文献类型：会议论文

英文题名：A Real-time Tracking System Used for Surface Detection

作者：Wang, Xiao[1] Li, Bo[1] Zhang, Zhankuan[1]

第一作者：王霄

机构：[1] Research Institute of Wood Industry, Chinese Academy of Forestry, Beijing, China

会议论文集：Proceedings - 2020 7th International Conference on Information Science and Control Engineering, ICISCE 2020

会议日期：December 18, 2020 - December 20, 2020

会议地点：Changsha, Hunan, China

语种：英文

外文关键词：Cameras - Cams - Motion analysis - Tracking (position)

年份：2020

摘要：In industrial production, the high speed of object on conveyor always causes motion blur when computer vision is used to achieve appearance inspection, results in difficulties in image processing. In this study, a real-time tracking system based on cam transmission was applied to eliminate the relative movement between camera and objects. The outline of the cam was specifically designed to meet the requirement of real-time tracking and simplified the motor control. To test the system, the velocities of both camera and object were collected using rotary encoders. It was found that the camera velocity would be almost equal to that of object in a certain period when cam rotated uniformly. Also, the images were captured by triggering the camera at the suitable moment, and it was shown that the object in image was very clear. Compared with conventional restoration algorithms, the proposed method resulted in clearer images without any distortion, and could save computation time, which was especially suitable for on-line inspection in industrial production such as boards. ? 2020 IEEE.