Abstract: The condition of the vascular network of the human eye is an important diagnostic factor in retinopathy. Its segmentation in fundus imaging is a nontrivial task due to the variable size of vessels, relatively low contrast, and potential presence of pathologies like microaneurysm and hemorrhages. The Project proposes the Retinal image analysis through efficient detection of vessels and exudates for retinal vasculature disorder analysis. It plays an important role in the detection of some diseases in their early stages, such as diabetes, which can be performed by comparison of the states of retinal blood vessels. Intrinsic characteristics of retinal images make the blood vessel detection process difficult. Here, we proposed a new algorithm to detect the retinal blood vessels effectively. The green channel will be selected for image analysis to extract vessels accurately. The Daubechies wavelet transform is used to enhance the image contrast for effective vessel detection. To increase the efficiency of the morphological operators by reconstruction, they were applied using multi-structure elements. A simple thresholding method along opening and closing indicates the remained ridges belonging to vessels. The experimental result proves that the blood vessels and exudates can be effectively detected by applying this method to the retinal images.
Keywords: Diabetic retinopathy, Pre-processing, Feature extraction, Classification, Discrete curvelet transform, Global contrast normalization, Digital image processing, Artificial neural network, Neural network classifier.
|
DOI:
10.17148/IJARCCE.2023.124156
