基於人臉五官特徵之唐氏症辨識｜國立清華大學博碩士論文庫

簡易檢索 / 詳目顯示

回結果列表

研究生：	彭國瑋 Guo-Wei Peng
論文名稱：	基於人臉五官特徵之唐氏症辨識 Down Syndrome Recognition Based on 2D Facial Features
指導教授：	陳永昌 Yung-Chang Chen 謝凱生 Kai-Sheng Hsieh
口試委員:
學位類別：	碩士 Master
系所名稱：	電機資訊學院 - 電機工程學系 Department of Electrical Engineering
論文出版年：	2008
畢業學年度：	96
語文別：	英文
論文頁數：	83
中文關鍵詞：	唐氏症、五官特徵、學習演算法、染色體異常、自動輪廓切割、主值分析
外文關鍵詞：	Down Syndrome, Facial Feature, Adaboost, Abnormal Chromosome, Active Contour Segmentation, Principal Component Analysis
相關次數：	點閱：3 下載：0
分享至:	分享至facebook 分享至twitter

查詢本校圖書館目錄查詢臺灣博碩士論文知識加值系統勘誤回報

摘要:
目前醫生對於罕見疾病患者的診斷，最準確的方法是採用染色體檢驗，但是，這需要昂貴的金錢，以及冗長的時間去做染色體的DNA分析• 在這篇論文的工作裡，我們希望建立一個僅僅需要人臉資訊的疾病辨識系統，來幫助醫生做最簡易、快速的疾病診斷•
某些染色體上的基因異常，常會造成五官或是身體上的畸變，我們可藉由某些特徵來做疾病的辨識•因為這是一個新的研究方向，為了驗證這個想法是否可行，初步的工作，我們一開始先將問題縮小在幼童的唐氏症辨識•
在新生兒出生時，大約會有1/500 的機會，因為染色體異常，而導致新生兒生長畸變，這些變異有時可直接從五官看出，例如 : 五官比例特殊、鼻子太扁、耳朵外觀怪異、臉頰過於肥厚扁平•我們利用人臉五官特徵而發展出一套醫療檢測系統，幫助醫生對於新生兒做疾病的臨床診斷•唐氏症又稱Trisomy 21，此種疾病是由於存在了三條第21號染色體•
在這個疾病辨識系統裡，我們針對正面及側面的兩張人臉，設計了15個重要臉部特徵，其中8個特徵是從正面擷取，剩下7個是從側面，分別針對兩眼的相對位置、兩眼大小、眼睛的傾斜程度、鼻子大小、鼻子的凹陷程度、耳朵與鼻子的相對位置、耳朵的形狀比例，並且提出了快速搜尋這些特徵的方式• 實驗結果的顯示，在做唐氏症患者的辨識上，我們具有9成以上的成功辨識率•

Abstract
In recent years, doctors adopt the most accurate method which is called “The examination of the chromosome” for diagnosis of rare diseases. But it costs much and long time to analyze the DNA on the chromosome. In the work of this thesis, we propose the disease recognition system that uses the information of the facial features to help doctors diagnose simply and rapidly.
Since the specific abnormality of the chromosome often affects the face or body as to be deformed, we can recognize the disease by the specific facial deformation features. Since this study is a novel research topic, in order to prove the feasibility of the idea, first of all, we begin to focus on the recognition of the Down syndrome in the infant faces.
When a baby is born, there is 1/500 possibility to be deformed because of the abnormal chromosome. Sometimes these differences can discern from the facial features, for instance, strange ratio of the facial feature, flat nose, strange contour of the ear, plump chin and so on. We develop the system of the medical diagnosis by the facial features to help doctors make the clinic diagnosis of the disease of infant babies. Down-Syndrome, also known as Trisomy-21 due to the presence of a third twenty-first chromosome, is one of the most common and well known birth anomalies.
In our disease recognition system, we consider fifteen important facial features from the front face and lateral face. Eight of which are extracted from the front face, and the after seven features are extracted from the lateral face. They stand for the relative position of the both eyes, size of the both eyes, the slope degree of the eye, the size of the nose, the concave degree of the nose, the relative position between ear and nose, and the shape ratio of the ear. We design a fast algorithm for feature recognition effectively. Experimental results show that the accuracy of our system to recognize Down-Syndrome case correctly is about 90%.

Abstract    i
Table of Contents    ii
List of Figures    v
List of Tables    x
Chapter 1 Introduction 1
        1.1  Background 1
        1.2 Motivation 4
        1.3 Thesis Organization 5
Chapter 2 Facial Features 6
    2.1 Preview 6
    2.2 The Characteristic of the Down Syndrome in
             the Facial Features 7
    2.2.1 Brushfield Spots 7
    2.3 Facial Features 9
    2.3.1 Feature (1~2) 10
2.3.2 Feature (3~4) 11
2.3.3 Feature (5~8) 12
2.3.4 Feature (9~10) 14
2.3.1 Feature (11~13) 15
2.3.1 Feature (14) 16
2.3.1 Feature (15) 17
Chapter 3 Feature Extraction 18
    3.1 System Flow 18
    3.1.1 Flowchart for extracting the frontal
               features 18
3.1.2 Flowchart for extracting the lateral features 19
    3.2 Face Detection     20
    3.3 The Eye Corner Detection 24
    3.4 Segmentation of the Eye Contour    25
3.4.1 Level Set Evolution Without Re-initialization 25
3.4.2 A Threshold Selection Method from Gray-Level
        Histograms 27
3.4.3 Combination of the active contour segmentation of
      level set evolution and a threshold selection method
      from gray-level histograms 28
3.5 Centroid of the Eye 32
3.6 Pupil & Iris Extraction    33
3.7 Principal Component Analysis 36
3.8 The Marking of the Eye Corner 38
3.9 The Nose Tip Detection in the Frontal Face 41
3.10 Nose Extraction in the Lateral Face 43
3.10.1 Removal of Eye 43
3.10.2 Extraction of nose features 45
3.11 Ear Extraction in the Lateral Face 48
3.11.1 Scale-Invariant Feature Transform 48
3.11.1.1 Detection of Scale-Space Extrema 48
3.11.1.2 Orientation Assignment 50
3.11.1.3 Generation of keypoint descriptors 51
3.11.1.4 Implementation of SIFT 51
3.11.2 Fourier Descriptor 55
3.11.3 Extraction of the Outer Contour of the Ear 57
Chapter 4 Classification system - AdaBoost 60
4.1 Boosting 60
4.2 AdaBoost 61
4.3 Classification and Regression Trees (CART) 65
Chapter 5 Experimental Results 67
5.1 Feature Extraction 71
5.2 Training Results of AdaBoost 73
5.3 Receiver Operating Characteristics 77
5.3 Area Under ROC Curve 78
Chapter 6 Conclusion & Future Works    80
References 82

                                

References
[1] White, M.J.D, "The chromosome," 6th , Chapman & Hall, London. p28. 1973.
[2] Sieben, V. J., Debes-Marun C. S., Pilarski P. M., Kaigala G. V., Pilarski L. M., and Backhouse C., "FISH and chips: chromosomal analysis on microfluidic platforms," IET Nanobiotechnology, Vol. 1(3), pp. 27-35, June 2007.
[3] Sanez, R.B, "Primary care of infants and young children with down syndrome," American Family Physician, Vol. 59 (2), Jan. 1999.
[4] Kim, J.H., "Characteristic ocular findings in Asian children with Down syndrome," Nature 16 (6), Nov. 2002.
[5] Kovac, J., Peer, P., Solina, F., "Human skin color clustering for face detection, "EUROCON2003, 2003.
[6] Chunming Li, Chenyang Xu, Changfeng Gui, and Martin D. Fox, "Level Set Evolution Without Re-initialization: A New Variational Formulation," IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), Vol. 1, pp. 430-436, San Diego, 2005.
[7] Otsu, N., "A Threshold Selection Method from Gray-Level Histograms," IEEE Transactions on Systems, Man, and Cybernetics, Vol. 9, No. 1, pp. 62-66, 1979.
[8] Libor Masek, Peter Kovesi. MATLAB Source Code for a Biometric Identification System Based on Iris Patterns. The School of Computer Science and Software Engineering, The University of Western Australia. 2003.
[9] Soille, P., "Morphological Image Analysis: Principles and Applications, "Springer-Verlag, pp. 173-174, 1999.
[10] Pearson, K."On Lines and Planes of Closest Fit to Systems of Points in Space," Philosophical Magazine, Vol. 2, 6, pp.559–572.
[11] Lowe, D. G., "Distinctive Image Features from Scale-Invariant Keypoints," International Journal of Computer Vision,Vol. 60, 2, pp. 91-110, 2004.
[12] Zahn, C.T., Roskies, R.Z., "Fourier descriptors for plane close curves," IEEE Trans. Computers, Vol. C-21, pp. 269-281, Mar. 1972.
[13] Demiriz, A., Bennett, K.P., Shawe-Taylor, J., "Linear Programming Boosting via Column Generation," Kluwer Machine Learning, Vol. 46, pp. 25–254, 2002.
[14] Freund, Y., "An adaptive version of the boost by majority algorithm," Machine Learning, Vol.43, 3, pp.293-318, June 2001.
[15] Domingo, C., Watanabe. O., "MadaBoost: A Modification of AdaBoost," Morgan Kaufmann, pp.180-189, 2000.
[16] Friedman, J., Hastie, T., Tibshirani, R., "Additive logistic regression: a statistical view of boosting," Annals of Statistics, Vol. 28(2), pp.337-407, 2000.
[17] Zhang, T., "Convex Risk Minimization," Annals of Statistics, 2004.
[18] Breiman, L., Friedman, J. H., Olshen,R. A., and Stone, C. J., "Classification and Regression Trees," Statistics/Probability Series. Wadsworth Publishing Company, Belmont, California, U.S.A., 1984.
[19] Fawcett, T., "An introduction to ROC analysis," Pattern Recognition Letters, 27, pp.861-874, 2006.

全文公開日期本全文未授權公開 (校內網路)
全文公開日期本全文未授權公開 (校外網路)

簡易檢索 / 詳目顯示

相關論文