Study of Brightness Preservation Histogram Equalization Techniques

Image captured under various environment conditions with low contrast results in visual deterioration. To enhance the contrast of images a number of techniques have been proposed over the years. Histogram equalization is one of those methods which could be used for this purpose. For the contrast enhancement purpose histogram equalization (HE) attempts reduction in the number of gray levels. Contrast enhancement by histogram equalization method involve mapping of gray levels on the basis of changeover function that is derived from the probability distribution of gray levels embedded in input image. For this, the neighbouring gray levels in image having low probability density values are combined into single gray level value. Whereas for neighbouring gray levels having high probability density value the gap between them is increased, HE often makes over-enhancement of images with frequent gray level region resulting in loss of contrast for less frequent area of the images. Thus, the HE technique is not able to preserve image brightness. So wherever the preservation of image brightness is required this method is not preferred. This paper review various histogram equalization techniques used in preserving image brightness as well as the contrast enhancement is presented. The key approach for this aim is to use image segmentation. The criteria used are decomposition of images and the presentation of the basic difference between these techniques. For comparative study of performance of these techniques, absolute mean brightness error (AMBE) is used to determine degree of brightness preservation while peak signal to noise ratio (PSNR) have been used to fine\ degree of contrast enhancement.