De-noising of retinal image using crafty edge detection (CED)

Abstract

In all fields Image processing doing excellent analysis to diagnosis the diseases. Especially image processing is relevant to modern ophthalmology. It is a field that heavily dependent on visual data. In medical filed ophthalmologists used retinal images for diagnostic purpose. So these images frequently need visual enhancement prior to apply a digital analysis for pathological risk or damage detection. In this work we propose a image enhancement techniques to compensate the noise in retinal images. The quality of retinal images affected by various noise and it can be denoised by proposed algorithm Crafty Edge Detection(CED). This new work helps to increase SNR (Signal-to-Noise Ratio) value. Also, Median filter, Gaussian filter, Mean filter, Average and proposed had taken for the comparative study. PSNR and MSE metrics are used to analyze the performance and quality of the retinal image for further processing. Experimental results proved that the proposed filter produce better PSNR (Peak Signal to Noise ratio) and reduces MSE (Mean Square Error) than other filter.