簡易檢索 / 詳目顯示

回結果列表
研究生： 林祐慶
You-ching Lin
論文名稱： 應用於壓縮視訊環境之跌倒偵測系統
Fall Detection in Compressed Video
指導教授： 陳郁堂
Yie-Tarng Chen
口試委員: 陳省隆
Hsing-Lung Chen
方文賢
W.-H. Fang
吳乾彌
Chen-Mie Wu
學位類別： 碩士
Master
系所名稱： 電資學院 - 電子工程系
Department of Electronic and Computer Engineering
論文出版年： 2009
畢業學年度： 97
語文別： 英文
論文頁數： 62
中文關鍵詞： 跌倒偵測 、三角形幾何顏色統計圖 、人體骨架 、雙分圖 、電腦視覺
外文關鍵詞： fall detection, triangular geometric histogram, human skeleton, bi-partite graph, computer vision
相關次數： 點閱：156  下載：4 
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

    • 近年來，因全世界的人口逐漸老年化，也越來越多獨居老人，所以與老人看護相關的研究日益備受重視。因此使用智慧型影像監視系統來偵測跌倒事件或分析人類行為也變成了熱門的研究議題。在本研究中，我們實做一個在壓縮視訊下以人體骨架為基礎的跌倒偵測系統。在這個系統中的主要議題包含了在壓縮視訊領域下的影像重建，多重物件追蹤與遮蔽效應的處理和偵測跌倒的演算法。首先，我們使用壓縮影像環境中的DC+2AC資訊重建影像像素，然後使用以貝氏分類 (Bayesian classification) 分離出前景物件及背景。為了增強空間顏色資訊，我們利用三角形幾何顏色統計圖 (Triangular geometric histogram) 來計算前景物件的相似度。另外我們把多重物件追蹤的問題轉成在雙分圖 (bi-partite graph) 上找最大權重配對的問題，並利用匈牙利演算法取得最佳配對結果。最後利用人體骨架辨識及人體外型的變化率來偵測行人跌倒事件。為了驗證這個跌倒偵測系統的效能，我們利用自行拍攝的跌倒影像做了一些實驗。實驗顯示，我們所提出的偵測系統有高偵測率及低誤報率的結果。

    In recent years, the growth of world’s aging population make more elderly people living alone, and elderly care becomes a serious problem. Hence developing an intelligent video surveillance system, which can detect fall incidence or human behaviors, becomes a hot research topic. In this research, we implement a skeleton-based fall detection system for the compressed video. Main issues in this research include image reconstruction from the compressed video domain, multiple objects tracking with occlusion handling, and fall detection. First, we reconstruct pixel values by using DC+2AC values in the compressed video. Then, we use Bayesian classification to discriminate foreground object and background. To enhance the spatial color information, we use triangular geometric histogram to measure the similarity in object tracking. Simultaneously, we convert the multiple objects tracking problem to the problem of finding maximum weight matching on a bi-partite graph, and we use the Hungarian algorithm to solve this problem. Finally, we combine the skeleton analysis, the ellipse of human body, and the change ratio of human shape to detect the fall incident. To verify the performance of the fall detection system, we perform intensive experiments based on videos. The experiment results reveal that the proposed fall detection system can achieve high detection rate and low false positive.

    摘要 I Abstract II 誌謝 III Contents IV List of Tables VI List of Figures VII 1. Introduction 9 1.1 Overview 9 1.2 Problem Statements 12 1.3 System Overview 13 1.4 Related work 14 1.5 Contributions 15 1.6 Organization of the thesis 15 2. Background 16 2.1 MPEG overview 16 2.2 MPEG Encoding / Decoding 18 2.3 Build DC+2AC image 19 2.4 Distance Metrics 21 2.4.1 Bhattacharyya distance 21 2.4.2 Euclidean distance 21 3. Object Detection Model 22 3.1 Foreground Extraction 24 3.2 Pre-processing 27 4. Object Tracking Model 28 4.1 Object Features Extraction 30 4.1.1 Ellipse fitting 30 4.1.2 Triangular Geometric Histogram 32 4.2.1 Object Distance Measurement 34 4.2.2 Minimum Weight Matching 35 4.3 Objects Merge/Split handling 38 5. Fall-down Detection Model 40 5.1 Fall-down Incident Detection 41 5.2 Posture Change Detection 43 5.2.1 Extract control points use the Douglas-Peucker algorithm 44 5.2.2 Delaunay Triangulation 46 5.2.3 Human Skeleton Extraction 48 5.2.4 Posture Matching use the Distance Map of the Skeleton 50 5.3 Analysis Shape’s Change Ratio 52 5.4 Fall-down Conformation 54 6. Experimental 55 6.1 Performance Metrics 55 6.2 Experimental Result 56 6.2.1 Fully decoded image V.S DC+2AC image 56 6.2.2 Detect object merge/split use TGH 57 6.3.3 Fall Detection Result 58 7. Conclusion and Future Works 60 8. Reference 61

    [1] 蔡益堅, 葉純志, 藍祚運, 張文瓊, and 曾德運, "1999年與2005年台灣地區老人跌倒狀況之比較," 行政院衛生署國民健康局, 2006.
    [2] Glen Debard, Bert Bonroy, Bart Vanrumste, Toon Goedemé Jared Willems, "How to detect human fall in video? An overview," Positioning and Context-Awareness International Conference, Antwerp, 2009.
    [3] Roger Haslam and David Stubbs, Understanding and preventing falls.: CRC Press, 2006.
    [4] Chia-Wen Lin, Zhi-Hong Ling, Yeng-Cheng Chang, and Chung J. Kuo, "Compressed-domain Fall Incident Detection for Intelligent Homecare," Journal of VLSI Signal Processing Systems, vol. 49, no. 3, pp. 393 - 408, 2007.
    [5] C.,Meunier, J., St-Arnaud, A.,Rousseau, J. Rougier, "Fall Detection from Human Shape and Motion History Using Video Surveillance," Advanced Information Networking and Applications Workshops, pp. 875-880, 2007.
    [6] Toshiki Iwasaki, Toshinori Watanabe Takanori Yokoyama, "Motion Vector Based Moving Object Detection and Tracking in the MPEG Compressed Domain," Seventh International Workshop on Content-Based Multimedia Indexing, pp. 201-206, 2009.
    [7] Chittaranjan Manda and Shamik Sural Vinay Vishwakarma, Automatic Detection of Human Fall in Video.: Springer Berlin / Heidelberg, 2007.
    [8] U. Lindemann, A. Hock, M. Stuber, and et al., "Evaluation of a fall detector based on accelerometers: a pilot study," Medical & Biological Engineering & Computing, vol. 43, pp. 548-551, 2005.
    [9] T. Zhang, J. Wang, L. Xu, and et al., "Using wearable sensor and NMF algorithm to realize ambulatory fall detection," Lecture Notes in Computer Science, Advances in Natural Computation, vol. 4222, pp. 488-491, 2006.
    [10] Soo-Chang PeiChing-Min ChengLung-Feng Ho, "Limited color display for compressed image and video," IEEE Transactions on Circuits and Systems for Video Technology, vol. 10, no. 6, pp. 913-922, 2000.
    [11] Yang Gaobo, Chen Weiwei, Zhang Zhaoyang Zhou Qiya, "A Fast and Accurate Moving Object Extraction Scheme in the MPEG Compressed Domain," Proceedings of the Fourth International Conference on Image and Graphics, pp. 592-597, 2007.
    [12] International Standard Organization. (1993) ISO/IEC 11172-2:Information technology -- Coding of moving pictures and associated audio for digital storage media at up to about 1,5 Mbit/s -- Part 2: Video.
    [13] Lena Gorelick, Moshe Blank, Eli Shechtman, and Michal. Weizmann dataset. [Online]. http://www.wisdom.weizmann.ac.il/~vision/SpaceTimeActions.html

    [14] D.-T. Chen , C.-W. Su and H.-R. Tyan H.-Y. M. Liao, "Real-time event detection and its application to surveillance systems," Proc. IEEE Comp. Sci. Conf. Comput. Vis. Pattern Recogni, vol. 3, pp. 512-515, Sep. 2005.
    [15] Gary Rost Bradski and Adrian Kaehler, Learning OpenCV : Computer Vision with the OpenCV Library.: O'Reilly, 2008.
    [16] W. Huang, I. Y. H. Gu, and Q. Tian L. Li, "Foreground object detection from videos containing complex background," Proceedings of the eleventh ACM international conference on Multimedia, pp. 2 - 10, 2003.
    [17] Aibing Rao, Rohini K. Srihari, and Zhongfei Zhang, "Geometric histogram: a distribution of geometric configurations of color subsets," Internet Imaging, no. 3964, pp. 91-101, 2000.
    [18] Alper Yilmaz, Omar Javed, and Mubarak Shah, "Object tracking: A survey," ACM Computing Surveys, vol. 38, no. 4, p. Article 13th, 2006.
    [19] Johannes Kilian. (2001) Simple Image Analysis By Moments: Version 0.2.
    [20] Caroline Rougier, Jean Meunier, Alain St-Arnaud, and Jacqueline Rousseau, "Procrustes Shape Analysis for Fall Detection," in The Eighth International Workshop on Visual Surveillance, 2008, pp. xx-yy.
    [21] Daniel Keysers, Thomas Deselaers, and Hermann Ney, Pixel-to-Pixel Matching for Image Recognition Using Hungarian Graph Matching.
    [22] Y. T. Hsu, H. Y. M. Liao, and C. C. Chen J. W. Hsieh, "Video-based Human Movement Analysis and its Application to Surveillance Systems," IEEE Transactions on Multimedia, vol. 10, no. 3, pp. 372-392, April 2008.
    [23] Gunilla Borgefors, "Distance Transformations in Digital Images," COMPUTER VISION, GRAPHICS, AND IMAGE PROCESSING, no. 34, pp. 344-371, 1986.

    QR CODE