Description

Enhancement of the images will be more helpful in medical applications inorder to identify the different regions in the medical images. In the enhanced images the affected portions were more clearly visible. The denoising process removes unwanted pixels in medical images so that the organ portions and the disease affected regions were identified exactly in the images. The defects in the images can be identified based on the features extracted from the denoised images. The noise commonly occurring in OCT images were salt and pepper, Gaussian noises. A process that identifies noises in the medical images and removes the noisy pixels based on K-SVD denoising algorithm is proposed. The denoised images were then clustered and edge regions were identified in the medical images. The input OCT images were taken as input. Noises were added to the input images producing noisy images. The input images were denoised based on the K-SVD process. K-SVD is a generalization of the k-means clustering method, and it works by iteratively alternating between sparse coding the input data based on the current dictionary, and updating the atoms in the dictionary to better fit the data. It aims to partition n observations into few clusters in which each observation belongs to the cluster with the nearest mean, serving as a prototype of the cluster. The denoised images were then clustered and edges of the clusters were identified. The clusters were estimated based on the FCM clustering process. The edge regions of the different portions in the images were identified and they were marked. The defects in the images were identified based on the statistical features extracted from the images. The statistics used for the extraction of features were mean, standard deviation and variance. The extracted features were classified using KNN classifier inorder to identify the defects in the images. The performance of the process is then measured using the performance parameters like PSNR, MSE and SSIM.