以影像技術為基礎之遙控機器人操作介面之設計

Design of a vision-based human-machine interface device for a remote control robot

黃健誠(Chien-Cheng Huang)

指導教授 ： 何明果

大同大學/工程學院/機械工程學系所/博士(2012年)

摘要

在本論文中我們把CCD擷取進來的影像利用特徵向量法抓取特徵物的顏色經由影像辨識法則的處理之後把結果傳輸給遠端的伺服馬達輸出相對應的動作，達到人機互動的目的。本論文中所採用影像追蹤的方法是基於特徵法，但與特徵法不同的是，在取得目標物的方法；在取樣前先針對目標物進行取得適當的RGB特徵向量，也就是說如符合此特徵向量的區塊即為所求之目標。而與一般的特徵法比較之下多花費了0.014秒，但是出來的結果可見其可信度增加與後續處理的時間減低許多，更符合往後研究上所要應用的設計。

關鍵字

介面 ； 機器人 ； 視覺

並列摘要

In this thesis, we use eigenvector method to capture the color of special object on the image retrieved by CCD. After processing with image recognition rules, the result is transmitted to remote servo motor to produce respective movement, and the goal of human machine interactive is achieved.
The image tracking method used in this thesis is based on characteristic method; however, the method to obtain an object is different from characteristic method. Before sampling, the focus was to obtain appropriate RGB eigenvector. That is, if colored tag confirms with eigenvector, it is the target needed. Comparing with ordinary characteristic method, this method takes 0.0014 seconds more. But its credibility is increased and processing time needed afterward is reduced. This meets the design application for future research.

並列關鍵字

robot ； vision ； interface

參考文獻

9. Hamdoun, O., Moutarde, F., Stanciulescu, B., and Steux, B. Person re-identification in multi-camera system by signature based on interest point descriptors, Proceedings of the 2nd ACM/IEEE International

10. Lim, S. N., Elgammal, A., and Davis, L. S. Imagebased pan-tilt camera control in multi-camera surveillance environment collected on short video sequences, In. Proceedings of the IEEE ICME 2003,

12. Zana, Y. and Cesar, R. Face recognition based on polar frequency features. ACM Trans.Appl. Percept., 2006, 3(1January 2006, pp. 62–82.

13. Huang, H.P. and Lin, C.T. Multi-CAMSHIFT for Multi-View Faces Tracking and Recognition. ROBIO, Proceedings of the IEEE International Conference on Robotics and Biomimetics, Kunming,

15. Comaniciu, D. and Meer, P. Mean shift: A robust approach toward feature space analysis. IEEE Trans. Pattern Anal. Mach. Intell., 2002, 24(5), 603–619.