即時物體追蹤之立體視覺導引自走車

A Stereo Vision-Guided Mobile Robot for Real-Time Object Tracking

傅培耕(Pei-Ken Fu)

指導教授 ： 張永鵬

中原大學/工學院/機械工程研究所/碩士(2004年)

https://doi.org/10.6840/cycu200400490

摘要

本文的研究目的在設計並實現ㄧ部具有立體視覺的影像伺服控制之自走車。傳統的影像伺服使用單眼視覺，無法得知目標物與攝影機的距離，本文採用雙眼立體視覺架構，可以求出目標物與攝影機的距離，使自走車得以追蹤移動目標物。首先，雙CCD攝影機連續擷取成對影像，經過影像處理找出符合顏色特徵的部份，再經影像分割與面積過濾後框選出目標物，藉由立體視覺演算法計算目標物與自走車的相對距離，實驗結果顯示自走車能夠即時追蹤目標物。本文設計之立體視覺導引自走車，結合機械視覺與自走車運動控制，可以即時追蹤特定顏色目標物，賦予自走車具有視覺感知的能力。

關鍵字

視覺感知 ； 自走車 ； 立體視覺 ； 影像伺服

並列摘要

In this dissertation, our research goal is to design and realize the visual servoing control of a mobile robot using the stereo vision. As opposed to conventional visual servoing techniques, which often used monocular vision can not estimate the distance between a target object and a camera. In this dissertation, the mobile robot can track a moving target object because we extract the 3-D distance information of the target object using the binocular stereo vision. First, the pair images acquisition are accomplished by two CCD cameras continuously, and then we can extract the color characteristic of the pair images using image processing methods. Second, we draw a box around the target object in the binary pair images after using image segmentation and size filtering. Moreover, the 3-D distance information between the target object and the mobile robot can be calculated by stereo vision algorithm. Finally, the experiments described in the dissertation have shown that the mobile robot carried out for tracking the target object real-time. In this dissertation, we have designed a stereo vision-guided mobile robot, which combines machine vision and motion control, for tracking a target object with color characteristic real-time. In fact, we give the mobile robot ability in visual perception.

並列關鍵字

stereo vision ； mobile robot ； visual perception ； visual servoing

參考文獻

[3] Horn, K.P. and Schunck, B.G., “Determining Optical Flow,” Artificial Intelligence, Vol. 17, pp.185-203, 1981.

[7] S.T. Barnard and W.B. Thompson, “Disparity analysis of images,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 2, pp. 330-340, 1980.

[9] A.P. Tirumalai, B.G. Schunck, and R.C. Jain, “Dynamic stereo with self-calibration,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 14, pp. 1184-1189, 1992.

[11] Wei Yun Yau and Han Wang, “Fast relative depth computation for an active stereo vision system,” Real-Time Imaging, Vol. 5, pp. 189-202, 1999.

[12] 陳冠宇，張永鵬，傅培耕，即時影像追蹤系統之設計與實現，中原學報，Vol. 31, No. 1, pp. 21-29, 2003.