Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/19915
標題: 在眼底影像中進行滲漏分割
An Exudate Segmentation Method on Retinal Images
作者: 蔡政哲
Tsai, Jheng-Jhe
關鍵字: 糖尿病視網膜病變
Diabetic retinopathy
滲漏偵測
對比受限自適應直方圖等化
梯度向量流蛇形
隨機漫步演算法
exudate detection
CLAHE
GVF snake
random walk
出版社: 資訊科學與工程學系
引用: [1] Pizer, Stephen M., et al. “Adaptive histogram equalization and its variations.” Computer vision, graphics, and image processing, 1987, 39(3), pp. 355-368. [2] Reza, Ali M. “Realization of the contrast limited adaptive histogram equalization (CLAHE) for real-time image enhancement.” Journal of VLSI signal processing systems for signal, image and video technology, 2004, 38(1), pp. 35-44. [3] Kauppi, T., Kalesnykiene, V., Kamarainen, J.K., L, L., Sorri, I., Uusitalo, H., Pietila, J.,Kalviainen, H., Uusitalo, H.,“The DIARETDB1 diabetic retinopathy database and evaluation protocol.” Proceedings of British Machine Vision Conference, 2007. [4] H. K. Hsiao, C. C. Liu, C. Y. Yu, S. W. Kuo, S. S. Yu, “A Novel Optic Disc Detection Scheme on Retinal Images,” Expert System with Applications,2012, Vol. 39, pp. 10600–10606. [5] GRADY, Leo. “Random walks for image segmentation. Pattern Analysis and Machine Intelligence.” IEEE Transactions on, 2006, 28(11), pp. 1768-1783. [6] A. Karamalis, W. Wein, T. Klein, N. Navab, “Ultrasound confidence maps using random walks,” Medical Image Analysis,2012, 16, pp.1101–1112. [7] Modzelewski, Romain, et al. “Brain perfusion heterogeneity measurement based on Random Walk algorithm: choice and influence of inner parameters,” Computerized Medical Imaging and Graphics, 2010, 34(4), pp. 289-297. [8] Kass Michael, Witkin Andrew, Terzopoulos Demetri. “Snakes: Active contour models.” International journal of computer vision, 1988, 1(4), pp. 321-331. [9] Xu Chenyang, Prince Jerry L. “Snakes, shapes, and gradient vector flow.” Image Processing,IEEE Transactions on, 1998, 7(3), pp. 359-369. [10] Leo Grady, Gareth Funka-Lea, “Multi-Label Image Segmentation for Medical Applications Based on Graph-Theoretic Electrical Potentials.” Computer Vision and Mathematical Methods in Medical and Biomedical Image Analysis. Springer Berlin Heidelberg, 2004, pp. 230-245. [11] R.C. Gonzalez, R.E. Woods, “Digital Image Processing”, 2nd ed, Prentice-Hall, 2002. [12] Lobregt, S., Viergever, M. A., “A discrete dynamic contour model”, Medical Imaging, IEEE Transactions on, 1995, 14(1), pp. 12-24. [13] S. W. Yang, M. H. Sheu, H. H. Wu, H. E. Chien, P. K. Weng, and Y. Y. Wu, “VLSI Architecture Design for a Fast Parallel Label Assignment in Binary Image.” Circuits and Systems, 2005. ISCAS 2005. IEEE International Symposium on, pp. 2393-2396. [14] R. Modzelewski, E. Janvresse, T. de la Rue, P. Vera, “Brain perfusion heterogeneity measurement based on Random Walk algorithm: Choice and influence of inner parameters,” Computerized Medical Imaging and Graphics, 2010, 34(4), pp. 289-297. [15] A. Karamalis, W. Wein, T. Klein, N. Navab, “Ultrasound confidence maps using random walks,” Medical Image Analysis ,2012, 16, pp. 1101–1112. [16] E.D. Pisano, et al., “Contrast limited adaptive histogram equalization image processing to improve the detection of simulated speculations in dense mammograms,” J. Digital Imaging , 2010, 11(4), pp. 193–200. [17] I. K. Maitra, S. Nag, S. K. Bandyopadhyay, “Technique for preprocessing of digital mammogram,” Computer methods and programs in biomedicine, 2012, 107(2), pp. 175-188. [18] A. F. Amos, D. J. McCarty, and P. Zimmet, “The rising global burden of diabetes and its complications: Estimates and Projections to the year 2010,” Diabetic medicine, 1997, 14(5), pp. 7-85. [19] (2013, Jane 30)Centers for Disease Control and Prevention National Diabetes Fact Sheet [Online]. Available: http://www.cdc.gov200. [20] Carl R, Tom S. C, Mark W. J, et al., “Diabetic Retinopathy. “In:AAO, ed. Retina and Vitreous. San Francisco, AmericanAcademy of Ophthalmology, 2007, pp. 99-119. [21] Chew Ey, Mils Jl, Metzger Be, et al., “Metabolic control and progression of retinopathy: The Diabetes in Early Pregnancy Study.” Diabetes care, 1995, 18(5), pp. 631-637. [22] K. W. Tobin, E. Chaum, V. P. Govindasamy, “Detection of Anatomic Structure in Human Retinal Inagery,” Medical Imaging, IEEE Transactions on, 2007, 26(12), pp. 1729-1739. [23] S. Kavitha, K. Duraiswamy, “Automatic Detection of Hard and Soft Exudates in Fundus Images Using Color Histogram Thresholding,” European Journal of Scientific Research, 2011 48(3), pp. 493-504. [24] H. E. Cline, W. E. Lorensen, S. Ludke, “High-resolution 3D reconstruction of CT images of the head,” Schenectady, NY, GE Research & Development Center, 1986. [25] C. I. S’anchez, R. Hornero, M. I. Lopez, M. Aboy, J. Poza, D. Abasolo, “A novel automatic image processing algorithm for detection of hard exudates based on retinal image analysis.” Med Eng Phys, 2008, 30(3), pp. 350-357. [26] Walter Thomas, et al. “A contribution of image processing to the diagnosis of diabetic retinopathy-detection of exudates in color fundus images of the human retina.” Medical Imaging, IEEE Transactions on, 2002, 21(10), pp. 1236-1243. [27] Welfer Daniel, Scharcanski Jacob, Marinho Diane Ruschel. “A coarse-to-fine strategy for automatically detecting exudates in color eye fundus images.” Computerized Medical Imaging and Graphics, 2010, 34(3), pp. 228-235. [28] Osareh Alireza, Shadgar Bita, Markham Richard. “A computational-intelligence-based approach for detection of exudates in diabetic retinopathy images.” Information Technology in Biomedicine, IEEE Transactions on, 2009, 13(4), pp. 535-545. [29] Giancardo Luca, et al. “Exudate-based diabetic macular edema detection in fundus images using publicly available datasets.” Medical Image Analysis, 2012, 16(1), pp. 216-226. [30] Sopharak Akara, et al. “Automatic detection of diabetic retinopathy exudates from non-dilated retinal images using mathematical morphology methods.” Computerized Medical Imaging and Graphics, 2008, 32(8), pp. 720-727. [31] Kim Yeong-Taeg. “Contrast enhancement using brightness preserving bi-histogram equalization.” Consumer Electronics, IEEE Transactions on, 1997, 43(1), pp. 1-8. [32] Wang Yu, Chen Qian, Zhang Baeomin. “Image enhancement based on equal area dualistic sub-image histogram equalization method.” Consumer Electronics, IEEE Transactions on, 1999, 45(1), pp. 68-75. [33] Sim K. S., Tso C. P., Tan Y. Y. “Recursive sub-image histogram equalization applied to gray scale images.” Pattern Recognition Letters, 2007, 28(10), pp. 1209-1221. [34] Ravishankar Saiprasad, Jain Arpit, Mittal Anurag. “Automated feature extraction for early detection of diabetic retinopathy in fundus images.” Computer Vision and Pattern Recognition, 2009. CVPR 2009. IEEE Conference on. IEEE, 2009, pp. 210-217.
摘要: 滲漏是糖尿病視網膜病變的早期徵兆之一。當它出現在黃斑區,可能會導致視力衰退,嚴重時甚至失明。因此,有必要建立早期預警系統,以有效控制病情。在以往的一些文獻中,往往只利用影像中的亮度資訊,同時也缺乏在一些較大型的資料集的評估結果。本篇文章提出一個新的滲漏檢測方法,在演算法中使用了對比受限自適應直方圖等化(CLAHE)來銳化黃斑區的影像,使用上限離群值及型態學以得到滲漏區域初始輪廓。最後使用了GVF snake及random walk來精確分割出滲漏輪廓。本篇文章採用了公開資料庫DIARETDB1 (Standard Diabetic Retinopathy Database Calibration level 1),並以misclassification error (ME)、accuracy、RFAE、sensitivity、及specificity來評估效能。在與專家手繪的ground truth比較之後,GVF snake算法的平均ME、accuracy、RFAE、sensitivity、及specificity分別為0.011、0.994、0.06461、0.871、及0.9972。實驗結果證明本篇文章提出的演算法可以準確偵測出滲漏區域。
Exudate is one of the earliest clinical signs of diabetic retinopathy. When it appears in the macular, it may lead to visual loss. Therefore, it is necessary to establish an early warning system in order to effectively control the illness. In the previous literature on exudates identification, mainly relied on gray-level information, and were not evaluated on large datasets. This paper presents a novel method for automated identification of exudate in retinal images. This algorithm uses contrast limited adaptive histogram equalization (CLAHE) to sharp the macular region, the upper outlier detection and morphological processing step to segment the rough border of exudates, GVF-snake and random walk scheme to get the accurate exudate contours. The proposed algorithm is tested on the available public dataset DIARETDB1. With compare to the expert ophthalmologists’ hand-drawn ground-truths, the misclassification error (ME), accuracy, RFAE, sensitivity, and specificity for the proposed algorithm is 0.011, 0.994, 0.06461, 0.871, and 0.9972, respectively. The experimental results show that the proposed algorithm can detect the exudate area accurately. Keyword: Diabetic retinopathy, exudate detection, CLAHE, GVF snake, random walk.
URI: http://hdl.handle.net/11455/19915
其他識別: U0005-1807201314080200
文章連結: http://www.airitilibrary.com/Publication/alDetailedMesh1?DocID=U0005-1807201314080200
Appears in Collections:資訊科學與工程學系所

文件中的檔案:

取得全文請前往華藝線上圖書館



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.