Efficient Retinal Image Segmentation Using Wavelets and Neural Networks

Abstract

Retinopathy has turned into a commonly spread disease in the world and it causes many complications. One of the common vision threatening debilitating complexities of Diabetic Retinopathy. It occurs when blood vessels in the patient's retina begin to leak into the macula region of eye. The purpose of this paper is to extricate features from retina digital images based on a further analysis of high frequency components (HH) obtained with the Discrete Wavelet Transform (DWT).  In particular, the DWT is applied to the retina photograph to obtain its high-high (HH) image sub band using db1,symlet,biorthogonal wavelet transform. Then, a further decomposition by DWT is applied to the HH image subband of the previous step to obtain HH*. Finally, statistical features are computed from HH* Discrete Wavelet Transform (DWT) based features and Adaptive Neural  Inference System   is reported. The computational results show that present stage(i.e., normal or abnormal) and gives  overall accuracy and sensitivity, specificity.