الفهرس | Only 14 pages are availabe for public view |
Abstract In the present work, a novel algorithm to enhance low contrast images is developed. The algorithm is a combination of the Histogram Equalization (HE) technique and the Fast Gray-Level Grouping (FGLG) technique. It is designed to perform effectively with images that have the position of the highest amplitude histogram components lies in the left of nonzero histogram components (NZHC) region. The proposed algorithm was implemented and applied to forty-six low contrast gray-scale and fourteen colored images. The performance of the proposed algorithm was compared with five existing contrast enhancement techniques: histogram equalization (HE), gray level grouping (GLG), fast gray level grouping (FGLG), weighted thresholded histogram equalization (WTHE) and low-complexity histogram modification (LCHM). Assessment of image contrast enhancement was performed via two approaches: subjective and objective criteria. An attempt has been made to investigate the effect of noise on the performance of the proposed algorithm. Different levels of Gaussian and Impulsive noise types were added to the data used and the same quality measures were calculated. It provides the highest performance through the qualitative visual inspection and the image quantitative analysis. Moreover, it outperforms other algorithms for clear and noisy low contrast images and can be conducted in a fully-automated manner. |