以嵌入式系統為基礎之電動輪椅人眼偵測及控制

本論文針對智慧型輪椅開發一個即時人眼辨識及控制系統。本論文提出一個針對智慧型輪椅的控制及自動充電方法。在輪椅控制方面，基於電動輪椅本身硬體架構上，利用Raspberry Pi為開發平台，對輪椅控制器以控制輪椅的電壓以達到輪椅控制的目的。利用Raspberry Pi的硬體架構，透過擷取影像畫面，系統可辨識人眼特徵及偵測雙眼的位置以控制輪椅的行進方向。除此之外，本系統利用Haar-like 特徵演算法來偵測人眼位置。不需要透過尺度不變特徵轉換或膚色來辨識人臉，大幅的減少在嵌入式系統執行複雜演算法的效能。由於Haar-like 特徵演算法並非採用色彩分析辨識特徵，本系統在前置處理部分又應用了直方圖等化，所以可減少光線不足或光線不平均等對於影像辨識的影響。本論文在影像前置處理的部分加入了直方圖等化來降低因光線不足或光線不平均等對於影像辨識的影響。最後，透過實驗的結果可驗證我們提出的控制方法和架構是有效及可行的。

關鍵字

哈爾特徵；人眼辨識；智慧型輪椅

並列摘要

Based on Raspberry Pi embedded platform and camera module, a real-time eyes detection control system for power wheelchair is implemented in this thesis. By utilizing the primal hardware of Raspberry Pi, the eyes feature and position can be recognized from the captured image to control the direction of wheelchair. In addition, Haar-like feature recognition algorithm is adopted in this system to detect the position of the eyes. Since the Scale-invariant feature transformation or skin color recognition is not taken into consideration, the burden of embedded platform can be decreased significantly. In addition, this system utilizes histogram equalization in the pre-processing part; therefore, the affections are reduced for insufficient light and different brightness environments. Lastly, the experiment results show the effectiveness and validity of the developed eye detection control system and the proposed control approach.

並列關鍵字

Haar-like Feature ； Eyes Recognition ； Intelligent Wheelchair

參考文獻

[10] P. Viola and M. Jones, "Robust Real-time Object Detection," ICCV 2nd International Workshop on Statistical and Computational Theories of Vision-Modeling, Learning, Computing and Sampling, July 2001.

[11] Papageorgiou, Oren and Poggio, "A General Framework for Object Detection ", International Conference on Computer Vision, 1998.

[12] R. Lienhart and J. Maydt, “An Extended Set of Haar-like Features for Rapid Object Detection,” International Conference on Image Processing, 2002, pp. 900-903.

[14] Y. Fruend and R. Schapire, "A Decision-Theoretic Generalization of On-Line Learning and an Application to Boosting," Journal of Computer and System Sciences, vol.55, 1997, pp.119-139.

[15] P. Viola and M. J. Jones, "Robust Real-Time Face Detection," International Journal of Computer Vision, 2004, pp. 137–154.

國際替代計量

以嵌入式系統為基礎之電動輪椅人眼偵測及控制

全文下載

主題瀏覽