問、聞、望、切四診為中醫作為疾病辯證之依據，而望診居為四診之首，其中舌診為望診中首要項目。中醫透過舌診可瞭解人體臟腑健康，辨别病症深淺，判斷病勢輕重變化，但往往舌診判讀結果易囿於知識、經驗、思維模式、診斷技巧、對顏色感知及詮釋等主觀判定影響，不同醫生對同一舌象可能做出互異判斷，重複性不佳；再者，中醫以八綱辨證闡述生理概況，由人體所展現之病徵概觀身體疾病，多無法明確診斷單一病症，臨床上對特定疾病亦缺乏中醫客觀判斷指標。近幾十年，傳統中醫舌診逐漸與現代化科學結合，隨著電腦科學及網路技術之發展與應用日益普及，中醫舌診已朝向現代化、資訊化及標準化發展。過去自動化舌診系統(ATDS)已透過科學方法，研發具標準化判別程序之系統，經科學、數據化方式輔助醫師診斷，提升診查結果之一致性、可靠性，使醫師能及早發現疾病與對症下藥。
為提升舌正面特徵擷取準確性，本研究在自動化舌診系統之基礎上加以改進，透過高析解度之拍攝裝置拍攝受試者舌部影像，經色彩校正與色彩補償技術統一影像品質；而後，經由分析舌頭、嘴唇及皮膚部位之RGB色彩分量表現，轉換為易於人眼認知之HSI色彩空間，依色調、飽和度及亮度特性，利用移除皮膚、矩形檢測、牙齒區域與口腔暗區檢測、控制點偵測後，使用主動輪廓技術擷取舌體，之後藉由型態學、臨界值法、連通物件標記等技術，依序擷取津液、苔色、舌色、舌苔厚薄、舌形、裂紋、瘀點、朱點及齒痕特徵，最後依五個臟器分區，進行量化特徵之數目、面積或長度，並利用統計分析結果產生舌診報告表，提供中醫師作為臨床上診斷之參考。

The diagnosis of TCM methods is generally based on four standard but not validated approaches, that is, inquiry, smelling/listening, observation, and palpation. During the process of observation, the diagnosis of the tongue is one of the crucial diagnosing steps. TCM diagnoses to understand human body internal organs health and recognize a disease depth through the tongue. But usually as a result, the diagnosis tends to be limited by knowledge, experience, train of thought, and diagnostic techniques. The subjective determination is likely to be affected by the doctor’s color sensitivity and interpretation. Different doctors may come to drastically different judgments on the same tongue presentation with little overlap. Moreover, doctors explain the physiological profile by eight principles which can’t clearly diagnose a single disease. In clinical research of specific diseases is a lack of objective TCM indices. Over the past decade, TCM tongue gradually combined with modern science. The development and application of computer science and network technology is increasingly popular, tongue inspection has been toward modernization, technology and standardization. In the past, ATDS has come up with an assisting system with standardized judging procedures by scientific methods in order to help doctors diagnose in a scientific and systematic way to boost the consistency and reliability of diagnosing,as well as to help find disease and give therapy at the first moment.
In order to improve the extracted accuracy of the tongue surface feature extraction, this research to be improved on the basis of ATDS. System can take the patient's tongue through the high-resolution camera, and unify the image quality by technique of color correction and color compensation. Then, the original tongue images go through HSI color space conversion, detection of the control points inside and outside the surface of the tongue, curve smoothness modification and active contour model to capture images of the tongue. After that, the system will sequentially capture fluid, moss color, tongue color, tongue thickness, tongue shape, cracks, petechiae, red tip of tongue and tooth mark by morphology, statics threshold and connected component technology. Finally, according to five viscera partition, the system can quantify the number, area or length of features. The statistical analysis report form the system, which can provide a diagnosis of Chinese medicine as a clinical reference.

論文審定書 i
論文公開授權書 ii
中文摘要 iii
Abstract iv
目錄 v
圖目錄 vii
表目錄 viii
第 一 章 緒論 1
1.1 中醫舌診辯證 1
1.2 舌診特徵 2
1.3 自動化舌診系統 3
第 二 章 相關研究 6
2.1 HSI色彩模型 6
2.2 型態學 7
2.3 連通物件標記 8
2.4 舌診腑臟分劃法 9
第 三 章 研究方法 11
3.1 取像設備 11
3.2 影像預處理 13
3.3 特徵擷取 14
3.3.1 津液 15
3.3.2 舌色 16
3.3.3 苔色 18
3.3.4 舌苔厚薄度 20
3.3.5 舌形 21
3.3.6 裂紋 23
3.3.7 瘀點 25
3.3.8 朱點 27
3.3.9 齒痕 28
第 四 章 實驗結果 30
第 五 章 結論與未來展望 42
參考文獻 43
附錄一、機構同意人體研究/人體試驗證明書 45
附錄二、ATDS收錄之可能病徵斷語 46

[1] Lun-chien Lo, Mark Chun-cheng Hou, Ying-ling Chen, John Y. Chiang and Jeng-jiun Shyu, "Automatic Tongue Diagnosis System," The 2nd Int'l Conf. on BioMedical Engineering and Informatics (BMEI'09), pp. 140-144, Tianjin, China, Oct. 17-19, 2009.
[2] 沈淑敏、葉美玲，“中醫舌診介紹，”慈濟護理雜誌，第5卷，第6期，頁73-78，2006。
[3] 巢元方，《諸病源侯論》上冊，第一版。北京，人民衛生出版社，1980，頁389。
[4] 梁嶸，“1949年以前中醫舌診學術發展歷程的研究，” 自然科學史研究，第23卷，第3期，頁257-273，2004。
[5] 羅綸謙，蔣依吾，"臨床望舌彩色圖解，" April 2010。
[6] Yu Cheng Hsu, Ying-ching Chen, Lun chien Lo and John Y. Chiang, "Automatic Tongue Feature Extraction," Proc. Conf. Int'l Computer Symp. (ICS2010), pp 936-941, Tainan, Taiwan, Dec. 16-18 2010.
[7] John Y. Chiang, Lun-chien Lo, Wen-jiuan Chen and Yu-Cheng Hsu, "Computerized Tongue Feature Extraction for Traditional Chinese Medicine," Proc. The 15th Int'l Congress of Oriental Medicine, pp. 112, Chiba, Japan, Feb. 26-28, 2010.
[8] Lun-Chien Lo, Tsung-Lin Cheng, Wen-Jiuan Chen, Yung-Fu Chen and *John Y. Chiang, "The Study on the Agreement between Automatic Tongue Diagnosis System and Traditional Chinese Medicine Practitioners," Evidence-Based Complementary and Alternative Medicine, Vol. 2012 (2012).
[9] Lun-chien Lo, Tsung-Lin Cheng, *John Y. Chiang and Natsagdorj Damdinsuren, "Breast Cancer Index - A Perspective on Tongue Diagnosis in Traditional Chinese Medicine," Journal of Traditional and Complementary Medicine, Vol. 3, No. 3, pp. 194-203, 2013.
[10] Pei-Yung Liao, Po-Chi Hsu, Jia-ming Chen, John Y. Chiang, Lun-Chien Lo, "Diabetes with Pyogenic Liver Abscess - A Perspective on Tongue Assessment in Traditional Chinese Medicine," Complementary Therapies in Medicine,Volume 22, Issue 2, pp. 341-348, Apr. 2014.
[11] Lun-Chien Lo, Chia-Yun Chen, John Y. Chiang, Tsung-Lin Cheng, Hong-Jen Lin and Hen-Hong Chang, "Tongue Diagnosis of Traditional Chinese Medicine for Rheumatoid Arthritis," Afr. J. Tradit. Complement Altern. Med., Vol. 10, Issue 5, pp. 360-369, 2013.
[12] 李乃民等，《中國舌診大全》。北京：學苑出版社，1995，頁 1-525。
[13] 王季藜等，《舌診源鑑》。台北：立得出版社，1993，頁 2-15。
[14] 洪禎徽，《舌診》。台北：立得出版社，1996。
[15] 劉華生，《中醫診斷學》。北京：清華大學出版社，2003，頁53。
[16] 許家佗、張志楓、任宏福、屠立平、張利、陸璐明，“一種舌體胖瘦分析的圖像診斷方法，” 中國中西醫結合影像學雜誌，第7卷，第6期，頁407-409，2009年12月。
[17] Laura Liu, David Zhang, and Jane You, "Detecting Wide Lines Using Isotropic Nonlinear Filtering, " IEEE Transactions on Image Processing, Vol. 16, No. 6, pp. 1584-1595, June 2007.