An Image Enhancement Based on RGB Color Channels with Fuzzy C-Means Clustering

Abstract

Image enhancement is used to improve the visual quality of an image. This paper proposes an image enhancement technique, which is based on the clustering of pixels using Fuzzy C Means (FCM). The aim of this paper is to provide a system to process the input image in such a way that the resultant image is well suited for interpretation by the humans and also by machines. Initially, the given input image is preprocessed based on the Median filtering algorithm. It helps to remove the unwanted and noisy images on the input image, which helps to provide the enhanced image. After that, we extract the RGB colors from the preprocessed image based on the color image transformation. Then, the image is clustered separately as red channel, green channel and blue channel. Finally, the FCM technique is used to clustering the image. The resultant image is the enhanced image which highlights certain characteristics than the original image. The experimental results are compared with the existing OSGFB approach in terms of SNR, MSE, entropy, contrast, correlation, dissimilarity, homogeneity and PSNR. The proposed methods outperforms better and enhance image than the exiting OSGFB method.