Fast, Optimal Algorithm for Real Time Image Enhancement

Abstract

This paper presents a fast and efficient algorithm to enhance digital image for real time image processing. The true quality of image and minor details in the image are of serious importance in any image analysis. In electronic applications, these are usually affected by varying depth of camera as well as magnitude and direction of sunlight on camera; Resulting digital image contains inaccurate color level, contrast levels, intensity noises in pixels. Paper presents a proposed algorithm and its comparison with traditional enhancement techniques. Unlike traditional histogram approaches, Boosted and Dualistic Contrast Intensity Enhancement enhances color image in three major operations: Global Recursive Separated and Weighted Histogram Stretching, Local Sliding Window of Contrast Limited Adaptive Histogram Equalization with the cancellation of pixel noises, and Luminance Adjustment Function. Proposed technique enhances the image in such a way that it makes minor details more prominent by making dark pixels brighter, sharp pixels lighter and keeping light pixels of same intensity as it is; Overall process maintains the balance among the regions of varying contrast and intensity level without the amplification of unwanted pixels noises. Algorithm is designed to perform under low light in real time, when images start losing the color strength of minor details.