Glaucoma and diabetic retinopathy diagnosis using image mining B.V.Baiju 1 , V.Ceronmani Sharmila 2 ,MuthurajB 3 ,Abdul Hasan M 4 ,Mahendar V 5 (bvbaiju@hindustanuniv.ac.in 1 , csharmila@hindustanuniv.ac.in 2 , muthuviz143@gmail.com 3 , mohabdul20@gmail.com 4 , mahendar0605@gmail.com 5 ) Department of Information Technology, Hindustan Institute of Technology and Sciences, Chennai 1,2,3,4,5 Abstract Diabetes is an overall unavoidable sickness that can cause recognizable microvascular complexities like diabetic retinopathy and macular edema in the normal eye retina, the pictures of which are today used for manual disease screening and assurance. This work genuine task could inconceivably productive by customized acknowledgment using a Deep learning methodology. Here we present a profound learning structure that perceives referable diabetic retinopathy comparably or better than presented in the past investigations. The proposed strategy evades the need of sore division or applicant map age before the arrangement stage. Neighbourhood parallel examples and granulometric profiles are privately registered to extricate surface and morphological data from retinal images. Various blends of this data feed arrangement calculations to ideally separate brilliant and dark lesions from solid tissues. These outcomes propose, radial basis function in classification could build the expense viability of screening and finding, while at the same time accomplishing higher than suggested execution, and that the framework could be applied in clinical assessments requiring better reviewing. Key words: Diabetic retinopathy, image processing, feature extraction, deep learning techniques. 1.INTRODUCTION Diabetic retinopathy is the most notable microvascular intricacy in diabetes, for the screening of which retinal imaging is the most by and large used system in view of its high affectability in distinctive retinopathy. The evaluation of the earnestness and level of retinopathy identified with an individual having diabetes is correct currently performed by clinical experts subject to the fundus or retinal pictures of the patient's eyes. As the amount of patients with diabetes is rapidly growing, the quantity of retinal pictures made by the screening tasks will moreover fabricate, which along these lines presents a tremendous work heightened inconvenience on the clinical experts similarly as an expense to the clinical consideration organizations. This could be helped with a robotized structure either as help for clinical experts' work or as a full I3CAC 2021, June 07-08, Chennai, India Copyright © 2021 EAI DOI 10.4108/eai.7-6-2021.2308568