论文部分内容阅读
FIRA MiroSot机器人足球比赛中,对足球机器人位姿的识别对最终决策起着极为重要的作用,视觉系统的识别速度、精度直接影响到比赛的胜负。本文提出了一种基于特征区域补偿的位姿快速校正方法,以弥补由于光源、场地等条件造成的识别误差。实验结果表明:该设计方案能够提高比赛中的识别精度,并具有良好的适应性。
In the FIRA MiroSot robot soccer match, recognizing the posture of a soccer robot plays a very important role in the final decision-making. The recognition speed and accuracy of the vision system directly affect the outcome of the match. In this paper, a fast pose correction method based on feature area compensation is proposed to compensate for the recognition error caused by light source, site and other conditions. The experimental results show that this design scheme can improve the recognition accuracy in the competition and has good adaptability.