簡易檢索 / 詳目顯示

回結果列表
研究生： 粘光裕
Nien, Kuang-Yu
論文名稱： 具前景分佈趨勢過濾與追蹤機制之輪廓為基礎的行人偵測
Contour-based Human Detection with Foreground Distribution Trend Filtering and Tracking
指導教授： 陳培殷
Chen, Pei-Yin
共同指導教授: 胡敏君
Hu, Min-Chun
學位類別： 碩士
Master
系所名稱： 電機資訊學院 - 資訊工程學系
Department of Computer Science and Information Engineering
論文出版年： 2014
畢業學年度： 102
語文別： 英文
論文頁數： 38
中文關鍵詞： 行人偵測 、行人追蹤 、監視影片 、前景分布趨勢 、擁擠環境
外文關鍵詞： Human detection, pedestrian detection, surveillance video, foreground distribution trend, crowded scene
相關次數： 點閱：127  下載：0 
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

    • 在監視系統的影片的分析系統中，在擁擠環境中去偵測行人是一件有利且但挑戰性的事。我們以輪廓為基礎的行人偵測為基礎，加入兩個新的方法: 前景分布趨勢過濾, 追蹤行人。前景分布趨勢過濾是透過分析誤判的前景點在x方向的分佈趨勢，進而把那些可以辨別的false alarm刪除，提升偵測行人的準確率(precision)。而追蹤行人原本是把人的行徑路線記錄下來，但我們可以利用這個特性，把那些漏掉偵測的人透過插補位置的方式補回來，來提高召回率(recall)。我們也可以透過一個假設，如果是誤判的話，可能只會在幾楨(frame)中出現，我們把這種的情況也刪除，來提高行人偵測的準確率(precision)。實驗結果證明我們提出方法可以提高準確率(precision)與召回率(recall)。

    In video surveillance, detecting human in crowded environment is profitable but challenging. Based on contour-based human detection, we add two new methods: foreground distribution trend filter and tracking. Foreground distribution trend filter deletes those distinguishable false alarms by analyzing foreground distribution trend along x-axis of false alarms. Tracking is to record the trajectory a human pass. We use this feature to recover those missing detected humans by interpolating positions and consequently increase recall. We also make a hypothesis that false alarms may only occur in few frames so we can delete them to increase precision. Experimental results show that our proposed methods can improve precision and recall.

    摘要 III ABSTRACT IV 誌謝 V CONTENTS VI LIST OF FIGURES VIII LIST OF TABLES X CHAPTER 1. INTRODUCTION 1 CHAPTER 2. RELATED WORK 1 CHAPTER 3. PROPOSED METHOD 4 3.1 CONTOUR-BASED HUMAN DETECTION 4 3.1.1 Problem Formulation 4 3.1.2 Template Matching 5 3.1.3 Template Generation 7 3.1.4 Sparse Contour Template Sampling 7 3.1.5 Template Scaling 9 3.1.6 Foreground Ratio Filtering 10 3.1.7 GPU-Assisted Gabor-Based Edge Detection 10 3.2 FOREGROUND FILTERING REFINEMENT 11 3.2.1 Investigation of filtering size for scanning window 11 3.2.2 Foreground distribution trend filter 12 3.3 TRACKING ALGORITHM 16 3.3.1 The way to establish tracking lists 16 3.3.2 Reduce false alarms 19 3.3.3 Reduce missing detections 20 CHAPTER 4. EXPERIMENT RESULTS 21 4.1 HUMAN DETECTION ACCURACY 23 4.1.1 Influences of different filtering size for scanning window 24 4.1.2 Influences of foreground distribution trend filter 25 4.1.3 Influences of deletion mechanism in our tracking algorithm 27 4.1.4 Influences of interpolation mechanism in our tracking algorithm 28 4.1.5 Influences of deletion and interpolation 29 4.1.6 Result of combining above method 30 4.2 TIME COMPLEXITY 31 4.2.1 Influences of filtering size 31 4.2.2 Influences of foreground distribution trend filter 32 4.2.3 Influences of tracking algorithm 32 CHAPTER 5. CONCLUSIONS 34 REFERENCES 35

    [1] C. R. Wren, A. Azarbayejani, T. Darrell and A. P. Pentland, "Pfinder: Real-time tracking of the human body," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 19, no. 7, pp. 780-785, 1997.
    [2] C. Stauffer and W. E. L. Grimson, "Adaptive background mixture models for real-time tracking," in Proc. IEEE Computer Society Conf. on Computer Vision and Pattern Recognition, pp. 246-252, 1999.
    [3] B. Wu and R. Nevatia, "Detection of multiple, partially occluded humans in a single image by bayesian combination of edgelet part detectors," in Proc. IEEE Int. Conf. on Computer Vision, pp. 90-97, 2005.
    [4] A. Opelt and A. Zisserman, "A boundary-fragment-model for object detection," in Proc. European Conference on Computer Vision, pp. 575-588, 2006.
    [5] V. Ferrari, T. Tuytelaars and L. V. Gool, "Object detection by contour segment networks," in Proc. European Conference on Computer Vision, 2006.
    [6] C. Papageorgiou and T. Poggio, "A trainable system for object detection," IInt. J. Comput. Vision, vol. 38, no. 1, pp. 15-33, 2000.
    [7] N. Dalal and B. Triggs, "Histograms of oriented gradients for human detection," in Proc. IEEE Computer Society Conf. on Computer Vision and Pattern Recognition, pp. 886-893, 2005.
    [8] D. M. Gavrila and J. Giebel, "Shape-based pedestrian detection and tracking," in Proc. IEEE Intelligent Vehicle Symposium, pp. 8-14, 2002.
    [9] D. Toth and T. Aach, "Detection and recognition of moving objects using statistical motion detection and fourier descriptors," in Proc. Int. Conf. on Image Analysis and Processing, 2003.
    [10] J. Zhou and J. Hoang, "Real time robust human detection and tracking system," in Proc. IEEE Computer Society Conf. on Computer Vision and Pattern Recognition, pp. 149-149, 2005.
    [11] J. Rittscher, P. H. Tu and N. Krahnstoever, "Simultaneous estimation of segmentation and shape," in Proc. IEEE Computer Society Conf. on Computer Vision and Pattern Recognition, pp. 486-493, 2005.
    [12] T. Zhao and R. Nevatia, "Tracking multiple humans in crowded environment," in Proc. IEEE Computer Society Conf. on Computer Vision and Pattern Recognition, pp. 406-413, 2004.
    [13] T. K. Svoboda, Jan/ Hlavac, Vaclav, "Image processing: Analysis and machine vision," 2007.
    [14] C. Beleznai and H. Bischof, "Fast human detection in crowded scenes by contour integration and local shape estimation," in Proc. IEEE Computer Society Conf. on Computer Vision and Pattern Recognition, pp. 2246-2253, 2009.
    [15] Caviar dataset: http://homepages.inf.ed.ac.uk/rbf/CAVIAR/.
    [16] Qt 4.6: http://qt.nokia.com.
    [17] Open cv library 2,0: http://opencv.willowgarage.com/wiki/Welcome.
    [18] T. F. Cootes, C. J. Taylor, D. H. Cooper and J. Graham, "Active shape models—their training and application," Computer Vision and Image Understanding, vol. 61, no. 1, pp. 38–59, 1995.
    [19] C. M. Bishop, "Pattern recognition and machine learning," 2006.
    [20] I. T. Jolliffe, "Principal component analysis," 1986.
    [21] K. Kim, T. Chalidabhongse, D. Harwood and L. Davis, "Real-time foreground–background segmentation using codebook model," Real-Time Imaging In Special Issue on Video Object Processing, vol. 11, no. 3, pp. 172-185, 2005.
    [22] K. G. Derpanis, York University, memo, "Integral image-based representations," 2007.
    [23] F. C. Crow, "Summed-area tables for texture mapping," in Proc. Int. Conf. on Computer Graphics and Interactive Techniques, pp. 207-212, 1984.
    [24] P. Y. Simard, L. Bottou, P. Haffner and Y. LeCun, "Boxlets: A fast convolution algorithm for signal processing and neural networks," Advances in neural information, M. Kearns, S. Solla and D. Cohn, MIT Press, vol. 11, pp. 571–577, 1999.
    [25] P. Viola and M. Jones, "Rapid object detection using a boosted cascade of simple features," in Proc. IEEE Computer Society Conf. on Computer Vision and Pattern Recognition, pp. 511-518, 2001.
    [26] O. Tuzel, F. Porikli and P. Meer, "Region covariance: A fast descriptor for detection and classification," in Proc. European Conference on Computer Vision, pp. 589–600, 2006.
    [27] X. Wang, G. Doretto, T. Sebastian, J. Rittscher and P. Tu, "Shape and appearance context modeling," in Proc. IEEE Int. Conf. on Computer Vision, pp. 1-8, 2007.
    [28] Tutorial on gabor filters: http://mplab.ucsd.edu/tutorials/gabor.pdf.
    [29] Cuda: http://www.nvidia.com/object/cuda_home_new.html.
    [30] M. T. PC Chang, JL Wu, CS Hu, "Real-time gender classification from human gait for arbitrary view angles," in Proc. IEEE ISM, 2009.
    [31] B. L. HC Lian, "Multi-view gender classification using local binary patterns and support vector machines," in Proc. ISNN, 2006.
    [32] M. S. J. Xuelong Li, Shuicheng Yan, Dacheng Tao, Dong Xu "Gait components and their application to gender recognition " IEEE transactions on systems, man and cybernetics, vol. 38 (2) pp. 145-155 2008.
    [33] Min-Chun Hu and Wen-Huang Cheng and Chuan-Shen Hu and Ja-Ling Wu and Jhe-Wei Li "Efficient human detection in crowded environment" Multimedia Systems, Springer Berlin Heidelberg, pp. 1-11, 2014
    [34] Chuan-Shen Hu, Min-Chun Hu, Wen-Huang Cheng, and Ja-Ling Wu. 2013. Efficient human detection in crowded environment based on motion and appearance information. In Proceedings of the Fifth International Conference on Internet Multimedia Computing and Service (ICIMCS '13). ACM, New York, NY, USA, 97-100. DOI=10.1145/2499788.2499837 http://doi.acm.org/10.1145/2499788.2499837
    [35] Dalal, N.; Triggs, B., "Histograms of oriented gradients for human detection," Computer Vision and Pattern Recognition, 2005. CVPR 2005. IEEE Computer Society Conference on , vol.1, no., pp.886,893 vol. 1, 25-25 June 2005

    無法下載圖示 校內：立即公開
    校外：不公開
    電子論文尚未授權公開，紙本請查館藏目錄
    QR CODE