臺灣博碩士論文加值系統

隨著醫學造影與成像技術的進步，醫學影像在臨床的診斷上扮演著重要的輔助角色，可選擇使用不同的醫學成像工具，例如X射線二維的成像到各式三維的斷層掃瞄技術，能夠以非侵入人體的方式來透視人體器官或部分組織的具體輪廓來獲得視覺化的人體器官影像資訊，並利用醫療用電腦影像系統來輔助醫生進行病情的診斷。而現今使用的醫療用影像分析技術中，我們需要分析出影像中的邊緣資訊作為診斷病情的輔助資訊，因此精準地切割出器官的三維曲面 (3D surface) 是首要任務，正確的分割結果將會影響後續的分析，如何將三維的邊緣曲面正確地標記出來則是對於三維醫學影像前處理十分關鍵的課題。現今最廣泛使用的Canny邊緣偵測法，是依據三方向相鄰像素的差分數值來計算梯度大小和方向，此方法對於雜訊的變化較為敏感，導致邊緣影像出現許多不必要的偵測結果。因此，我們提出多角度的邊緣遮罩解決此問題，以不同平面切割方向的立體遮罩偵測出在像素上的差分變化，並利用降取樣排除雜訊的干擾，達到邊緣追蹤的效果，避免顆粒性的雜訊被誤判成邊緣影像，再藉由區塊性標準化克服區域內對比度不佳的情形，提升邊緣偵測的準確性。實驗結果顯示本論文提出的技術能有效提升影像邊緣偵測的準確性，除了在醫學影像使用之外，這項技術也可以應用在其他三維物件邊緣偵測。

With the progress of the technology of medical imaging, it plays an important role in clinical diagnosis. From X-ray imaging technology to various types of tomography, medical image shows the information of human body or part of tissue. Medical image shows the visualized and perspective information of concretized contour of human organ or partial tissue through noninvasive diagnosed method. Analyze the edge information from the contour image that can be used as supplementary information for disease diagnosis. However, due to uneven gray-scale distribution of computed tomography images, three-dimensional Canny edge detection method based on gradient values may easily cause chaotic noise messages in the edge images. Therefore, we derive a three-dimensional operator, which use the different cutting directions, will find the best orientation at each point in the image. This operator is directly based on the three-dimensional problem. In addition, our algorithm uses down sampling method to overcome the chaotic noise messages. Our experimental result shows that the algorithm we proposed in this paper could effectively enhance the accuracy of contour detection of edge image. In addition to the usage of medical images, these techniques can also be applied to the contour detection of edge images in other three-dimensional objects.

目錄 IV
致謝 VI
中文摘要 VII
ABSTRACT VIII
圖目錄 IX
表目錄 XII
第一章 緒論 1
1.1 研究動機與目的 1
1.2 研究背景與發展現況 1
1.3 論文簡介 2
1.4 論文架構 3
第二章 背景介紹 4
2.1 醫學影像 4
2.1.1 X光 4
2.1.2 超音波 (Ultrasound) 5
2.1.3 核磁共振成像 (Magnetic Resonance Imaging) 6
2.1.4 電腦斷層掃描 (Computed Tomography) 6
2.2 二維影像邊緣偵測 7
2.2.1 空間濾波的基本原理 7
2.2.2 空間相關性(Cross-correlation)與卷積(Convolution) 10
2.3 常見的二維影像邊緣偵測運算子 14
2.3.1 Roberts 交叉梯度運算子 (Roberts Cross Gradient Operators) 17
2.3.2 蒲瑞維特運算子 (Prewitt Operator) 18
2.3.3 索貝爾運算子 (Sobel Operator) 19
2.3.4 Canny 邊緣檢測演算法 20
2.4 常見的三維影像分割之方法 22
2.4.1 三維Canny邊緣偵測演算法 22
2.4.2 三維表面檢測統計法 23
第三章 三維多方向性邊緣曲面偵測演算法 25
3.1 三維邊緣曲面偵測 25
3.1.1 產生多方向性遮罩的方法 27
3.1.2 三維面檢測遮罩之運算方法 30
3.1.3 三維遮罩大小之討論 32
3.1.4 降取樣 (Down sampling) 34
3.2 三維遮罩邊緣演算法和CANNY EDGE之比較 36
3.3 探討DOWN SAMPLING之次數與速度比較 39
第四章 實驗結果 43
4.1 邊緣偵測表現的評估方法 43
4.2 邊緣偵測結果的數值評估比較 44
4.2.1 基於電腦斷層影像 44
4.2.2 基於人造球型影像混高斯雜訊 48
4.2.3 基於人造雙球型 52
4.2.4 基於人造頭部假體影像 55
第五章 結論與未來展望 61
REFERENCE 62
附錄一：十三種方向的3_3遮罩 65
附錄二：有無降取樣之影像結果比較 69
附錄三：人造球型影像之偵測結果 70
附錄四：一至三次降取樣在不同遮罩下的偵測結果 71

[1] Kak, Avinash C., and Malcolm Slaney. Principles of computerized tomographic imaging. Society for Industrial and Applied Mathematics, 2001.
[2] Beckmann, Christian F., and Stephen M. Smith. "Probabilistic independent component analysis for functional magnetic resonance imaging." IEEE transactions on medical imag¬ing 23.2 (2004): 137-152.
[3] Nagarkar, V. V., et al. "Structured CsI (Tl) scintillators for X-ray imaging applications." IEEE transactions on nuclear science 45.3 (1998): 492-496.
[4] Ollinger, John M., and Jeffrey A. Fessler. "Positron-emission tomography." IEEE Signal Processing Magazine 14.1 (1997): 43-55.
[5] Lockwood, Geoffrey R., James R. Talman, and Shelby S. Brunke. "Real-time 3-D ultra-sound imaging using sparse synthetic aperture beamforming." IEEE transactions on ultrason¬ics, ferroelectrics, and frequency control 45.4 (1998): 980-988.
[6] Gonzalez, “Digital image processing‚” Prentice Hall, 2008.
[7] Peters, Jochen, et al. "Optimizing boundary Detection via simulated search with applications to multi-modal heart segmentation." Medical image analysis 14.1 (2010): 70-84.
[8] Heimann, Tobias, and Hans-Peter Meinzer. "Statistical shape models for 3D medical image segmentation: a review." Medical image analysis 13.4 (2009): 543-563.
[9] Zhang, Xing, et al. "Automatic liver segmentation using a statistical shape model with optimal surface Detection." IEEE Transactions on Biomedical Engineering 57.10 (2010): 2622-2626.
[10] Roberts, Lawrence G. Machine perception of three-dimensional solids. Diss. Massachusetts Institute of Technology, 1963.
[11] Prewitt, Judith MS. "Object enhancement and extraction." Picture processing and Psychopictorics 10.1 (1970): 15-19.
[12] Sobel, Irvin. "An isotropic 3× 3 image gradient operator." Machine vision for three-dimensional scenes (1990): 376-379.
[13] Canny, John. "A computational approach to edge Detection." IEEE Transactions on pattern analysis and machine intelligence6 (1986): 679-698.
[14] Bao, Paul, Lei Zhang, and Xiaolin Wu. "Canny Edge Detection enhancement by scale multiplication." IEEE transactions on pattern analysis and machine intelligence 27.9 (2005): 1485-1490.
[15] Nikolic, Marina, Eva Tuba, and Milan Tuba. "Edge Detection in medical ultrasound images using adjusted Canny Edge Detection algorithm." Telecommunications Forum (TELFOR), 2016 24th. IEEE, 2016.
[16] Arbelaez, Pablo, et al. "Contour Detection and hierarchical image segmentation." IEEE transactions on pattern analysis and machine intelligence 33.5 (2011): 898-916.
[17] Abdou, Ikram E., and William K. Pratt. "Quantitative design and evaluation of enhancement/thresholding edge detectors." Proceedings of the IEEE 67.5 (1979): 753-763.
[18] Zucker, Steven W., and Robert A. Hummel. "A three-dimensional edge operator." IEEE Transactions on Pattern Analysis and Machine Intelligence 3 (1981): 324-331.
[19] Halder, Amiya, et al. "Edge Detection: A statistical approach." Electronics Computer Technology (ICECT), 2011 3rd International Conference on. Vol. 2. IEEE, 2011.
[20] Williams, Ian, Nicholas Bowring, and David Svoboda. "A performance evaluation of statistical tests for edge Detection in textured images." Computer Vision and Image Understanding122 (2014): 115-130.
[21] Luo, Limin M., et al. "A moment-based three-dimensional edge operator." IEEE transactions on bio-medical engineering 40.7 (1993): 693.
[22] Smith, Samuel, and Ian Williams. "A statistical method for improved 3D surface Detection." IEEE Signal Processing Letters22.8 (2015): 1045-1049.
[23] Ecabert, Olivier, et al. "Automatic model-based segmentation of the heart in CT images." IEEE transactions on medical imaging27.9 (2008): 1189-1201.
[24] Brejl, Marek, and Milan Sonka. "Object localization and border Detection criteria design in edge-based image segmentation: automated learning from examples." IEEE Transactions on Medical imaging 19.10 (2000): 973-985.
[25] Dollar, Piotr, Zhuowen Tu, and Serge Belongie. "Supervised learning of edges and object boundaries." Computer Vision and Pattern Recognition, 2006 IEEE Computer Society Conference on. Vol. 2. IEEE, 2006.
[26] Rong, Weibin, et al. "An improved Canny Edge Detection algorithm." Mechatronics and Automation (ICMA), 2014 IEEE International Conference on. IEEE, 2014.
[27] Marr, David, and Ellen Hildreth. "Theory of edge Detection." Proceedings of the Royal Society of London B: Biological Sciences 207.1167 (1980): 187-217.
[28] Lindeberg, Tony. "Edge Detection and ridge Detection with automatic scale selection." International Journal of Computer Vision 30.2 (1998): 117-156.
[29] Haralick, Robert M. "Digital step edges from zero crossing of second directional derivatives." IEEE Transactions on Pattern Analysis and Machine Intelligence 1 (1984): 58-68.
[30] Zheng, Yefeng, et al. "Four-chamber heart modeling and automatic segmentation for 3-D cardiac CT volumes using marginal space learning and steerable features." IEEE transactions on medical imaging 27.11 (2008): 1668-1681.
[31] Yamini, B., and R. Sabitha. "Adaptive Image Steganalysis for LSB Embedding Technique using Enhanced Canny Operator." Indian Journal of Science and Technology 9.10 (2016).