Performance Analysis of Underwater Image Enhancement with CLAHE 2D Median Filtering Technique On the Basis Of SNR, RMS Error, Mean Brightness

Abstract

The objects in the underwater images are faint, difficult to view and analyze because the images of such environment loses the details of the objects. In this paper a method of RMS error reduction & SNR improvement is presented for underwater images with keeping Mean brightness preserved in which the image quality is first enhanced using Contrast Limited Adaptive histogram equalization method and 2d median filter , and then histogram thresholdding is used for segmentation of the objects for analysis. The method is tested on various type of underwater image environment and signal to noise ratio (SNR), RMS error and mean brightness is used for comparison for various images and it is found that using this method is advantageous as it improves the SNR to greater values even after segmentation while RMS error represented by ERMS gets reduced leading to better visualization and keeping mean brightness constant.