Extraction of Non Proliferative Diabetic retinopathy using new designed wavelet and classification using KNN classifier

Abstract

Diabetic retinopathy is very serious condition in which increased blood sugar level causes swelling in the blood vessels and it may leak in to the retina. It is the effect of diabetes on the eye. There are four stages of diabetic retinopathy that are Mild non-proliferative diabetic retinopathy, Moderate non-proliferative diabetic retinopathy, severe non-proliferative diabetic retinopathy, Proliferative diabetic retinopathy. In this paper we have done the extraction of non-proliferative diabetic retinopathy. We have extracted microaneurysms and exudates using new designed wavelet (DR), and classification is done using KNN classifier. KNN classifier classifies in to four classes that are mild, moderate, severe, and healthy. For this work we have used three publically available fundus image databases that are DIARETDB0, DIARETDB1 and STARE databases having 130, 89, and 397 fundus images respectively. Using this technique we have got 99.03% accuracy for detection of NPDR.

Authors and Affiliations

Deepali D. Rathod, Ramesh R. Manza, Yogesh M. Rajput, Manoj Saswade, Neha Deshpande

Keywords

.

Related Articles

The Effect of Curing Period on Strength Behaviour of Sand Bentonite Mixes

A landfill is a secured engineered pit in the ground into which the wastes are dumped. Proper landfill liners need to be used to avoid any hydraulic connections between waste and surrounding environment. These liners can...

Comparing Optimal Homotopy Asymptotic Method and Adomian Decomposition Method for the Solution of Integro-Differential Equations

In this work, the author compares optimal homotopy asymptotic method with Adomian decomposition method for the solution of integro-differential equations. These methods are applied to first, second and third order proble...

NFC based Secure Mobile Health Care System

In this paper, a novel architecture for improving healthcare system with the help of Android based mobile devices with NFC and Bluetooth interfaces, smartcard technology on tamper resistant secure element (SE) for storin...

Zero Discharge Concepts in Five Star Hotel – A Case Study

Concept of zero liquid discharge (ZLD) not only offers a solution to protect water bodies from pollution by the disposal of large quantity of industrial and municipal wastewater but also provides a sustainable approach f...

Download PDF file
  • EP ID EP393012
  • DOI 10.9790/9622-0712040613.
  • Views 106
  • Downloads 0

How To Cite

Deepali D. Rathod, Ramesh R. Manza, Yogesh M. Rajput, Manoj Saswade, Neha Deshpande (2017). Extraction of Non Proliferative Diabetic retinopathy using new designed wavelet and classification using KNN classifier. International Journal of engineering Research and Applications, 7(12), 6-13. https://europub.co.uk/articles/-A-393012