Robust and Accurate Detection of Object Orientation and ID without Color Segmentation

Abstract

To give optimal visual-feedback, which helps to control a robot, it is important to make its vision system more robust and accurate. In the RoboCup Small Sized League (F180), a global vision system that is robust to unknown and varying lighting conditions is especially important. The vision system, which is in common use, processes an image that identifies and locates robots and the ball. For low-level vision, the color segmentation library called CMVision (J. After color is segmented, objects are identified based on the color segmentation results, and then the robot's pose is estimated. To improve the vison system's robustness to varying light conditions, color In this chapter, we describe a robust and accurate pattern matching method for simultaneously identifying robots and estimating their orientations that does not use color segmentation. To search for similar patterns, our approach uses continuous DP matching, which is obtained by scanning at a constant radius from the center of the robot. The DP similarity value is used to identify object, and to obtain the optimal route by back tracing to estimate its orientation. We found that our system's ability to identify objects was robust to variation in light conditions. This is because it can take advantage of the changes in intensity only. Related work and our approach are described in section 2. Section 3 describes the method for robust and accurate object identification. The experimental results are presented in section 4. Section 5 discusses some of the advantages of the proposed method. Finally, section 6 concludes the chapter.