臺灣博碩士論文加值系統

移動偵測在以電腦視覺為基礎的各種監控系統中是相當重要的一個步驟，目前已有許多方法針對這個部分進行研究，如連續畫面差異(Frame Difference)、背景相減(Background Subtraction)等，在本文中，我們提出一個有效地方法，利用小波轉換在XY平面運算取得影像空間上 (Spatial Domain) 的特徵，並對時間軸做一次小波轉換取得時間上 (Temporal Domain) 的特徵，使用特徵合成及連接元件標記法 (Connected Component Labeling) 完成移動區域偵測，並提出影像校正方法克服各種小波轉換造成偵測區域不準確的問題，使用最小範圍矩形標示偵測區域，而後導入疊代式演算法(Iterative Algorithm)對偵測出的矩形執行後處理，最後加入影像金字塔修正物件破碎情形，使系統正確找出移動物件所在區塊，並以實驗驗證我們的方法。實驗影像序列包含室內與室外人車環境、一般道路與高速公路，實驗結果顯示本方法在陰影明顯的情況下，仍能正確標記出物件位置，並且對亮度突然改變及非移動物件之晃動具有很高的容忍性。

Motion detection is one of the most important procedures in computer vision based surveillance system. Recently, there are lots of related researches about motion detection, such as frame difference and background subtraction. In this paper, we propose an efficient method which uses 2+1D wavelet transformation on video streams to obtain features in spatial and temporal domains. Feature combination and connected component labeling are then performed for moving regions detection. The detected moving region are incorrect due to regions of support of wavelet bases, so we present a location calibration method to conquer this problem, and use minimum bounding rectangles to label the calibrated detecting regions. Finally, we refine the minimum bounding rectangles by an iterative algorithm and adopt a pyramid method to reduce the fragmentation of objects. Our method is verified by experiments. The experimental videos include indoor and outdoor scenes with pedestrians, planar road and expressway with moving vehicles. Our system can find precise object position even when severe shadow effect exists. It is demonstrated by experiments that our method is robust to shadows, sudden light change, and shaking of background or camera.

中文摘要 i
英文摘要 ii
目錄 iv
圖目錄 vi
第一章 導論 1
1.1 研究動機 1
1.2 研究目的 1
1.3 相關研究 2
第二章 論文架構 5
2.1 轉換彩色模型 6
2.2 特徵抽取 6
2.3 前景萃取 7
2.4 前景描述 7
第三章 特徵抽取 9
3.1 小波轉換 9
3.1.1 一維離散小波轉換 9
3.1.2 二維離散小波轉換 10
3.1.3 2+1D小波轉換 14
3.2 特徵合成 17
第四章 前景萃取 19
4.1臨界值選取 19
4.2 位置校正 23
4.3 雜訊去除與破碎修補 26
第五章 前景描述 28
5.1前景標記 28
5.2最小範圍矩形精鍊演算法 29
5.3影像金字塔 31
第六章 實驗結果 33
6.1 實驗環境介紹 33
6.2 實驗結果 35
6.2.1 小波轉換基底與執行過程比較 35
6.2.2 影像金字塔結果 38
6.2.3 與影像差異法比較 40
6.3 特殊情境模擬實驗 41
6.3.1 強烈陰影 41
6.3.2 亮度改變 42
6.4 行人偵測 49
6.5 錯誤類型分析與討論 51
第七章 結論 54
參考文獻 55

[1] B. K. P. Horn, B. G. Schunck., “Determining optical flow,” AI Memo 572. Massachusetts Institue of Technology, 1980.
[2] K. Toyama, J. Krumm, B. Brumitt, and B. Meyers, “Wallflower : Principles and Practice of Background Maintenance,” Internal Conference on Computer Vision, Vol. 1, Page(s): 255-261, Sept. 1999.
[3] Y. Ivanov, A. Bobick, J. Liu, “Fast lighting independent background subtraction,” IEEE Workshop on Visual Surveillance, Page(s): 49 -55, Jan. 1998.
[4] S. Huwer, H. Niemann, “Adaptive change detection for real-time surveillance applications,” IEEE International Workshop on Visual Surveillance, Page(s): 37 -46, July 2000.
[5] M. Yu, G. Jiang, B. Yu, “An integrative method for video based traffic parameter extraction in ITS,” IEEE Asia-Pacific Conference on Circuits and Systems, Page(s): 136 -139, Dec. 2000.
[6] C. Stauffer, W. E. L. Grimson, “Adaptive background mixture models for real-time tracking ,” IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Vol. 2, Page(s): 246-252, June 1999.
[7] M. Tsuchikawa, A. Sato, H. Koike, A. Tomono, “A moving-object extraction method robust against illumination level changes for a pedestrian counting system,” International Symposium on Computer Vision, Page(s): 563 -568, Nov. 1995.
[8] R. Cucchiara, C. Grana,, M. Piccardi, A. Prati, “Statistic and knowledge-based moving object detection in traffic scenes,” IEEE International Conference on Intelligent Transportation Systems, Page(s): 27 -32, Oct. 2000.
[9] M. Tsuchikawa, A. Sato, H. Koike, A. Tomono, “A moving-object extraction method robust against illumination level changes for a pedestrian counting system,” International Symposium on Computer Vision, Page(s): 563 -568, Nov. 1995.
[10] G. L. Foresti, V. Murino and C. Regazzoni, “Vehicle recognition and tracking from road image sequences,” IEEE Transactions on Vehicular Technology, Vol. 48, No. 1, Page(s): 301-318, Jan. 1999.
[11] R. Cucchiara, P. Mello, M. Piccardi, "Image analysis and rule-based reasoning for a traffic monitoring" IEEE Transactions on Intelligent Transportation Systems, Vol. 1, No. 2, Page(s):119-130, June 2000.
[12] A. N. Rajagopalan, R. Chellappa, ”Vehicle detection and tracking in video,” IEEE International Conference on Image Processing, Vol. 1, Page(s): 351 -354, 10-13 Sept. 2000.
[13] S. Pumrin, D. J. Dailey, “Roadside camera motion detection for automated speed measurement,” IEEE International Conference on Intelligent Transportation Systems, Singapore, Page(s):147 — 151, Sept. 2002.
[14] N. Srinivasa, “Vision-based vehicle detection and tracking method for forward collision warning in automobiles,” IEEE International Conference on Intelligent Vehicles, Vol. 2, Page(s): 626 -631, June 2002.
[15] J. B. Kim, C.W. Lee, K. M. Lee, T.S. Yun, H. J. Kim, “Wavelet-based vehicle tracking for automatic traffic surveillance,” TENCON Proceedings of IEEE Region 10 International Conference on Electrical and Electronic Technology, Vol. 1, Page(s): 313 -316, Aug. 2001.
[16] D. M. Weber, D. P. Casasent, “Quadratic Gabor filters for object detection,” IEEE Transactions on Image Processing, Vol. 10, No. 2, Page(s): 218 -230, Feb. 2001.
[17] Z. Sun, G. Bebis, R. Miller, “On-road vehicle detection using Gabor filters and support vector machines,” 14th International Conference on Digital Signal Processing, Vol. 2, Page(s): 1019 -1022, July 2002.
[18] S. Mallat and S. Zhong, “Characterization of signals from multiscale edges,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 14, No. 7, Page(s): 710-732, July 1992.
[19] R. L. Joshi, V. J. Crump, T. R. Fisher, “Image subband coding using arithmetic and trellis coded quantization,” IEEE Transactions on Circuits System and Video Technology, Vol. 5, Page(s): 515—523, Dec. 1995.
[20] V. Wu, R. Manmatha, E. M. Riseman, “Textfinder: an automatic system to detect and recognize text in images,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 21, No. 11, Page(s): 1224 -1229, Nov. 1999.
[21] G. S. K. Fung, N. H. C. Yung, G. K. H. Pang, A. H. S. Lai, “Effective moving cast shadow detection for monocular color image sequences,” 11th International Conference on Image Analysis and Processing, Page(s): 404 -409, Sept. 2001
[22] J. Kato, T. Watanabe, S. Joga, J. Rittscher, A. Blake, “An HMM-based segmentation method for traffic monitoring movies,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 24, No. 9, Page(s): 1291 -1296, Sept. 2002.
[23] IEEE International Workshop on Visual Surveillance and Performance Evaluation of Tracking and Surveillance 2003, http://vspets.visualsurveillance.org/