內視鏡影像病兆之辨識與分析主要的目的就是利用計算機高速運算的能力，將原本經由廣角鏡頭取得之扭曲影像，透過修正程式與整合影像處理的方法技術和學問。提供內視鏡病兆影像病兆區域辨識之資訊。
本研究主要針對內視鏡影像，建立一完整影像處理過程包括：彩色模型轉換之探討、影像直方圖運算、影像濾波器選用、邊緣檢測、區域成長、二值影像轉換與影像形態學之應用。於影像前處理之過程中建立具有明確差異性質之區域，供影像分割以形成簡單清楚之邊界，可供內視鏡影像病兆之辨視。
本研究所有的原理探討均套用待處理之內視鏡影像，經由電腦視覺之檢測，獲取最佳之影像處理步驟，以正確辨識病兆。供醫師用藥之參考。

The main purpose of “the Identification and Analysis of Lesion in Endoscopic Image” is to provide information concerning identification of endoscopic lesions, by making use of the high speed computing capability of computers and taking the distorted images from wide angle lens, through adjusting programs and incorporating image processing techniques and knowledge.
The main purpose of “the Identification and Analysis of Lesion in Endoscopic Image” is to make use of the high speed computing capability of computers, take the distorted images from wide angle lens , through adjusting programs and incorporating image processing techniques and knowledge, and provide information concerning identification of endoscopic lesions.
The research concentrates on endoscopic images, establishing a complete image managing process which includes investigation of color model transformation, counting of histogram, selection of image filter, edge detect , region grow , conversion of binary image, and the application of morphology. By setting up regions with distinct differences before the image managing process, images are divided and a clear-cut boundary is formed to provide identification of lesions in endoscopic images.
All the images used in this research are unprocessed images. Through the examination of computer vision and premium image processing procedures, lesions are correctly identified and used as references for doctors’ prescriptions.
Key words: histogram, image filter, edge detection, region grow, binary image , morphology

第一章　緒論
1.1 研究動機．．．．．．．．．．．．．．．．1
1.2 研究方法．．．．．．．．．．．．．．．．2
1.3 論文結構．．．．．．．．．．．．．．．．4　
第二章　內視鏡介紹與圓桶效應修正
2.1 圓桶效應修正法則．．．．．．．．．．．．6
第三章 影像處理
3.1 彩色影像模型．．．．．．．．．．．．．．9
3.1.1 RGB彩色模型 ．．．．．．．．．．．9
3.1.2 YIQ 彩色模型．．．．．．．．．．．11
3.1.3 H IS 彩色模型 ．．．．．．．．．．14
3.2 直方圖處理．．．．．．．．．．．．．．．16
3.2.1直方圖的運算 ．．．．．．．．．．．18
3.3 空間濾波．．．．．．．．．．．．．．．．23
3.3.1迴旋積與相關 ．．．．．．．．．．　24
3.3.2線性濾波．．．．．．．．．．．．． 26
3.3.3中值濾波．．．．．．．．．．．．． 31
第四章 影像分割
4.1邊緣檢測的概念．．．．．．．．．．．．． 35
4.1.1 Sobel運算子邊緣檢測．．．．．．．36
4.1.2 Canny運算子邊緣檢測．．．．．．．38
4.1.3 Laplacian 運算子邊緣檢測．．．． 38
4.2區域成長．．．．．．．．．．．．．．．． 40
第五章 二值影像處理
5.1 像素的近鄰．．．．．．．． ．． 43
5.2 連通性．．．．．．．．．．．． 44
5.3 臨界值
5.3.1 人工經驗臨界值法．．．．．47
5.3.2 峰值法．．．．．．．．．．47
5.3.3 谷值法．．．．．．．．．．48
5.3.4 最佳臨界．．．．．．．．．48
5.4 形態學
5.4.1 基本運算．．．．．．．．．51
5.4.2 膨脹．．．．．．．．．．．52
5.4.3 斷開與閉合．．．．．．．．53
5.4.4 邊界提取．．．．．．．．．54
5.4.5 區域埴充．．．．．．．．．54
第六章 實驗結果
6.1實驗影像．．．．．．．．．．．．．56
6.2實驗流程．．．．．．．．．．．．．58
6.3實驗結果．．．．．．．．．．．．．58
第七章 結論．．．．．．．．．．．．．．．．71
參考文獻．．．．．．．．．．．．．．72

參考文獻
【1】 Margulies C. Krevsky B,Catalano M:”How accurate are endoscopic estimates of size?” Gastrointest Endosc.,vol 40: pp. 174-177,1994.
【2】 林志鴻，民國九十年。內視鏡影像之扭曲校正與病兆面積之量測，大葉大學電機工程學系研究所，彰化。
【3】 Sonnenberg A, Giger M, Kern L, et al.:”How reliable is determination of ulcer size by endoscopy ? “ Br .Med.J.,vol. 2:pp.1322-1324,1979.
【4】 Gonzalez,R. C. and Wood,R. E. Digital Image Processing .Addison Wesley,Reading,MA.[1992].
【5】 D. Androutsos,K. N. Plataniotis,and A.N. Venetsanopoulos,”A Novel Vector-Based Approach to Color Image Retrieval Using a Vector Angular-Based Distance Measure,” Computer Vision and Image Understanding,Vol75 ,Nos.1/2,July/August,pp.46-58,1999.
【6】 Milan Sonka,Vaclav Hlavac, Roger Boyle “Image Processing ,Analysis,and Machine Vision “ ITP pp.57-62 ,1993
【7】 D. Androutsos,K. N. Plataniotis,and A.N. Venetsanopoulos,”A Novel Vector-Based Approach to Color Image Retrieval Using a Vector Angular-Based Distance Measure,” Computer Vision and Image Understanding,Vol75 ,Nos.1/2,July/August,pp.46-58,1999.
【8】 Image Processing Toolbox For Use with MATLAB，The MATH WORKS Inc.
【9】 數位影像處理活用-Matlab。全華，謬紹綱【1999】
【10】 Haralick,R.M. and Shapiro,L. G. urvey: Image Segmentation,”Computer,Vision,Graphics,Image Processing, vol.29,pp.100-132 [1980]
【11】 數位信號-影像與語音處理。全華，林宸生。【1997】
【12】 Lim,Jae S.Two-Dimensional Signal and Image Processing. Englewood Cliffs,NJ:Prentice Hall,1990.pp.469-476
【13】 Haralick,R.M. and Shaprio,L. G. Computer and Robot Vision,vols.Ⅰ and Ⅱ.Addison Wesley,Reading,MA.[1993]
【14】 Phillips.D.Image Processing in C.Prentice-Hall,Englewood Cliffs[1994]
【15】 Canny,John. ”A Computational Approach to Edge Detection,” IEEE Transactions on Pattern Analysis and Machine Intelligence ,1986.Vol.PAMI-8,No.6,pp. 679-698
【16】 蔡騰興，民國八十九年。影像分割技術之深入探討及摭拾，國立台灣海洋大學電機工程學系，台北。
【17】 Matlab 程式設計與應用。清蔚科技2000，張智星。