In this paper, we present a study of the usage of intensity exposure in histogram segmentation and its performance in histogram equalization. Two techniques are proposed: the mean-based bi-histogram equalization plateau limit (mean-BHEPL) or median-based BHEPL (median-BHEPL) and adaptive bi-histogram equalization algorithm (ABHE). Both techniques initially divide the input histogram into two subhistograms through a threshold value computed from the intensity exposure of the image. Histogram clipping for mean-BHEPL and median-BHEPL is then performed on these subhistograms using the mean and median values, respectively. Conventional histogram equalization is also implemented on each clipped subhistogram. The second proposed technique, ABHE, applies the modified version of the adaptive histogram equalization algorithm (AHEA) on both subhistograms. Results of extensive simulations reveal that mean-BHEPL and median-BHEPL perform comparably to the conventional BHEPL technique. ABHE exhibits excellent performance in image quality, naturalness, and mean brightness preservation. However, it is slightly inferior in image detail preservation to the conventional AHEA technique. In conclusion, segmenting the input histogram through the threshold value calculated based on the intensity exposure of the image yields good enhancement results.
Histogram equalization, gray-scale image, histogram segmentation, histogram clipping, image details
RUI, TANG JING and ISA, NOR ASHIDI MAT
"Intensity exposure-based bi-histogram equalization for image enhancement,"
Turkish Journal of Electrical Engineering and Computer Sciences: Vol. 24:
5, Article 19.
Available at: https://journals.tubitak.gov.tr/elektrik/vol24/iss5/19