臺灣博碩士論文加值系統

本論文目的是以數位影像處理方式來去除內視鏡鏡面反射，提供醫師一個更高品質的內視鏡影像，以利其對病症的處理。本論文提出適應性內視鏡鏡面反射偵測方法，根據灰階值分析，並利用形態學膨脹，準確地將鏡面反射區域標記。我們改良Image Inpainting 的演算法，使其適合於內視鏡影像之特性，並給予鏡面反射區域類別與各類別代表性修補參數。以此基礎，我們應用倒傳遞類神經網路進行鏡面反射區域的分類並修補。本論文成功的應用類神經網路自動選擇參數修補鏡面反射區域，並在實驗結果上皆比原Inpainting文獻方法佳，相信未來可進一步嘗試修補更多相關色彩影像。

In this thesis, our research goal is removal specular reflection in endoscopy using digital image processing techniques. We offer to a high-quality endoscopy image for doctor to do management of diseases in the Minimally Invasive Surgery (MIS). This thesis proposed adaptive detection method based on gray-scale value analysis and morphological dilation of specular reflection in endoscopy, as a result, the proposed method can accurately mask the specular reflection. We improved Image Inpainting algorithms. Each specular reflection area was given a category and representative parameters so that suited to the characteristics of the endoscopy image. As mentioned above we used back propagation neural network to classify the specular reflection area and specular reflection area will be inpainting. This thesis successfully applied neural network to adaptive selection parameters inpainting of specular reflection area. On experimental results, we proposed method better than original literature method and we believe that can be further attempts to inpainting more color image in the future.

摘要
ABSTRACT
誌謝
目錄
表目錄
圖目錄
第一章緒論
1.1 研究背景
1.2 研究動機與目標
1.3 論文主要貢獻
1.4 章節概述
1.5 實驗平台規格
第二章 背景簡介與文獻回顧
2.1 國外微創手術內視鏡影片研究文獻評述
2.2 國內微創手術內視鏡鏡面反射修補研究文獻評述
2.3 影像修補INPAINTING的相關研究文獻
2.4 結論 6
第三章 內視鏡鏡面反射觀察分析與偵測標記
3.1 內視鏡鏡面反射觀察與分析
3.1.1 內視鏡鏡面反射變化情況與灰階值分佈分析
3.1.2 除霧技術
3.1.3 直方圖等化技術
3.1.4 除霧技術與直方圖等化技術實驗於鏡面反射結果與分析
3.2 PROPOSED 內視鏡鏡面反射偵測標記演算法架構
3.2.1 二值化內視鏡鏡面反射區域
3.2.2 鏡面反射區域膨脹-運用形態學方法
3.3 內視鏡鏡面反射區域標記實驗成果
3.4 本章結論
第四章 檢驗整合現有影像修補INPAINTING演算法
4.1 影像修補INPAINTING
4.1.1 影像修補INPAINTING簡介
4.2 CRIMINISI EXEMPLAR-BASED IMAGE INPAINTING演算法系列探討
4.2.1 CRIMINISI影像修補INPAINTING研究探討
4.2.2 ANUPAM影像修補INPAINTING研究探討
4.2.3 LIXIN YIN影像修補INPAINTING研究探討
4.2.4 各文獻修補演算法實驗於內視鏡鏡面反射結果與討論
4.3 MODIFIED EXEMPLAR-BASED IMAGE INPAINTING演算法
4.4 調整參數修補INPAINTING演算法與原各文獻方法比較
4.5 調整參數執行影像修補INPAINTING缺點
4.6 本章結論
第五章 自適應鏡面反射區域分類-運用類神經網路
5.1 運用連通區域法自動LABELING各個鏡面反射區域
5.2 鏡面反射區域類別說明
5.3 鏡面反射區域外部特徵擷取方法
5.3.1 鏡面反射外部特徵擷取範圍
5.3.2 灰階值標準差偏度峰度特徵
5.3.3 CANNY EDGE數量特徵
5.3.4 任意鏡面反射形狀HOG特徵
5.4 倒傳遞類神經網路訓練測試階段
5.4.1 訓練階段
5.4.1.1 輸入視訊規格與擷取影像數量相關類神經網路資訊
5.4.1.2 人工分類鏡面反射區域與外部特徵擷取資料庫說明
5.4.1.3 倒傳遞類神經網路創建網路描述與訓練流程
5.4.2 測試階段
5.4.2.1 鏡面反射區域測試類別資料庫數量說明
5.4.2.2 倒傳遞類神經網路測試步驟流程與準確率
5.5 本章結論
第六章 實驗結果與性能評估
6.1 運用類神經自適應選擇參數修補實驗結果
6.2 內視鏡影像修補INPAINTING主客觀評估方式
6.2.1 客觀SSIM評估方式
6.2.2 主觀評估方式
6.3 延伸到一般彩色影像修補實驗結果
6.3.1 一般彩色影像修補INPAINTING主客觀評估方式
6.3.2 客觀SSIM評估方式
6.3.3 主觀評估方式
6.4 本章結論
第七章 結論與未來展望
7.1 結論
7.2 未來研究方向
參考文獻
附錄

[1]Tchoulack, S., J.M. Pierre Langlois, and F. Cheriet, “A Video Stream Processor for Real-time Detection and Correction of Specular Reflections in Endoscopic Images,” in Proc. Circuits and Systems and TAISA Conference, pp.49-52 June 2008.
[2]M. Bertalmio, A.L. Bertozzi and G. Sapiro, “Navier-stokes, fluid dynamics, and image and video inpainting,” in Proc. the IEEE Conference on Computer Vision and Pattern Recognition, vol. 1 , pp. I-355- I-362,2001.
[3]G. Karapetyan and H. Sarukhanyan, “Automatic detection and concealment of specular reflections for endoscopic images,” in Computer Science and Information Technologies (CSIT), 2013, pp. 1–8. Sept 2013.
[4]T. Stehle, “Removal of specular reflections in endoscopic images”, Journal of Advanced Engineering, pp. 32-36, 2006.
[5]Kai-Tai Song and Chun-Ju Chen, “Autonomous and Stable Tracking of Endoscope Instrument Tools with Monocular Camera,” in Proc. IEEE International Conference on Advanced Intelligent Mechatronics, pp.39-44, Kaohsiung, Taiwan, 11-14 July 2012.
[6]Chun-Ju Chen, Wayne Shin-Wei Huang, and Kai-Tai Song, “Image Tracking of Laparoscopic Instrument Using Spiking Neural Networks,” in Proc. 2013 13th International Conference on Control, Automation and Systems (ICCAS 2013 ), pp.951-955, Gwangju, Korea Oct. 20-23, 2013.
[7]Jungle Chi-Hsiang Wu, Guo-Shiang Lin, Hsiao-Ting Hsu, You-Peng Liao, Kai-Che Liu, and Wen-Nung Lie, “Quality enhancement based on retinex and pseudo-HDR synthesis algorithms for endoscopic images,”in Proc. IEEE Conference on Visual Communications and Image Processing (VCIP), pp. 1-5, 2013.
[8]Yen-Chin Lee, “Assisted 3D Display System for Endoscopy,” Master thesis of National Yang-Ming University, July 2012.
[9]A. Criminisi, P. Perez, and K. Toyama, “Region filling and object removal by exemplar-based image inpainting,” IEEE Trans. Image Process., vol. 13, no. 9 pp. 1200–1212, Sep. 2004.
[10]Anupam, P. Goyal,S. Diwakar, “Fast and Enhanced Algorithm for Exemplar Based Image Inpainting,” in Proc. 2010 Fourth Pacific-Rim Symposium on Image and Video Technology (PSIVT), pp. 325-330, 2010.
[11]Lixin Yin and Chen Chang, “An Effective Exemplar-based Image Inpainting Method” in 2012 IEEE 14th International Conference on Communication Technology (ICCT), pp. 739 – 743, 2012.
[12]Shu-Chiang Chung, Ta-Wen Kuan, Chuan-Pin Lu and Hsin-Yi Lin, “A New Approach of Image Inpainting Based on PSO Algorithm,” in Proc. 2013 International Conference on ICOT, pp.205-209, 2013.
[13]B.Wohlberg, “Inpainting With Sparse Linear Combinations Of Exemplars,” in Proc. IEEE International Conference on ICASSP, pp. 689-692, 2009.
[14]R.Gomathi and A.V.A. Kumar. “An Efficient GEM Model for Image Inpainting Using a New Directional Sparse Representation: Discrete Shearlet Transform,” in Proc. IEEE International Conference on Computational Intelligence and Computing Research (ICCIC), pp. 1- 4, 2010.
[15]Jingang Shi and Chun Qi. “Sparse Modeling Based Image Inpainting With Local Similarity Constraint,” in Proc. IEEE International Conference on (ICIP), pp. 1371-1375, 2013.
[16]Lin, C.W. and Cheng, N.C. “Video background inpainting using dynamic texture synthesis,” in Proc. IEEE International Symposium on Circuits and Systems (ISCAS), pp.1559 – 1562, 2010.
[17]Guo-Shiang Lin, Chia-Yung Hsieh, and Wen-Nung Lie, “Sprite Generation For Hole Filling In Depth Image-Based Rendering,” in Proc. IEEE Conference on Visual Communications and Image Processing (VCIP), pp.5402-5406, 2014.
[18]K. He, J. Sun, and X. Tang. “Single image haze removal using dark channel prior,” in Proc. IEEE Conference on CVPR, pp.1956-1963, 2009.
[19]Acharya and Ray. “Principles and Applications,” Wiley-Interscience 2005, ISBN 0-471-71998-6.
[20]繆紹綱(2008)，Rafael C.Gonzalez, Richard E.Woods, Steven L.Eddins.原著，『數位影像處理-運用MATLAB』"Digital Image Processing Using MATLAB"，台北，台灣培生教育、台灣東華書局合作出版，pp. 364- 365.
[21]Zhou Wang, A.C. Bovik, H.R. Sheikh and E.P. Simoncelli. “Image Quality Assessment From Error Visibility to Structural Similarity”, in Proc. IEEE Transactions, pp.600-612, 2004.
[22]M. ertalmio, G. Sapiro, V. Caselles and C. Ballester. “Image Inpainting,” in Proc. ACM Conf. Comp. Graphics (SIGGRAPH), pp. 417- 424, 2000.
[23]沈岱範.李仲明, “數位化老舊膠卷影像斑點及刮痕的修復研究”, 2014 全國電信研討會, 亞洲大學, 2014.
[24]Mao-Cheng,Chen, “A GA-based Feature Selection for Breast Lumps under Various Conditions,” Master thesis of National Yunlin University of Science & Technology, July 2014.
[25]沈岱範.李柏漢, “應用消失點及深度模型產生二維影像之深度圖”, 2014 全國電信研討會, 亞洲大學, 2014.
[26]N. Dalal and B. Triggs, “Histograms of oriented gradients for human detection,” in Proc. IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pp. 886–893, 2005.
[27]Rumelhart, D.E., Hinton, G.E. and Williams, R.J. “Learning internal representations by backpropagating errors,” Nature, pp.533–536, 1986.
[28]Back Propagation Neural Network 倒傳遞類神經網路http://www.dotblogs.com.tw/dragon229/archive/2013/01/23/88750.aspx