臺灣博碩士論文加值系統

辨識技術在電腦視覺的領域中，是非常重要的一項課題。在生活中處處可見，如指紋辨識、眼睛辨識、姿勢辨識等應用。人體姿勢由開始到完成可以視為連續的靜態姿勢組成，利用一般攝影機進行辨識會因為其高複雜度的資訊而有所限制。近年來由於體感控制加入姿勢辨識中，原理為利用更具特徵性的骨架資訊和深度資訊，使用者只需要站在攝影機前面就可以達到控制滑鼠和鍵盤的效果。
因此本研究使用Kinect作為輸入端，讓使用者能夠自行輸入動作後經由系統進行辨識，使用者不需侷限於特定的姿勢，利用Kinect所提供的人體關節點座標 值，經過正規化後，可得到以人體肩膀中央為原點的新座標系統。利用動態時間扭曲(Dynamic Time Warping)演算法和最近鄰居法(1-Nearest Neighbor)對使用者的姿勢進行辨識，辨識時計算目前姿勢和訓練樣本的歐幾里得距離，挑選差距最小的為辨識結果，並設計增量(Incremental)演算法讓使用者不需額外訓練就可以維持理想的辨識率。
本系統的固定姿勢在概括樣本(General Sample)平均辨識率可達86.02%，辨識五種自訂姿勢時也可達到75.60%的辨識率，使用增量法比未使用提升了近3%的辨識率，証明此法可行性。

Recognition technology is a very important issue in the area of computer vision.
The applications include fingerprint recognition, iris recognition and gestures recognition in our lives. Body gestures are composed of many continuously static poses that are segmented from the action, it’s difficult to work because high complexity information by using common webcam. In recent years, somatosensory system had become popular in gestures recognition, user could control the object in the screen like using mouse or keyboard.
This paper establishes a human body gestures recognition system, which can recognized the gestures from user defined. Using Kinect to generated 3D coordinates value of skeleton joints, and using translation method to normalized coordinates, after that, we get a new original point by Shoulder Center point. We use Dynamic Time Warping (DTW) algorithm and 1-Nearest Neighbor to compare and classify body gestures. Calculating the Euclidean Distance between training data and testing data. The minimum distance is the result. Finally, we design an Incremental method to keep recognition rate without any extra training by user.
In our system, the average recognition rate of static gestures in general sample is 86.02%, five gestures defined by user is 75.60%, and increasing almost 3% after using incremental method.

附圖目錄..................................................................................................................V
附表目錄................................................................................................................ VI
第一章 簡介........................................................................................................... 1
1.1 研究動機.................................................................................................. 1
1.2 研究目的.................................................................................................. 2
1.3 研究範圍及限制...................................................................................... 2
1.4 論文內容架構.......................................................................................... 3
第二章 文獻探討................................................................................................... 4
2.1 姿勢辨識文獻.......................................................................................... 4
2.1.1 偵測人體資訊.............................................................................. 4
2.1.2 人體姿勢表示.............................................................................. 6
2.1.3 姿勢行為分析.............................................................................. 7
2.2 Kinect 攝影機架構與原理....................................................................... 9
第三章 研究方法................................................................................................. 13
3.1 系統概述和流程圖.................................................................................. 13
3.2 資料前處理………………………………………………………………..14
3.3 特徵擷取…………………………………………………………………..16
3.4 動態時間扭曲......................................................................................... 18
3.5 姿勢辨識................................................................................................. 20
3.6 系統實作................................................................................................. 21
第四章 實驗和結果............................................................................................. 23
4.1 實驗說明................................................................................................. 23
4.2 固定姿勢辨識......................................................................................... 24
4.2.1 單一使用者樣本辨識率.............................................................. 24
4.2.2 概括樣本辨識率.......................................................................... 26
4.3 自訂姿勢辨識......................................................................................... 29
4.3.1 自訂姿勢辨識率.......................................................................... 29
4.3.2 增量辨識率.................................................................................. 31
4.4 結果分析................................................................................................. 33
第五章 結論與未來研究..................................................................................... 34
5.1 結論......................................................................................................... 34
5.2 未來研究................................................................................................. 35
參考文獻................................................................................................................ 36

[1] Kinect for Windows官網. Available: http://www.microsoft.com/en-us/kinectforwindows/.
[2] 李韋廷, “人體姿勢判斷系統應用於投影片控制,”碩士論文,國立台灣師範大學資訊工程所,2010.
[3] S. Yuping and H. Foroosh, “Recognition From Point Triplets,” IEEE Transactions on pattern Analysis and Machine Intelligence, vol. 31, no. 10, pp.1898-1905, 2009.
[4] B. Lo and S. Velastin, “Automatic Congestion Detection System For Underground Platform,” in Proceedings of International Symposium Intelligent Multimedia, Video and Speech, pp.158-161, 2001.
[5] S. Murali, R. Girisha, “Segmentation Of Motion Objects From Surveillance Video Sequences Using Temporal Differencing Combined With Multiple Correlation,” in Proceedings of the International Conference on
Advanced Video and Signal Based Surveillance, pp.472-477, 2009.
[6] A. Efros, A. Berg, G. Mori and J.Malik, “Recognizing Action At Distance,” in Proceedings of the International Conference on Computer Vision, pp.726-733, 2003.
[7] J. Xiaofei, and L. Honghai, “Advances In View-Invariant Human Motion Analysis: A Review,” IEEE Transactions on Systems, Man, and Cybernetics, Part C: Applications and Reviews, vol. 40, no. 1, pp.13-24, 2010.
[8] Y. Zhang, K. Huang, Y. Huang and T.Tan, “View-Invariant Action Recognition Using Cross Ratios Acorss Frames,” in Proceedings of the International Conference on Image Processing, pp.3549-3552, 2009.
[9] M. Ahmad and S. Lee, “HMM-based Human Action Recognition Using Multiview Image Sequences,” in Proceedings of the International Conference on Pattern Recognition, pp.263-266, 2006.
[10] R. Poppe, “Vision-Based Human Motion Analysis: An Overview,” in Proceedings of the International Conference on Computer Vision and Image Understanding, pp.4-18, 2007.
[11] J. Wu, G. Pan, D. Zhang, G. Qi, and S. Li, “Gesture Recognition With A 3-D Accelerometer,” in Proceedings of the International Conference on Ubiquitous Intelligence and Computing, pp.25-38, 2009.
[12] M. A. Turk and A. P. Pentland, “Face Recognition Using Eigenfaces,” in Proceedings of the International Conference on Computer Vision and Pattern Recognition, pp.586-591. 1991.
[13] M. Uddin and T. Kim, “Continuous Hidden Markov Models for Depth Map-Based Human Activity Recognition,” Department of Biomedical Engineering Kyung Hee University, 2011.
[14] S. Yoon and A. Kuijper, “Human Action Recognition Using Segmented Skeletal Features,” in Proceedings of the International Conference on Pattern Recognition, pp.3740-3743, 2010.
[15] E. Baum, T. Prtrie, “Statistical Interference For Probability Functions Of Finite State Markov Chains,” Institute of Defense Analysis, Princetion, N.J., 1966.
[16] B.-W. Min, H.-S. Yoon, J. Soh, Y.-M. Yang, and T. Ejima, “Hand Gesture Recognition Using Hidden Markov Models,” IEEE Transactions on Systems, Man, and Cybernetics, Computational Cybernetics and Simulation,
pp.4232-4235, 1997.
[17] J. Yamato, J. Ohya and K. Ishill, “Recognizing Human Action In Time-Sequential Images Using Hidden Markov Model,” in Proceedings of the International Conference on Computer Vision and Pattern Recognition,
pp.379-385, 1992.
[18] C.-D. Liu, Y.-N Chung and P.-C C, “An Interaction-Embedded HMM Framework for Human Behavior Understanding,” in Proceedings of the International Conference on Information Technology in Biomedicine, vol. 14
no. 5, pp.1236-1246, 2010.
[19] W.Wang, X. Deng, X. Qiu, S. Xia and Z.Wang, “Learning Local Models For 2D Human Motion Tracking,” in Proceedings of the International Conference on Image Processing, pp.2589-2592, 2009.
[20] Kinect in depth. Available: http://www.tomshardware.co.uk/game-developers-conference-gdc-2011-world-of-warcraft,review-32131-2.html.
[21] 吳貴崗, “利用Kinect體感攝影機藉由人體骨架進行人類動作識別,” 碩士論文,國立台北科技大學資訊工程所,2011.
[22] Guiling Li, Wang, Yuanzhen, Min Li and Zongda Wu, “Similarity Match in Time Series Streams under Dynamic Time Warping Distance,” in Proceedings of the International Conference on Computer Science and Software
Engineering, pp.399-402, 2008.
[23] S. Sempena, P.R Aryan and N.U. Maulidevi, “Human Action Recognition Using Dynamic Time Warping,” in Proceedings of the International Conference on Electrical Engineering and Informatics, pp.1-5, 2011.
[24] 周敬恩, “基於軌跡辨識技術之人體姿勢自定與分辨研究,” 碩士論文,國立台灣師範大學資訊工程所,2012.