在壅塞環境中多人的追蹤｜國立清華大學博碩士論文庫

簡易檢索 / 詳目顯示

回結果列表

研究生：	蔡耀德 Yao-Te Tsai
論文名稱：	在壅塞環境中多人的追蹤 Multiple Human Object Tracking in Cluttered Scene
指導教授：	黃仲陵 Chung-Lin Huang
口試委員:
學位類別：	碩士 Master
系所名稱：	電機資訊學院 - 電機工程學系 Department of Electrical Engineering
論文出版年：	2005
畢業學年度：	93
語文別：	英文
論文頁數：	45
中文關鍵詞：	追蹤、遮蔽、混合高斯模型、像素分類
外文關鍵詞：	human tracking, occlusion, Gaussian mixture model, pixel classification
相關次數：	點閱：3 下載：0
分享至:	分享至facebook 分享至twitter

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

多人追蹤是在電腦視覺領域中十分重要的研究課題，而多個人彼此之間經常互相遮蔽使得追蹤變得十分困難。在此篇論文，我們發展了可以追蹤多人的系統並解決遮蔽問題。
我們使用顏色與位移兩個顯著的特徵來區分並追蹤每個人。我們將一個人分為上半身與下半身兩個區域，各自有其獨立的混合高斯色彩模型。為了排除環境亮度的影響，我們使用HIS色彩表示法。每個像素有其色調與飽和度的二維向量。對於每個影像中的像素，我們根據它與某個人的相對位置與顏色的相似度來判斷它是否屬於這個人。每個像素相對於某個人的垂直中央軸與水平參考軸會有一個相對座標，在此相對座標有一”出現機率”，結合此出現機率與混合高斯模型的機率來判別每個像素分別屬於哪一個人。利用此像素分類的概念，我們可將每個人的區域分割出來。分割出每個人的區域後，結合光流，我們就可以獨立的追蹤每一個人。
當發生遮蔽現象時，我們分析遮蔽區域的顏色分佈，來分離每個發生遮蔽的人，並計算出一群遮蔽的人彼此之間的距離。將發生遮蔽的人的色彩模型代入來分割出每個人，移除雜訊後，估測出每個人的垂直中央軸。若某兩人之垂直中央軸的距離超過某門檻值，表示此兩人距離夠遠，我們就將兩人分離。對於每個被分開的人，我們作獨立的追蹤。

Multiple human tracking is an important research topic in computer vision. In the cluttered scene, an occlusion occurs frequently that makes the tracking problem even more challenging. In this thesis, we propose a tracking system to track each human object and solve the occlusion problem. We use color and motion, the two significant features, to distinguish and track different human objects. We classify each pixel to which human object it belongs by its relative position about an object and its color model. Therefore, each object region can be segmented effectively. Combining each object region and optical flow, we can track each object independently. When an occlusion happens, we analyze the color distribution of the occlusion group to differentiate each object in an occlusion. By calculating the distances between each objects, we can determine whether an object separate from the occlusion or not. For an object leaves an occlusion, we treat it as an individual and track it afterward.

Contents
Chapter 1 Introduction.....................................1
1.1 Motivation.............................................1
1.2 Related works..........................................1
1.3 System overview........................................3
Chapter 2 Gaussian Mixture Color Model.....................6
2.1 Gaussian mixture model.................................6
2.2 Color model............................................7
Chapter 3 Initialization and Pixel Classification.........10
3.1 Initialization........................................10
3.2 Identifying a newcomer................................11
3.3 Initialize the presence map...........................13
3.4 Bayesian classification...............................15
Chapter 4 Tracking Individual Object......................16
4.1 Object tracking by optical flow analysis..............17
4.2 The tracking system...................................18
4.3 Update the presence map...............................20
Chapter 5 Tracking Occluded Objects.......................23
5.1 Occlusion detection...................................23
5.2 Tracking under occlusion..............................25
Chapter 6 Experimental Results............................36
6.1 Tracking examples.....................................36
6.2 System evaluation and error analysis..................40
Chapter 7 Conclusion......................................44
Reference.................................................45

                                

[1] I. Haritaoglu, D. Harwood and L. S. Davis, “W4: Real time surveillance of People and Their Activities,” IEEE Trans. on PAMI, vol.22, no.8, Aug 2000.
[2] T. Zhao and R. Nevatia, “Tracking multiple humans in complex situations” IEEE Trans. on Pattern Analysis and Machine Intelligence, Volume 26, Issue 9, pp. 1208 – 1221, Sept. 2004.
[3] A. Mittal and L. S. Davis, “M2tracker: A multi-view approach to segmenting and tracking people in a cluttered scene,” International Journal of Computer Vision 51(3), pp. 189-203, 2003.
[4] H. Tsutsui, J. Miura and Y. Shirai, “Optical flow-based person tracking by multiple cameras”, Proc. International Conference on Multisensor Fusion and Integration for Intelligent Systems, pp. 91-96 Germany, Aug. 2001.
[5] S. Kang, B. Hwang and S. W. Lee, “Multiple people tracking based on temporal color feature,” International Journal of Pattern Recognition and Artificial Intelligence Vol. 17, No. 6 , 2003.
[6] J. Krumm, S. Harris, B. Meyers, B. Brumitt, M. Hale, and S. Shafer, “Multi-camera multi-person tracking for EasyLiving” Proceedings of the Third IEEE International Workshop on Visual Surveillance, pages 3-10, 2000.
[7] T. Chang and S. Gong, “Tracking multiple people with a multi-camera system” IEEE Workshop on Multi-Object Tracking, July, 8, Page(s):19 – 26, 2001.
[8] M. Isard and A. Blake, “CONDENSATION-Conditional density propagation for visual tracking,” International Journal Computer Vision, Vol. 29, No.1, 1998.
[9] M. Isard and J. MacCormick, “BraMBLe: Bayesian Multiple-Blob Tracker,” Proc. of Int. Conf. on Computer Vision, vol.2, pp.34-41, 2001.
[10] S. J. McKenna, Y. Raja, and S. Gong, “Tracking color objects using adaptive mixture models,” Image and Vision Computing, vol. 19, pp.225-231.
[11] Y. Huang and X. Ding, “Real-time object tracking with occlusion reasoning”, Proceedings of the 6th Asian Conference on Computer Vision, pp. 776-781, Jeju, Korea, Jan. 27-30, 2004.
[12] R.A. Redner and H.F. Walker, “Mixture densities, maximum likelihood and the EM algorithm” in SIAM Review 26 (2) (1984) 195-239.
[13] N. Peterfreund, “Robust tracking of position and velocity with Kalman snakes” IEEE Trans. on PAMI, Volume 21, Issue 6, June 1999 Page(s):564 – 569
[14] N. T Siebel and S. Maybank, “Fusion of Multiple Tracking Algorithms for Robust People Tracking” in Proc. of the 7th European Conference on Computer Vision-Part IV Vol. 4, pp 373 - 387 .
[15] Y. Wang, J. Ostermann, and Y. Zhang, “Video Processing and Communications”, pp 142-154, 2002.
[16] R. C. Gonzalez, R. E. Woods, “Digital image procession”, 2nd Edition, 2002.

全文公開日期本全文未授權公開 (校內網路)
全文公開日期本全文未授權公開 (校外網路)

簡易檢索 / 詳目顯示

相關論文