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：2018
卷号：6
期号：2
页码：1007
DOI：10.15680/IJIRCCE.2018.0602058
出版社：S&S Publications
摘要：Diabetic retinopathy (DR) is a serious eye disease originating from diabetes mellitus and the mostcommon cause of blindness in the developed countries. Early treatment can prevent patients to become affected fromthis condition or at least the progression of DR can be slowed down. The key to the early detection is to recognizemicroaneurysms (MAs) in the fundus of the eye in time. Thus, mass screening of diabetic patients is highly desired, butmanual grading is slow and resource demanding. Microaneurysms (MAs) are early signs of DR, so the detection ofthese lesions is essential in an efficient screening program to meet clinical protocols. Early micro aneurysm detectioncan help reduce the incidence of blindness and Micro aneurysm detection is the first step in automated screening ofdiabetic retinopathy. A reliable screening system for the detection of MAs on digital fundus images can provide greatassistance to ophthalmologists in difficult diagnoses. This project presents image processing techniques such as darkobject detection to analyze the condition or enhance the input image in order to make it suitable for further processingand improve the visibility of Microaneyrysm in color fundus images. The correlation coefficient between eachprocessed profile and a typical microaneurysm profile is measured and used as a scale factor to adjust the shape of thecandidate profile. Each candidate is then classified based on spread spectrum analysis features. We implement thisretinal imaging in real time environments.
