Semi dynamic fuzzy histogram equalization

Conventional histogram equalization is a widely used technique for contrast enhancement, but it usually over-enhances the contrast of the images resulting into loss to their natural appearance. To overcome this problem, histogram segmentation prior to equalization process has been suggested in the literature. Although these methods significantly enhance the contrast but usually fail to preserve the natural appearance of the images. In this paper, a fuzzy-HE method to preserve the brightness and natural appearance of the images is proposed. It decomposes the fuzzy histogram of the input image into two segments using median value of occupied intensities. Prior to equalization process, the narrow segment is identified and allocated with new dynamic range. Finally the combined equalized sub-histograms are normalized to avoid intensity saturation and un-even distribution of bins. The simulation results show that the proposed method yields better results both qualitatively and quantitatively.