請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/23334
完整後設資料紀錄
DC 欄位 | 值 | 語言 |
---|---|---|
dc.contributor.advisor | 宋孔彬(Kung-Bin Sung) | |
dc.contributor.author | Ting-Jia Chang | en |
dc.contributor.author | 張庭嘉 | zh_TW |
dc.date.accessioned | 2021-06-08T04:59:21Z | - |
dc.date.copyright | 2010-08-20 | |
dc.date.issued | 2010 | |
dc.date.submitted | 2010-08-18 | |
dc.identifier.citation | [1] BLODI, F. C., and ALLEN, L., 'Stereoscopic Manual of the Ocular Fundus in Local and Systemic Disease,' 1964
[2] Azriel Rosenfeld, “Picture Processing by Computer,” New York: Academic Press, 1969. [3] Samuel J. Biondo, “Fundamentals of expert systems technology: principles and concepts,” p. 1, New Jersey, ISBN 0-89391-701-X, 1990. [4] L. D. Andrews and S. Sinclair, “Macro-vascular topography in diabetic retinopathy: Analysis of fundus pictures using high resolution digital image processing,” in ARVO Abstract, p. 262, 1988. [5] M. Tanaka and K. Tanaka, “An automatic technique for fundus-photograph mosaic and vascular net reconstruction,” in MEDlNFO ’80. Amsterdam, the Netherlands: North-Holland, 1980. pp, I 16-120. [6] Y. Okamoto, S. Tamura, and K. Yanashima, “Zero-crossing interval correction in tracing eye-fundus blood vessels,” Pattern Recognition, vol. 21, pp. 227-233, 1988. [7] S. Chauduri, S. Chatterjee, N. Katz, M. Nelson, and M. Goldbaum, “Detection of blood vessels in retinal images using two-dimensional matched filters,” IEEE Trans. Med. Imag., vol. 8, no. 3, pp. 263–269, Mar. 1989. [8] Chen Ko-Yan, “High dynamic range image processing to screen diabetic retinopathy,” master thesis, Jun. 2009. [9] John Moran Eye Center, University of Utah, “Simple Anatomy of the Retina,” http://webvision.med.utah.edu/sretina.html, Oct. 2003. [10] Peter K. Kaiser, 'Ophthalmoscope,” http://www.yorku.ca/eye/ophthal.htm, 1996. [11] Henry Schneiderman, 'The Funduscopic Examination,' http://www.ncbi.nlm.nih.gov/bookshelf/br.fcgi?book=cm∂=A3572, 1990. [12] H. D. Riley, 'Binocular Indirect Ophthalmoscopy,' http://www.opt.indiana.edu/riley/HomePage/BIO_lecture_notes/TEXT_BIO_Exam3.html, 2005. [13] Sudhalkar Eye Hospital, “Video Indirect ophthalmoscopy of ROP,” http://dietcookingrecipes.com/video/eCPmtsLRaRk/Sudhalkar-Eye-Hospital-Video-Indirect-ophthalmoscopy-of-ROP.html, 2010. [14] Image-Acquire.com, 'Canon CF-1 Digital Retinal Camera,' http://www.image-acquire.com/canon-cf-1-digital-retinal-camera/, 2010. [15] Sharat Laser Eye Hospital, 'Retinal Clinic,' http://sharateyehospital.com/sharath-laser-photo-gallery-1.html, 2009. [16] MacLean AL, Maumenee AE, “Hemangioma of the choroid,” American Journal of Ophthalmology, Vol. 50, pp. 3-11, 1960. [17] Buffalo-Niagara Retina Associates, 'Digital Fluorescein Angiography,' http://www.bnretina.com/OurPractice/services.jsp, 1992. [18] Linda G. Shapiro and George C. Stockman, “Computer Vision,” pp 279-325, New Jersey, Prentice-Hall, ISBN 0-13-030796-3, 2001. [19] Wikipedia, “Thresholding (image processing),” http://en.wikipedia.org/wiki/Thresholding_(image_processing), 2010. [20] A. Hoover, V. Kouznetsova, and M. Goldbaum, “Locating blood vessels in retinal images by piecewise threshold probing of a matched Filter response,” IEEE Trans. Medical imaging, vol. 19, no. 3, pp. 203.210, March 2000. [21] Edoardo Ardizzone, et al, “Automatic Extraction of Blood Vessels, Bifurcations and End Points in the Retinal Vascular Tree,' ICBME 2008, Proceedings 23, pp. 22–26, 2009. [22] Thitiporn Chanwimaluang, Guoliang Fan, “An Efficient Blood Vessel Detection Algorithm for Retinal Images using Local Entropy Thresholding,” in Proceedings of the 2003 IEEE International Symposium on Circuits and Systems, Bangkok, Thailand, 2003. [23] Ricci E, Perfetti R., “Retinal blood vessel segmentation using line operators and support vector classification,” IEEE Trans Med Imaging. vol. 16, no. 10, pp. 1357-65, Oct, 2007. [24] R.A. Haddad and A.N. Akansu, 'A Class of Fast Gaussian Binomial Filters for Speech and Image Processing,' IEEE Transactions on Acoustics, Speech and Signal Processing, vol. 39, pp 723-727, March 1991. [25] Image Processing Learning Resources, “Laplacian/Laplacian of Gaussian,” http://homepages.inf.ed.ac.uk/rbf/HIPR2/log.htm, 2004. [26] Li, H. and Chutatape, O., “Fundus image feature extraction,” in Proceedings of the 22nd Annual International Conference of the IEEE Engineering on Medicine and Biology Society, 2000, pp. 3071–3073 (IEEE, New York), 2000. [27] David Marr and Ellen Hildreth, “Theory of Edge Detection,” Proc. R. Soc. London, B 207, 187-217, 1980. [28] E. Ardizzone, et al, “Blood Vessels and Feature Points Detection on Retinal Images,’ 30th Annual International IEEE EMBS Conference, 2008 [29] Lassada Sukkaew, Bunyarit Uyyanonvara, and Sarah Barman, “Comparison of Edge Detection Techniques on Vessel Detection of Infant’s Retinal Image,” Proceedings of the International Conference on Computer and Industrial Management, ICIM, 2005 [30] William K. Pratt, “Digital image processing,” pp.503., John Wiley & Sons, Inc., 1991. [31] Canny, J., “A Computational Approach to Edge Detection,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 8, no. 6, pp. 679–698, 1986. | |
dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/23334 | - |
dc.description.abstract | 前言:視網膜的影像在眼科學中,具有相當大的臨床與研究價值。從單純的彩色影像儲存,到影像特徵之分類,皆有助於臨床診斷與後續之追蹤比較。
方法:本論文中,我們以一階與二階導數演算法處理原始影像,偵測血管、視神經盤與其他特徵及病灶。藉由第一階導數確定其邊緣位置與厚度,並由第二階導數限制各構造與病灶之範圍勿使之溢出其邊界,將兩者做邏輯運算即可將所欲分離之特徵由高雜訊之背景分離出來;此外我們亦使用一高斯平滑之拉普拉斯濾波器進行影像處理,以了解其加速的效能。最後,我們以現今之主流方法 - 匹配濾波器進行速度、效能與統計值之比較。 結果:高斯平滑拉普拉斯濾波器法可得到最快之處理速度(平均2.257秒),導數法次之(4.237秒),匹配濾波器法最慢(12.186秒)。但從整體接受者操作曲線來看,匹配濾波器法於高敏感性區域明顯優於其他兩法,於高特異性區域則三法十分接近,甚至導數法略優於標準之匹配濾波器法。 結論:我們發展合併一階與二階導數之演算法以偵測視網膜血管與其他特徵,在合理可接受的敏感度設定之下可為匹配濾波器法之替代方案,以便達到加速、甚至增加效能之目的,其結果可為電腦自動分類與輔助診斷系統之基石。 | zh_TW |
dc.description.abstract | Introduction: It is valuable to ophthalmology with digitalized retinal images. Ophthalmologists can improve their ability in clinical diagnoses and follow-up in disease progress by the spectra from simple storage of color photographs to automatically computerized classifications of vessels and other architectures on these images.
Methods: We apply a method in the automatic extraction of important architectures of a retina from a highly noisy background by first- and second-order derivatives algorithm. The former detect the edge of the objects in a retinal image. The latter restricts the result of the former algorithm not to cross over its edge. We can isolate architectures from high-noisy background in a retinal image by combining these derivatives methods. Besides, we also use a Laplacian of Gaussian (LoG) filter to process these images. The performance of derivatives and LoG methods are compared with that of the matched response (MFR) method, which is the mainly modern algorithm for noise suppression in image processing. Results: The LoG method costs the shortest processing time (average in 2.257 seconds) the combination of derivatives method is the second (average in 4.237 seconds), and the MFR method is the slowest (average in 12.186 seconds). However, the analysis of ROC curves discloses that the MFR method is the best method within high sensitivity rate. Within low sensitivity rate, .the performance of these three algorithms is similar to that of each other Conclusions: We design an algorithm in combination of the first- and second-order derivatives to detect the vessels and other architectures in a retinal image. This algorithm is an alternative choice of the MFR method to obtain the information faster with a reasonable sensitivity and specificity rate. The information of pre-operated images is the basis and pilot to support an advanced automatic classification system. | en |
dc.description.provenance | Made available in DSpace on 2021-06-08T04:59:21Z (GMT). No. of bitstreams: 1 ntu-99-R97945036-1.pdf: 10431177 bytes, checksum: 560d4f8193cf44cc3cb82b6dbe72d5d5 (MD5) Previous issue date: 2010 | en |
dc.description.tableofcontents | 口試委員會審定書 i
誌 謝 ii 摘 要 iii Abstract iv Contents vi List of Figures viii List of Table x Chapter 1 Introduction 1 1.1 Motivation 1 1.2 The Structures of Retinae 3 Chapter 2 Methods of Retinal Examinations 5 2.1 Direct and Indirect Ophthalmoscopy 5 2.2 Color Photography 8 2.3 Fundus Fluorescein Angiography 10 Chapter 3 Proposed Method 12 3.1 Introduction of Image Segmentation 12 3.2 Matched Filter Response Method 17 3.3 Gaussian Lowpass Filter 20 3.4 Laplacian Filter 22 3.5 Protocol 22 Chapter 4 Experiments and Results 27 4.1 Experimental Environment 27 4.1.1 Environment 27 4.1.2 Experiments 28 4.2 Image Results 32 4.3 Evaluation of Performance by Comparison with the MFR Method 40 Chapter 5 Discussion and Conclusion 46 5.1 Discussion 46 5.2 Limitations 48 5.3 Future Work 48 Reference 49 | |
dc.language.iso | en | |
dc.title | 以一階及二階導數演算法偵測視網膜影像之血管與其他特徵 | zh_TW |
dc.title | Segmentation of Vessels and Other Architectures on Retinal
Images by First- and Second-Order Derivatives Algorithm | en |
dc.type | Thesis | |
dc.date.schoolyear | 98-2 | |
dc.description.degree | 碩士 | |
dc.contributor.oralexamcommittee | 蔡景耀(Ching-Yao Tsai),丁建均(Jian-Jiun Ding) | |
dc.subject.keyword | 影像處理,導數,邊緣偵測,視網膜,匹配濾波器,高斯低通濾波器,拉普拉斯濾波器, | zh_TW |
dc.subject.keyword | image processing,derivative,edge detection,retina,matched filter,Gaussian lowpass filter,Laplacian filter, | en |
dc.relation.page | 52 | |
dc.rights.note | 未授權 | |
dc.date.accepted | 2010-08-19 | |
dc.contributor.author-college | 電機資訊學院 | zh_TW |
dc.contributor.author-dept | 生醫電子與資訊學研究所 | zh_TW |
顯示於系所單位: | 生醫電子與資訊學研究所 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
ntu-99-1.pdf 目前未授權公開取用 | 10.19 MB | Adobe PDF |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。