本論文提出了一套應用於結核菌顯微影像上之自動顯微影像結核菌檢測系統，以達到自動取像並且偵測結核菌區域，並針對檢出區域加以辨識的目的。
整個系統可以細分為兩個子系統：自動對焦系統與結核菌偵測系統。在自動對焦系統方面，我們建置一三維空間之自動化取樣顯微平台，並且發展一個漸進式自動對焦演算法,以達到由電腦控制平台移動尋找焦點並且完成取像動作。在所提出自動對焦搜尋法的部份，我們使用不同取像解析度之漸進式對焦演算法，以減少在進行對焦時之時間消耗。整個搜尋演算法分為兩個部分：首先，我們使用低解析度的影像來進行二次函數逼近，找出一個近似的焦點位置。接著我們將輸入影像切換成高解析度影像，並使用二元搜尋法找到更精準的焦點位置。其中，對焦值的評估方式是採用Sum of modified Laplacian(SML)法計算出每張輸入影像的銳利度。經過實驗證明，我們提出的兩階段焦點搜尋演算法可以在短時間內找到準確的焦點位置。
在結核菌偵測與評估方面，我們提出了一套以色彩為基礎之自動化偵測結核菌系統，這個子系統在實做上分成兩大階段：結核菌區域偵測和候選區域分類。結核菌區域偵測的目的為將在顏色上與結核菌色彩分布相近的像素分割出來，其中包含了顯微影像亮度補償、影像色彩標準化、結核菌候選區域偵測和計算特徵參數等步驟。在影像前處理的部分，我們使用亮度補償的機制以解決顯微影像上因為光源不均勻照射產生的亮度分布不平衡問題。接著我們運用亮度變異程度對所有影像分成三大類，以避免因為顯微影像中細胞組織與背景的比例不同對後續處理的影響。而在後續處理中，針對每張影像所屬類別，分別訓練其所屬參數。接下來對每張影像進行色彩標準化，以降低在染色情況以及拍攝條件不同時所造成的顯微影像之色彩差異。接著我們使用了線性鑑別分析法(Linear Discriminant Analysis, LDA)以及主成分分析法(Principal Component Analysis, PCA)建構而成的色彩分類器進行分割。經過標記(Labeling)與型態學上的處理，結核菌候選區域即可被擷取出來，進而計算其特徵參數。而在結核菌候選區域分類的階段，為了解決之前研究在辨識時遭遇的偽陽性率過高的問題，我們使用了Ada-boost演算法自動挑選出辨識率高的弱分類器組合成一個強分類器，對輸入的特徵參數進行辨識。實驗結果顯示在使用了Ada-boost產生的強分類器之後，可以有效降低偽陽性率。

In this thesis, we proposed an automatic detection system for mycobacterium tuberculosis (TB) via coarse-to-fine auto-focusing and color-based segmentation. This fully automatic system can complete all the detection tasks, including focusing, detecting TB candidates and identification of TB bacilli. The system consisted of two sub-systems, auto-focusing system and mycobacterium tuberculosis identification system. In the auto-focusing system, a coarse-to-fine auto-focusing method was proposed based on a multi-resolution scheme which locates the focusing range by the quadratic polynomial fitting in low resolution and achieves the more accurate focal point by using binary search in high resolution. An algorithm was developed to identify the focal point under multiple image resolutions with the Sum of Modified Laplacian (SML) as the focusing measure. The main contribution of this sub-system was to obtain the initial focal range in low image resolution and refine the focal estimation in high image resolution and thus can reduce the auto-focusing time significantly. Comparing with the conventional binary search, the coarse-to-fine method needed only half of the time to accomplish auto-focusing. In the detection phase, we first used a lightening compensation step to reduce the effect of non-uniform brightness on microscopic images. Then, the microscopic images were classified into three types based on the variance of illumination. Next, a color normalization procedure was applied to reduce the color variation within the each type of images. From the normalized image, color features were extracted and evaluated by using a pixel classifier based on Linear Discriminant Analysis (LDA) and Principle Component Analysis (PCA) to detect the candidates of mycobacterium tuberculosis. By applying labeling and morphological methods, the corresponding features of each candidate could be extracted and input to the strong Ada-boost classifier - which was composed by several weak classifiers. In general, our system can automatically perform on-line mycobacterium tuberculosis detection automatically on the microscope stage with outstanding performance.

[1] A.S. deBoer, B. Blommerde, P.E.W. deHaas, M.M. G. Sebek, K.S. B. Lambregts-van Weezenbeek, M. Dessens, and D. van Soolingen, “False-Positive Mycobacterium tuberculosis Cultures in 44 Laboratories in Netherlands (1993 to 2000): Incidence, Risk Factors, and Consequences,” Journal of Clinical Microbiology, vol. 40, pp. 4004–4009, 2002.
[2] T. Hanscheid, C.M Ribeiro, H.M Shapiro, and N.G Perlmutter, "Fluorescence microscopy for tuberculosis diagnosis," The Lancet Infectious Diseases, vol. 7, pp. 236–237, 2007.
[3] V. H. Nguyen, N. S. Dinh, R. M. Anthony, F. G. Cobelens and D van Soolingen, "Fluorescence microscopy for tuberculosis diagnosis," The Lancet Infectious Diseases, vol. 7, pp. 238–240, 2007.
[4] 行政院衛生署,“結核病十年減半全民動員計畫,” 行政院衛生署, 2006。
[5] B.L. Stone, W.J. Burman, M.V. Hildred, E.A. Jarboe, R.R. Reves, and M.L. Wilson, “The Diagnostic Yield of Acid-Fast-Bacillus Smear-Positive Sputum Specimens,” Journal of Clinical Microbiology, vol. 35, no. 4, pp. 1030–1031, 1997.
[6] 呂喬洋、林碧芬, “建立標準結核菌實驗室,” 行政院衛生署疾病管制局九十一年度科技研究發展計畫研究報告, 2003。
[7] K. Veropoulos, C. Campbell, and G. Learnmonth, “Image processing and neural computing used in the diagnosis of tuberculosis,” IEE Colloquium on Intelligent Methods in Healthcare and Medical Applications, pp. 1–4, 1998.
[8] M.G. Forero, E.L. Sierra, J. Alvarez, J.L. Pech, G. Cristóbal, L. Alcalá, and M. Desco, “Automatic sputum color image segmentation for tuberculosis diagnosis, ” Proceedings of SPIE, vol. 4471, pp. 251–261, 2001.
[9] M.G. Forero, F. Sroubek, J. Alvarez, N. Malpica, G. Cristóbal, A. Santos, L. Alcala, M. Desco, and L. Cohen, “Segmentation, autofocusing and signature extraction of tuberculosis sputum images,” Proceedings of SPIE, vol. 4788, pp. 171–182, 2002.
[10] M.G. Forero, F. Sroubek, and G. Cristóbal, “Identification of tuberculosis bacteria based on shape and color,” Real-Time Imaging, vol. 10, no. 4, pp. 251–262, 2004.
[11] M.G. Forero and G. Cristóbal, “Automatic identification of Mycobacterium tuberculosis by Gaussian Mixture Models,” Journal of microscopy, vol. 223, no.2, pp. 120–132, 2006.
[12] M.G. Costa, C.F. Marly Filho, J.F. Sena, J. Salem, and M.O. de Lima, “Automatic identification of mycobacterium tuberculosis with conventional light microscopy,” IEEE Engineering in Medicine and Biology Society, pp. 382–385, 2008.
[13] N. OTSU, “A Threshold Selection Method from Gray-level Histogram,” IEEE Transactions on Systems, Man and Cybernetics, vol. 9, no. 1, pp. 62-66, 1979.
[14] E. Reinhard and M. Ashikhmin et al., “Color transfer between images,” IEEE Computer Graphics and Applications, vol. 21, no. 5, pp. 34–41, 2001.
[15] D.L. Ruderman, T.W. Cronin, and C.-C. Chiao, “Statistics of cone responses to natural images: implications for visual coding,” Journal of the Optical Society of America A, vol. 15, no. 8, pp. 2036–2045, 1998.
[16] T.Y. Zhang and C.Y. Suen, “A Fast Parallel Algorithm for Thinning Digital Patterns,” Communications of the ACM, vol. 27, pp. 236–239, 1984.
[17] R. C. Gonzalez and R. E. Woods, “Digital Image Processing,” Prentice Hall, 2002
[18] D.E. Rumelhart and J.L. McClelland, “Parallel Distribute processing: Exploration in the Microstructure of Cognition,” MIT Press Cambridge, MA, USA, vol. 1, 1986.
[19] A.P. Dempster, N.M. Laird, and D.B. Rubin, “Maximum Likelihood from Incomplete Data via the EM Algorithm”, Journal of the Royal Statistical Society. Series B (Methodological), vol. 39, no. 1, pp. 1-38, 1977.
[20] G. Mark, K. Nasser and R. H. Katie, “Low-Light Auto-Focus Enhancement for Digital and Cell-Phone Camera Image Pipelines,” IEEE Transactions on Consumer Electronics, vol. 53, no. 2, pp. 249-257, 2007.
[21] L. C. Ku, “Auto Focusing under Microscopic Views”, Proc. IEEE Workshop on Signal Processing Systems, 2007.
[22] Y. Yi, A. Besma, D. Narjes and A. Mongi, “Evaluation of Sharpness Measures and Search Algorithms for the Auto-Focusing of High Magnification Images,” Proc. SPIE, Vol. 6246, pp. 1-12, 2006.
[23] S. K, Nayar and Y., Nakagawa, “Shape from Focus: An Effective Approach for Rough Surfaces”, Proceedings of IEEE International Conference on Robotics and Automation, vol. 2, pp. 218-225, Cincinnati, USA, 1990.
[24] Y. C. Liu, F. Y. Hsu, H. C. Chen, Y. Y. Wang, and Y. N. Sun*, “A Coarse to Fine Auto-Focusing Algorithm for Microscope Image”, International Conference on System Science and Engineering, Macao, 2011.
[25] V. Makkaapati, R. Agrawl and R. Acharya, “ Segmentation of Tuberculosis Bacilli from ZN-stained sputum Smear Images”, the 5th Annual IEEE Conference on Automation Science and Engineering, Bangalore, 2009.
[26] J. Latin, D. Carroll, and P. E. Green, ”Analyzing Multivariate Data,” Duxbruy, 2003.
[27] A. C. Rencher, ”Multivariate Statistical Inference and Applications,” John Wily and Sons, 1998.
[28] R. Khutlang, S. Krishnan, A. Whitelaw, and T. S. Douglas, “Automated detection of tuberculosis in Ziehl-Neelsen stained sputum smears using two one-class classifiers,” J. Microscopy, vol. 237, pp. 96–102, 2010.
[29] Y.N Sun and M.H. Horng, “Assessing liver tissue fibrosis with an automatic computer morphometry system”, IEEE Engineering in Medicine and Biology, vol.16, no. 3, pp. 66-73, 1997.
[30] R. O. Duda, P. E. Hart, and D. G. Stork, Pattern Classification, New York: John Wiley & Sons, 2001.
[31] 章少謙, “應用色彩資訊於顯微影像之自動腫瘤細胞分割與參數統計”, 成功大學資訊工程所碩士論文, 2007。
[32] R. Khutlang, S. Krishnan, R. Dendere, A. Whitelaw, K. Veropoulos, G. Learmonth, and T.S. Douglas, “Classification of Mycobacterium Tuberculosis in Images of ZN-Stained Sputum Smears”, IEEE Trans. on Information Technology in Biomedicine, vol. 14, no. 4, 2010.
[33] 林展頤, “應用自動彩色顯微影像分割之結核菌偵測與評估”, 成功大學資訊工程所碩士論文, 2009。
[34] J. M. Tenenbaum, “Accommodation in computer vision”, Ph.D thesis, Stanford University, CA, USA, 1971.
[35] E. P. Krotkov, “Active Computer Vision by Cooperative Focus and Stereo”, New York: Springer-Verlag, 1989.
[36] R. A. Jarvis, “Focus optimization criteria for computer image processing”, Microscope, vol. 24, no. 2, pp. 163-180, 1976.
[37] T Acharya, W Metz , “Auto-focusing algorithm using discrete wavelet transform” , US Patent 6,151,415, 2000.
[38] N. K. Chern, P. A. Neow, and M. H. Ang, “Practical Issues in Pixel-Based Auto focusing for Machine Vision”, Proceedings of the IEEE International Conference on Robotics and Automation, vol. 3, pp. 2791-2796, 2001.
[39] G. S. Beveridge and R. S. Schechter, “Optimization: Theory and Practice”, McGraw-Hill, New York, 1970.
[40] J. Baina and J. Dublet, “Automatic Focus and Iris Control for Video Cameras”, the Fifth International Conference on Image Processing and its Applications, pp. 232-235, 1995.
[41] M. Moscaritolo, H. Jampel, F. Knezevich, R. Zeimer, “An Image Based Auto-Focusing Algorithm for Digital Fundus Photography”, IEEE Trans. on Medical Imaging, vol. 28, no. 11, 2009.
[42] A.M. Eskicioglu, P.S. Fisher, “Image quality measures and their performance”, IEEE Trans. on Communication, vol. 43, no. 12, 1995.
[43] M. Niemeijer, B. van Ginneken, J. Staal, M.S.A. Suttorp-Schulten, M.D Abramoff, “Automatic detection of red lesions in digital color fundus photographs”, IEEE Trans. on Medical Imaging, vol. 24, Issue 5, 2005.
[44] B. Zhanga, X. Wub, J. You, Qin Li, and F. Karraya, “Detection of microaneurysms using multi-scale correlation coefficients”, Pattern Recognition, vol. 43, Issue 6, 2010.
[45] I. T. Jolliffe, Principal Component Analysis, Springer-Verlag, pp. 487, 1986.
[46] R.A. Fisher, “The use of multiple measurements in taxonomic ”, Annals of Eugenics, vol. 7, pp. 179-188, 1936.
[47] F.Rosenblatt, The Perceptron: A Probabilistic Model for Information Storage and Organization in the Brain, Cornell Aeronautical Laboratory, Psychological Review, vol. 65, no. 6, pp. 386–408, 1958.
[48] Y.Freund and R.E. Schapire, “A decision-theoretic generalization of on-line learning and an application to boosting”, Journal of Computer and System Sciences, vol. 55, no. 1, pp.119--139, 1997.
[49] D.E. Rumelhart and J.L. McClelland, “Parallel Distribute processing: Exploration in the Microstructure of Cognition,” MIT Press Cambridge, MA, USA, vol. 1, 1986.
[50] P. A. Viola, M.J. Jones, “Rapid Object Detection using a Boosted Cascade of Simple Features”, Computer Vision and Pattern Recognition, pp.511-518, 2001.
[51] 呂金河, “多變量分析”, John Willey, 2005。
[52] I. Maglogiannis, CN. Doukas, “Overview of advanced computer vision systems for skin lesions characterization", IEEE Trans. on Information Technology in Biomedicine, vol. 13, no. 5, pp.721-33, 2009.
[53] K. Pearson, "On Lines and Planes of Closest Fit to Systems of Points in Space", Philosophical Magazine, vol. 2, no. 6, pp. 559–572, 1901.
[54] W. Huang and Z. Jing, “Evaluation of focus measures in multi-focus image fusion”, Pattern Recognition Letters, vol. 28, pp. 493-500, 2006.