• Mahalakshmi Bollimuntha
• Srinivasarao G
• CH.Geethanjali,
• P.Pavani,
• P.N.S.S Tejaswi

Diabetes problems can lead to a disease called diabetic retinopathy (DR), which damages the retina's blood vessels irreversibly. DR is a major cause of blindness, if not identified in it’s early stages. In recent years, Deep learning (DL) methods have emerged as powerful tools for image analysis, demonstrating remarkable success in various medical imaging applications. Combining multiple datasets, Accuracy, Sensitivity, Specificity and Processing difficulty are the major drawbacks of existing work by using Convolutional Neural Network (CNN) method. The proposed system classifies DR images into five stages - no DR, Mild, Moderate, Severe, Proliferative DR and it gives accurate result for multiple datasets lesions localization system for increasing many applications in the field of medical image processing and diagnosis process based feature analysis is performed to get the better system by using the application of De-noising Autoencoder (DAE) techniques for the classification of diabetic retinopathy with high accuracy and that exceeds the current state-of-the-art results.

Diabetic Retinopathy, Deep Learning, Convolutional Neural Network, Image Analysis, De-noising Autoencoder, Multiple datasets.

"CLASSIFICATION OF DIABETIC RETINOPATHY LEVELS BY USING DENOISING AUTOENCODER (DAE)", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 3, page no.k497-k500, March-2024, Available :http://www.jetir.org/papers/JETIR2403A64.pdf

"CLASSIFICATION OF DIABETIC RETINOPATHY LEVELS BY USING DENOISING AUTOENCODER (DAE)", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 3, page no. ppk497-k500, March-2024, Available at : http://www.jetir.org/papers/JETIR2403A64.pdf