期刊名称：Indian Journal of Computer Science and Engineering
印刷版ISSN：2231-3850
电子版ISSN：0976-5166
出版年度：2021
卷号：12
期号：6
页码：1875-1885
DOI：10.21817/indjcse/2021/v12i6/211206189
语种：English
出版社：Engg Journals Publications
摘要：The challenging problem of designing techniques that assist computer-aided medical diagnosis has developed as a result of the rapid expansion of computing and information technologies. Due to the obvious growing number of diabetic retinopathy (DR) patients, automated procedures for rapid screening are in greater demand. The accurate diagnosis of DR depends upon detecting and analyzing some features of human retina (i.e. blood vessels, fovea, optic disc, etc.) as well as several types of spot lesions (i.e. exudates, drusen, microaneurysms, hemorrhage, etc). This study focus on blood vessel segmentation in the human retina, which is an essential milestone in retinal image analyzation. To accomplish more efficient and optimum segmentation, the proposed approach employs a number of image processing techniques including contrast enhancement, normalisation, and thresholding. Experiments show that the presented technique produces high-accuracy segmentation and the algorithm is ideally suited for real-time screening applications and large data retrieval.
