论文部分内容阅读
A vision-based scheme for object recognition and transport with a mobile robot is proposed in this paper.First,camera calibration is experimentally performed with Zhenyou Zhangs method,and a distance measurement method with the monocular camera is presented and tested.Second,Kalman filtering algorithm is used to predict the movement of a target with HSI model as the input and the seed filling algorithm as the image segmentation approach.Finally,the motion control of the pan-tilt camera and mobile robot is designed to fulfill the tracking and transport task.The experiment results demonstrate the robust object recognition and fast tracking capabilities of the proposed scheme.