研究生: |
林克臻 Ke-jen Lin |
---|---|
論文名稱: |
高可靠度划拳手指數目視覺偵測系統研發 Development of High Reliability Image-based Finger Counting System for Finger Gaming |
指導教授: |
林其禹
Chyi-yeu Lin |
口試委員: |
范欽雄
Chin-shyurng Fahn 徐繼聖 Gee-sern Hsu |
學位類別: |
碩士 Master |
系所名稱: |
工程學院 - 機械工程系 Department of Mechanical Engineering |
論文出版年: | 2013 |
畢業學年度: | 101 |
語文別: | 中文 |
論文頁數: | 65 |
中文關鍵詞: | 划拳機器人 、手指數目計算 、Kinect感測器 、指尖角度計算法 |
外文關鍵詞: | finger gaming robot, finger counting, Kinect sensor, fingertip-angle computation method |
相關次數: | 點閱:201 下載:5 |
分享至: |
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報 |
本研究旨在開發出一套供划拳機器人計算玩家手指數目的高可靠度影像辨識系統。該影像辨識系統採用Kinect感測器,並結合指尖偵測程式,用以即時辨識出玩家所出的拳(即計算手指數目)。
影像辨識的策略,首先利用Kinect感測器所擷取的場景深度資訊濾除玩家手部後方的所有影像以節省分析時間。接著搜尋手部輪廓並運用指尖角度計算法依序偵測出輪廓序列中的指尖點,以便合計出手指數目。
為建立高可靠度的辨識系統,將Kinect感測器裝設在9個可能的裝配位置,並分別針對玩家的6個出拳位置和每個位置上的5個不同手掌角度,測試8種台灣划拳慣用手勢。最後,將所得測試結果彙整建立成數據庫後,交叉統計並探討各種情況下的最佳辨識率,進而找到系統在使用兩台Kinect感測器下的最佳辨識組合,其辨識率超過94%。
This research aims to develop a high reliability image-based system for finger gaming robot to count the fingers of the competitor. The image-based system includes the use of Kinect sensors and fingertip detection program so as to identify the exposed finger numbers of the competitor in the game real time.
The image recognition strategy starts with the use of Kinect sensor to screen out the image part behind the hand using depth information to save time. Afterwards, the hand contour will be searched and then detect all fingertips using fingertip-angle computation method so as to make a finger count.
In order to generate a high reliable recognition system, Kinect sensors are installed in 9 different locations. For each location the competitor's hand will present in 6 different positions, and for each hand position the hand will appear in 5 different angles. Finally, for each combination, 8 common finger postures are shown for recognition testing. After the large set of experiment data went through an extensive analysis process, the overall best combination with two Kinect sets is revealed, with a 94%+ correct recognition performance.
[1] C. Graetzel, T. Fong, S. Grange, and C. Baur, "A Non-contact Mouse for Surgeon-Computer Interaction," THC, 12(3):245-257, 2004.
[2] R. Guo and J. Quarles, "Exercise-based Interaction Techniques for a Virtual Reality Car Racing Game," IEEE VRW, 93-94, 2012.
[3] M. Van den Bergh, D. Carton, R. De Nijs, N. Mitsou, C. Landsiedel, K. Kuehnlenz, D. Wollherr, L. Van Gool, and M. Buss, "Real-Time 3D Hand Gesture Interaction with a Robot for Understanding Directions from Humans," IEEE RO-MAN, 357-362, 2011.
[4] J. Shotton, A. Fitzgibbon, M. Cook, T. Sharp, M. Finocchio, R. Moore, A. Kipman, and A. Blake, "Real-Time Human Pose Recognition in Parts from Single Depth Images," IEEE CVPR, 1297-1304, 2011.
[5] M. Tang, "Recognizing Hand Gestures with Microsoft's Kinect," Technical Report of Department of Electrical Engineering, Stanford University, 2011.
[6] V. Frati and D. Prattichizzo, "Using Kinect for Hand Tracking and Rendering in Wearable Haptics," IEEE WHC, 317-321, 2011.
[7] J. L. Raheja, A. Chaudhary, and K. Singal, "Tracking of Fingertips and Centres of Palm Using Kinect," IEEE CIMSiM, 248-252, 2011.
[8] C. Keskin, F. Kırac, Y. E. Kara, and L. Akarun, "Real Time Hand Pose Estimation Using Depth Sensors," IEEE ICCVW, 1228-1234, 2011.
[9] Z. Ren, J. Yuan, and Z. Zhang, "Robust Hand Gesture Recognition Based on Finger-Earth Mover's Distance with a Commodity Depth Camera," ACM, 1093-1096, 2011.
[10] Y. Li, "Multi-Scenario Gesture Recognition Using Kinect," IEEE CGAMES, 126-130, 2012.
[11] Y. Wen, C. Hu, G. Yu, and C. Wang, "A Robust Method of Detecting Hand Gestures Using Depth Sensors," IEEE HAVE, 72-77, 2012.
[12] 陳威詔,「基於前臂輪廓的即時掌心追蹤以及手勢判斷」,台灣科技大學碩士論文,2011。
[13] K. Khoshelham and S. O. Elberink, "Accuracy and Resolution of Kinect Depth Data for Indoor Mapping Applications," Sensors, 12:1437-1454, 2012.
[14] OpenNI. Available online: http://openni.org/ (accessed on 12 December 2012)
[15] PrimeSense. Available online: http://www.primesense.com/ (accessed on 14 December 2012)
[16] A. Maimone and H. Fuchs, "Reducing Interference between Multiple Structured Light Depth Sensors Using Motion," IEEE VR, 51-54, 2012.
[17] T. Dutta, "Evaluation of the KinectTM Sensor for 3-D Kinematic Measurement in the Workplace," AE, 43:645-649, 2012.
[18] J. Smisek, M. Jancosek, and T. Pajdla, "3D with Kinect," IEEE ICCVW, 1154-1160, 2011.
[19] X. Wen and Y. Niu, "A Method for Hand Gesture Recognition Based on Morphology and Fingertip-Angle," IEEE ICCAE, 1:688-691, 2010.
[20] OpenCV. Available online: http://opencv.org/ (accessed on 12 December 2012)
[21] R. C. Gonzalez and R. E. Woods, "Digital Image Processing," 2nd ed., Hong Kong: Prentice-Hall, 2002.
[22] M. Sonka, V. Hlavac, and R. Boyle, "Image Processing, Analysis, and Machine Vision," 3rd ed., Singapore: Thomson, 2008.
[23] S. Marchand-Maillet and Y. M. Sharaiha, "Binary Digital Image Processing," 1st ed., Tokyo: Academic Press, 2000.
[24] G. Bradski and A. Kaehler, "Learning OpenCV," 1st ed., Taipei: O'Reilly, 2008.
[25] R. Laganiere, "OpenCV 2 Computer Vision Application Programming Cookbook," 1st ed., Birmingham: Packt, 2011.
[26] 鍾國亮,「影像處理與電腦視覺」,第4版,台北:東華,2008。
[27] 劉瑞禎和于仕琪,「OpenCV教程—基礎篇」,第1版,北京:北京航空航天大學,2009。
[28] 陳勝勇、劉盛、姚春燕、毛國紅、王中杰、林陶、杜雲鶴、唐詩、范蕓、楊慶豐、方銀鋒和管秋,「基於OpenCV的計算機視覺技術實現」,第1版,北京:科學,2009。