透過您的圖書館登入
IP:216.73.216.100
  • 學位論文

基於三維深度之手勢手指互動系統與晶片設計

Three-Dimensional Finger and Hand Gesture Interaction System and Chip Design

指導教授 : 范育成
若您是本文的作者,可授權文章由華藝線上圖書館中協助推廣。

摘要


本論文提出一個基於三維深度之手勢手指互動系統與晶片設計,本系統使用較低成本的雙攝影機來計算深度影像,並且根據使用者手勢與手指辨識使用者控制狀態,提供使用者進行即時的手勢操控。 本論文提出的三維深度之手勢手指互動系統介面為單手操作,判別兩種特定手勢:單指與握拳手勢。當擺出單指手勢下,可以進行圖片的方位移動功能與畫面按鈕點擊功能,本系統單指手勢預設為判別八方向位移,並且圖片依方向移動,當單指移至按鈕進行點擊動作,將觸發按鈕功能。當擺出握拳手勢下,可以進行圖片的縮小與放大功能。經過多位使用者測試,八方向移動功能準確率為93%,縮放功能準確率為93.75%,點擊功能準確率達90%。 最後我們將本系統進行硬體架構設計,以數位積體電路設計流程實作成數位晶片,將左、右視角的彩色影像輸入後,經過晶片處理膚色轉換、形態學、深度運算後,輸出一張具有深度值的膚色遮罩圖,做為追蹤與手勢判斷的依據。

並列摘要


In this thesis, we proposed a Three-Dimensional Finger and Hand Gesture Interaction System and Chip Design. We use stereo camera to capture stereo image and calculate disparity map and depth map for supplying user a real-time interface based on hand gesture and finger action. Our three-dimension finger and hand gesture interaction interface is for one hand control, this system can distinguish two hand gesture types, one finger gesture and stone gesture. When the system is under one finger gesture mode, user can move hand to move display image by 8 directions, also can execute clicking function. When the system is under stone gesture mode, user can move hand to zoom the display image. Our accuracy of 8 direction function is 93%, accuracy of zoom function is 93.75% and accuracy of click function is 90%. We design the hardware architecture using digital chip flow to make real digital chip. This chip adopts left and right color image data and operates using skin color transform, morphology and depth map information. Finally, chip outputs a skin mask with depth value image for hand tracking.

參考文獻


[4] A. Bellarbi, S. Benbelkacem, N. Zenati-Henda, and M. Belhocine, ”Hand Gesture Interaction Using Color-based Method for Tabletop Interfaces,” in IEEE International Symposium on Intelligent Signal Processing (WISP), 2011, pp.1-6.
[6] L. Liu, N. Sang, S. Yang, and R. Huang, “Real-Time Skin Color Detection under Rapidly Changing Illumination Conditions,” IEEE Transactions on Consumer Electronics, vol. 57, no. 3, pp. 1295-1302, Aug. 2011.
[7] J. Geng, Z. Miao, and C. Zhong, “Skin Detection with Illumination Adaptation in Single Image” in IEEE International Conference on Multimedia and Expo (ICME), 2011, pp. 1-6.
[11] Qing Chen, Nicolas D. Georganas, and Emil M. Petriu, “Hand Gesture Recognition Using Haar-Like Features and a Stochastic Context-Free Grammar,” IEEE Transactions on Instrumentation and Measurement, vol. 57, No. 8, pp. 1562-1571, August, 2008.
[14] S. Bilal, Rini Akmeliawati, Momoh Jimoh El Salami, Amir A. Shafie, and El Mehdi Bouhabba, “A Hybrid Method Using Haar-like and Skin-color Algorithm for Hand Posture Detection, Recognition and Tracking,” in IEEE International Conference on Mechatronics and Automation (ICMA), 2010, pp. 934-939.

延伸閱讀