臺灣博碩士論文加值系統

輪廓萃取技術是一種從圖像中精確獲得該物件之輪廓的方法，這是影像處理中最重要的環節之一，在本文研究中提出了以自我調適遠光/近光探照燈濾波器擷取區域結構資訊之物件輪廓萃取方法。此方法的靈感來自於自身的駕車經驗，當我們在夜晚駕車時，需要開啟燈光來探照前方的道路，當前方的道路相當的直並且無任何的障礙物時，使用遠光燈來照射，可以提早預知何處會有危險，當行駛於蜿蜒曲折的道路時，則改用近光燈。本文所提出之方法是利用預先所設定的一對初始輪廓點，以此為探照的起點和終點進行區域結構資訊的探照運算，並將可能為邊緣的點於相對應之邊緣強度圖進行強化，最後以邊緣強度圖所累積強化的次數做為邊緣判斷的依據，將邊緣強度較弱的點過濾得到輪廓之寬帶圖後並進行細化，最後將取得一個像素點寬之封閉物件輪廓。
本研究選用兩張加入高斯雜訊的人造影像以及兩張真實醫學影像進行實驗並與其他相同之輪廓追蹤方法進行比較，實驗結果顯示出本文提出方法能夠在相同精準度下降低其他探照燈所產生之運算複雜度高的問題，亦能解決基礎探照燈會因展開角度不足而產生偵測失準的問題，且本方法產生之結果能優於其他傳統的方法。

Contour extraction is a method in exactly obtaining an object’s contour from images. It is considered as one of the most important pre-processing for image processing applications. In this study, an object contour extraction using local structure information captured by self-adaptive high beam and low beam spotlight filters is proposed. The proposed method is inspired from the users’ experiences (UE) in car driving, where when the road is very straight and clear, a low beam light is applied, while a high beam light will be utilized when the road is winding and/or the environment is dark. Utilizing the self-adaptive high beam and low beam spotlight filter, the local structural information between two pre-selected initial contour points are obtained and the candidate contour points are then determined using statistics based technique. The spotlight filter continues for all pairs of initial points of an object such that a broadband of the object’s contour is constructed. A thinning process is then applied to obtain the final one-pixel-wide exact object contour. Experimental results using artificial and real medical images showed that the proposed method can obtain better contour extraction performance than other existed methods.

中文摘要............................................i
Abstract..........................................ii
致謝..............................................iii
目錄...............................................iv
圖目錄.............................................vi
表目錄.............................................ix
第1章 緒論..........................................1
1.1 研究背景........................................1
1.2 問題描述........................................2
1.3 研究貢獻........................................3
1.4 章節架構........................................4
第2章 相關研究......................................5
2.1 邊緣偵測........................................5
2.1.1 Leclerc和Zucker方法[9]........................6
2.1.2 Higgins和Hsu的方法[11]........................9
2.2 輪廓擷取.......................................15
2.2.1 主動式輪廓模型[16]............................15
2.2.2 Somkantha 方法[12]...........................17
2.2.3 探照燈濾波器[21][22][23]......................19
2.3 其他既有方法....................................23
2.3.1 反曲特性之輪廓點偵測[13].......................23
第3章 研究架構......................................29
3.1 遠光/近光探照燈之定義............................32
第4章 實驗結果......................................43
4.1 人造影像........................................47
4.1.1 三圓人造圖實驗................................48
4.1.2 人造不規則星形................................64
4.2 真實影像實驗....................................79
第5章 結論與未來工作.................................89
5.1 結論...........................................89
5.2 未來工作........................................90
參考文獻............................................91

[1] T. M. Lehmann, C Goonner, and K. Spitzer,” Survey: Interpolation methods in medical image processing,” IEEE transactions on medical image, vol. 18, no. 11, pp. 1049-1075, 1999.
[2] Bankman, Isaac,” Handbook of medical image processing and analysis,” academic press, 2008.
[3] Y. R. Chen, K. Chao, M. S. Kim,” Machine vision technology for agricultural applications,” Computers and electronics in Agricultural, vol. 36, no. 2, pp. 173-191, 2002.
[4] E. N. Malamas, E. G. M. Petrakis, M. Zervakis, L. Petit, and J. D. Legat,” A survey on industrial vision systems, applications and tools,” Image and vision computing, vol. 21, no. 2, pp. 171-181, 2003.
[5] G. Ramkumar, M. Manikandan,” Face recognition-survey,” Sciencepublication. Org, pp. 260-268, 2013.
[6] W. Zhao, R. CHenllappa, P. J. Phillips, and A. Rosenfeld,” Face recognition: Aliterture survey,” ACM computing surveys, vol. 35, no. 4, pp. 399-458,2003.
[7] Q. Chen, B. Xue, Q. Sun, D. Xia,” Interactive image segmentation based on object contour feature image,” IEEE Conference on Image Processing, pp. 3605-3608, 2010.
[8] R. C. Gonzales,and R. E. Woods,” Digital image Processing 3/e,” prentice hall Upper Saddle River, NJ(2002)

[9] Y. G. LECLERC and S.W. ZUCKER,” The Local Structure of Image Discontinuties in One Dimension,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. PAMI-9, no3, pp. 341-355, May, 1987.
[10] Y. Leclerc,” Capturing the Local Structure of Image Discontinuities in two Dimensions,” Reading in computer vision: issues, problems, principles, and paradigms. Morgan Kaufmann Publishers Inc., 1987.
[11] W. E. Higgins, and C. Hsu,” Edge detection using two-dimensional local structure information,” Pattern Recognition., vol. 27. no. 2, pp. 277-294, 1994.
[12] K. Somkantha, N. Theera-Umpon, and S. Auephanwiriyakul, “Boundary detection in medical images using edge following algorithm based on intensity gradient and texture gradient features,” IEEE Transactions on Biomedica. Engineering, vol. 58, no. 3, pp. 567-573, March, 2011.
[13] R. C. Hsu., D. Y. Chan, W. C. Lai., and C. T. Liu, ”An improved automatic initial snaxel selection with corner tracing for object contour extraction in medical image,” IEEE Visual Communications and Image Processing., pp. 1-4, 2014.
[14] Z. Guo, and R. W. Hall,” Parallel thinning with two-subiteration algorithms,” Communications of the ACM , vol. 32, no. 3, pp. 359-373, 1989.
[15] W. Abu-Ain*, S. N. H. S. Abdullah, B. Bataineh, T. Abu-Ain, and K. Omar,”Skeletionization Algorithm for Binary Images,” Procedia Technology. International Conference on Electrical Engineering and Informatics(ICEEI), pp. 704-709, 2013.
[16] M. Kass, A. Witkin, and D Terzopoulos, ”Snakes: Active contour models, ”International journal of Computer Vision, vol. 1, no. 4, pp. 321-331, 1988.
[17] J. Canny, ”A computational approach to edge detection,”IEEE Transactions on pattern analysis and Machine Intelligence, vol 6, pp. 679-698, 1986.
[18] W. Gao, X. Zhang, L. Yang, and H. Liu,”An Improved Sobel Edge Detection,” IEEE International Conference on Computer Science and Information Technology(ICCSIT), vol. 5, July, 2010.
[19] T. F. Chan and L. A. Vese,”Active Contours Without Edges,”IEEE Transaction on Image Processing, vol. 10, no. 2, pp. 266-277 , Feb, 2001.
[20] C. H. Chuang, and W. N. Lie,”Automatic Snake Contours for the segmentation of multiple objects,” IEEE International Symposium on Circuits and system, vol. 2, pp. 389-392, 2001.
[21] C. T. Liu, C. C. Lin, R. C. Hsu,”An Object Contour Extraction Using Local Structure Information Captured by a Spotlight Filter,” IEEE International Conference on Acoustics, Speech, and Signal Processing(ICASSP), pp. 1177-1180, March, 2012.
[22] R. C. Hsu, C. H. Hsu, C. T. Liu and G.H. Qiu,”A tri-directional spotlight filter for object contour extraction,”IEEE Conference on Audio, Language and Image Processing(ICALIP), pp. 516-520, July, 2014.
[23] 林智全，”利用探照燈濾波器擷取物件邊緣區域結構資訊之物件輪廓萃取方法”，民國102年
[24] Confusion Matrix. [Online]. Available: https://en.wikipedia.org/wiki/Confusion_matrix
[25] The Cancer Image Archive. [Online]. Available: http://www.cancerimagingarchive.net/
[26] ACG Case Reports Journal. [Online]. Available: http://acgcasereports.gi.org/colonic-
diverticulitis-after-endoscopic-band-ligation-performed-for-colonic-diverticular-hemorrhage/
[27] PEIR Digital library. [Online]. Available: http://peir.path.uab.edu/library/
[28] U.S. National Library of Medicine. [Online]. Available: https://www.nlm.nih.gov/research/
visible/visible_human.html
[29] M. J Firbank, A. Coulthard, R. M. Harrison, and E.D. Williams,” A comparison of two methods for measuring the signal to noise ratio on MR images,” IOP SCIENCE Phisics in Medicine and Biology, vol. 44, no. 12, 1999.
[30] 方世榮，” 統計學導論” ， 華泰出版.，2005
[31] A. Jevtic, I. Melgar, D. Andina,” Ant based edge linking algorithm,” IEEE Conference on Industrial Electronics, pp. 3353-3358, 2009.