Meng Tianyou, Wang Yunjia. Ideal equalization of remote sensing images and quantitative assessment of image quality[J]. Journal of Image and Graphics, 2012, 17(5): 729-739.DOI: 10.11834/jig.20120519.
The intensity distribution of most original remote sensing images concentrated in a narrow range
resulting in unclear image details and low contrast of the image. Histogram equalization is commonly used to distribute the gray levels of an image uniformg which can increase the contrast of images and enhanced the details. Through the elaboration of the definition of comentropy
we draw forth the image enhancement algorithm of histogram equalization. Then
we put forward an improved algorithm of ideal histogram equalization
which is based on piecewise mapping thoughts through the analysis of the defects existing in the traditional histogram equalization. In addition
we relate and analyze the implementation of the improved algorithm in this paper. Meanwhile
in order to do the quantitative analysis and comparison between the traditional algorithm and the modified algorithm
we use synthetic average contrast
and details evaluation parameters
which are based on simultaneous contrast
the contrast resolution limit
and fuzzy mathematics.Finally
we use simultaneous contrast
synthetic average contrast
and details evaluation parameters as the indexes of quantitative evaluation and quantitative evaluation of the improved algorithm. The results of quantitative evaluation show that the image enhancement effect of the improved algorithm proposed in this paper is better than the algorithm of traditional histogram equalization.