Image enhancement by scaling the discrete wavelet coefficients based on segmentation

Histogram Equalization is one of the most effective technique to process the digital images over the years used for the purpose of image enhancement and normalization. But in some cases traditional Histogram Equalization (HE) may cause bad results. There is a remedy given in this paper which takes over the traditional HE. This paper presents a technique based on segmented histogram for contrast enhancement of the image by scaling the discrete wavelet transform coefficient. This utilizes the advantage of wavelet transform and preserves the color consistency in addition to improving the overall contrast of the image. This approach efficiently reduced the contaminated noise in the image by introducing wavelet shrinkage terms adaptively in the different scales. Experimental results and comparison is made between a number of techniques on the basis of Absolute mean brightness error (AMBE), PSNR, Entropy and Standard Deviation values.