臺灣博碩士論文加值系統

在這資訊科技突飛猛進的時代，隨著電腦的運算能力的日益精進，使得影像處理的相關應用蓬勃的發展。影像辨識於行人檢出應用於駕駛輔助系統與監控、保全方面，一直是各方重要的研究課題。
本研究探討了變化檢出(change detection)，小邊線元(ridgelets)描述，以及靜態影像的行人檢出的問題。利用變化檢出以偵測行人的方式可避免行人與背景間的混雜。不過，其方法的應用僅限於對固定背景下的連續影像。本研究提出利用鏈碼長度來濾除行人背景所造成的干擾，並由輸入行人邊緣影像訓練行人樣板，再選取樣板的特徵點以便進行比對。
本研究取72張行人影像作訓練，利用訓練後所得特徵分別對訓練內、外的影像進行測試，並針對其結果探討行人檢出所遭遇的問題。

With the rapid progress in information technology, computer is becoming more and more powerful, applications of digital image processing and machine vision are getting popular. The application of the image recognition to pedestrian detection in the security system and the advanced driver assistance system is an important and active research area.
In this thesis, change detection, ridgelets for the representation of pedestrian with edges, and pedestrian detection on static images are discussed. The change detection method is often used to segment the moving objects out of the scene for pedestrian detection, and must be in the same background. This paper presents a way to detect pedestrian on images in different scenes by limiting chain code length used to reduce the deterioration due to the background noise in the image. In pedestrian detection, we obtain edge templates from input images and extract the feature points by training edge templates. It detects pedestrian in different scenes by the feature points matching.
The research employs 72 pedestrian images for training and uses the extracted features to test. The problem in pedestrian detection is then discussed based on the experimental results.

英文摘要 I
中文摘要 II
誌 謝 III
目 錄 IV
圖表索引 VI
第一章 緒論 1
1.1研究背景與簡介 1
1.2論文架構及綱要 2
第二章 利用變化檢出的行人偵測 4
2.1差異影像法 4
2.2背景模型 6
2.2.1 背景抽離法 6
2.2.2 適應性變化偵測 8
第三章 行人特徵擷取及辨識 12
3.1輪廓擷取 12
3.2小波轉換 13
3.3小波樣板 16
3.4行人樣板 18
3.4.1細節線段與雜點的濾除 19
3.4.2行人樣板訓練 23
3.4.3行人檢出 25
第四章 小邊線元轉換 26
4.1雷登轉換 26
4.2小邊線元轉換 27
第五章 實驗結果 34
5.1實驗設備 34
5.2實驗過程 34
5.3實驗一：利用鏈碼長度濾除細節線段 36
5.3.1站立行人的檢出 36
5.3.2當行人右手舉出時的檢出 39
5.3.3對舉雙手的行人做檢出 42
5.4實驗二：室外環境的行人檢出 45
5.5討論 47
第六章 結論與未來研究方向 51
6.1 結論 51
6.2 未來研究方向 52
參考文獻 53

[1] A. Elgammal, R. Duraiswami, D. Harwood and L. S. Davis, “Background and foreground modeling using nonparametric kernel density estimation for visual surveillance,” Proceedings of the IEEE, vol.90, Issue 7, pp.1151-1163, July 2002.
[2] D. Harwood, A. Elgammal and L. Davis, “Non-parametric model for background subtraction,” Proceedings of the 6th European Conference on Computer Vision, vol.2, pp.751-767, May 2000.
[3] M. Seki, H. Fujiwara and K. Sumi, “A robust background subtraction method for changing background,”, Proceedings of the Fifth IEEE Workshop on Applications of Computer Vision, pp.207-213, 4-6 Dec 2000.
[4] N. Paragios and C. Tziritas, “Detection and location of moving objects using
deterministic relaxation algorithms,” Proceedings of the 13th International
Conference on Pattern Recognition, vol.1, pp.201-205, 25-29 Aug. 1996.
[5] S. Huwer and H. Niemann, “Adaptive change detection for real-time surveillance applications,” Proceedings of the Third IEEE International Workshop on Visual Surveillance, pp.37-46, 1 July 2000.
[6] Chia-Jung Pai, Hsiao-Rong Tyan, Yu-Ming Liang, Hong-Yuan Mark Liao and Sei-Wang Chen,“Pedestrian detection and tracking at crossroads,” Proceedings of the International Conference on Image Processing, vol.2, pp.101-104, 14-17 Sept. 2003.
[7] M. Oren, C. Papageorgiou, P. Sinha, E. Osuna and T. Poggio, “Pedestrian detection using wavelet templates.” Computer Vision and Pattern Recognition, pp.193-199, 1997.
[8] D. M. Gavrila and J. Giebel, “Shape-based pedestrian detection and tracking,” Proceedings of the IEEE Intelligent Vehicle Symposium, vol.1, pp.8-14, 17-21 June 2002.
[9] A. Broggi, M. Bertozzi, A. Fascioli and M. Sechi, “Shape-based pedestrian detection,” Proceedings of the IEEE Intelligent Vehicle Symposium, pp.215-220, 2000.
[10] J. L. Starck, E. Candes and D. L. Donoho, “The curvelet transform for image denoising,” Proceedings of the IEEE Image Processing, vol.11, pp.131–141, June 2002.
[11] R. Jain, R. Kasturi and B. Schunk, Machine Vision, McGraw Hill, 1995.
[12] P. Rosin, “Thresholding for change detection,” Proceedings of the Sixth International Conference on Computer Vision, pp.274-279, 4-7 Jan. 1998.
[13] Xiaolong Dai and S. Khorram, “The effects of image misregistration on the accuracy of remotely sensed change detection,” Proceedings of the IEEE on Geoscience and Remote Sensing, vol.36, no.5, pp.1566-1577, Sept. 1998.
[14] B. T. Phong, “Illumination for computer generated pictures,” Communications of the ACM, vol.18, pp.331-317, 1975.
[15] I. Niemeyer, M. Canty and D. Klaus, “Unsupervised change detection techniques using multispectral satellite images,” Proceedings of IEEE 1999 International Geoscience and Remote Sensing Symposium, IGARSS '99, pp.327-329, 1999
[16] N. Friedman and S. Russell, “Image segmentation in video sequences: A probabilistic approach,” Uncertainty in Articial Intelligence, pp. 175-181, 1997.
[17] W. E. L. Grimson and C. Stauer, “Adaptive background mixture models for real-time tracking,” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, vol.2, pp.252, 23-25 June 1999.
[18] B. B. Chaudhuri and S. Chandrashekhar, “Neighboring direction runlength coding: an efficient contour coding scheme,” Proceedings of the IEEE Systems, Man and Cybernetics, vol.20, no.4, pp.916-921, July-Aug 1990.
[19] 楊德雲，以能量頻譜為隱藏式馬可夫模型輸入的手勢辨識，台灣科技大學電機工程研究所碩士論文，2001。
[20] D. L. Donoho and M. R. Duncan, “Digital curvelet transform: Strategy, implementation and experiments,” Proceedings of SPIE, vol. 4056, pp. 12–29, 2000.
[21] E. J. Candes, “Ridgelets: theory and applications,” Ph.D. Thesis, Statistics, Stanford, 1998.
[22] David L. Donoho, “Orthonormal Ridgelets and Linear Singularities,” Society for Industrial and Applied Mathematics Journals, SIAM J., Math Anal., vol.31, no.5, pp.1062-1099, 2000.
[23] E. J. Candes, “Monoscale ridgelets for the representation of images with edges,” Technical Report, Department of Statistics, Stanford University, 1999.