Enhancement of overexposed color images

A highly overexposed color image is characterized by high brightness, low chromaticity and loss of detail. In this paper, a fuzzy based enhancement approach is developed to improve contrast and lightness. The brightness and color relative to the brightness of a similarly illuminated white are corrected by selecting the fuzzy operators. Hue, Saturation and intensity Value (HSV) color model is used to preserve the hue in the images. Based on the amount of intensity exposure, a threshold is determined to divide the image pixels into two regions, dark and bright. For contrast improvement, darker and brighter regions of V component are fuzzified and modified by selecting separate membership functions. For lightness correction, S component is fuzzified and suitably modified. A parameter is determined adaptively by conducting subjective evaluations to modify the power law membership function. The results are compared on the basis of subjective and objective evaluations. The proposed algorithm preservers the information content by reducing the overexposure and gives the best results in terms of processing time.