Abstract

The vision system is the key to the process of soccer robot competitions. While the key to the vision system is the camera which offers all the information related to the environment in the competition process. Therefore how to ensure the truth and accuracy of the pictures taken in by the camera is essential to the vision system of soccer robots. Based on the competition rules of RoboCup Small Size League, we corrected the errors caused by the camera's not being perpendicular to the ground or by the camera's not being right above the ground, as well as the serious barrel-shaped distortion because of the enlargement of the angle of view of the video camera. It turned out that the vision system corrected had been improved greatly in terms of recognition accuracy and stability.