期刊名称:International Journal of Innovative Research in Computer and Communication Engineering
印刷版ISSN:2320-9798
电子版ISSN:2320-9801
出版年度:2015
卷号:3
期号:11
DOI:10.15680/IJIRCCE.2015.0311196
出版社:S&S Publications
摘要:Identifying retinal pathologies in fundus images is an overwhelming issue. It can be solved usingupcoming process. This paper presents, the best way to segment a Fundus Image is unsupervised iterative vesselsegmentation algorithm. First, a enhanced vessel image is generated by Morphological reconstruction of the greenplane image. An initial estimate of the segmented vessel is extracted by the global thresholding the vessel enhancedimage. Then new pixels are identified iteratively by adjustable thresholding of the leftover image achieved by maskingout the existing segmented vessel estimate from the vessel enhanced image. The new vessel pixels are then regiongrown into the existing vessel, thereby resulting in an iterative enrichment of the segmented vessel structure. Thenumber of false edge pixels are identified as new vessel pixels compared with actual vessel pixels by the iterationprogress. The high segmentation accuracy can be achieved by means of novel stopping criteria in the iterative process
关键词:Fundus image; Morphological Operation and Iterative Segmentation