Image Contrast Enhancement Using Normal Matching Histogram Equalization
ABSTRACT Histogram Equalization (HE) is a very popular algorithm in the field of image enhancement. Its theory is very simple but effective and easy to implement. However, this algorithm can not get good result in some special cases. Furthermore, it will change the mean brightness of original image significantly. According to these drawbacks of HE, some novel algorithms have been proposed. The main target of these algorithms is trying to preserve the brightness and entropy of original image better. But they also decrease the enhancement efforts at the same time. In this paper, a novel algorithm, Normal Matching Histogram Equalization (NMHE), is proposed. Experimental results show that this algorithm can not only preserve the mean brightness and entropy of original image but also keep the enhancement efforts simultaneously.
- [Show abstract] [Hide abstract]
ABSTRACT: Image processing requires an excellent image contrast-enhancement technique to extract useful information invisible to the human or machine vision. Because of the histogram flattening, the widely used conventional histogram equalization image-enhancing technique suffers from severe brightness changes, rendering it undesirable. Hence, we introduce a contrast-enhancement dynamic histogram-equalization algorithm method that generates better output image by preserving the input mean brightness without introducing the unfavorable side effects of checkerboard effect, artefacts, and washed-out appearance. The first procedure of this technique is; normalizing input histogram and followed by smoothing process. Then, the break point detection process is done to divide the histogram into subhistograms before we can remap the gray level allocation. Lastly, the transformation function of each subhistogram is constructed independently. © 2011 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 21, 280-289, 2011;International Journal of Imaging Systems and Technology 08/2011; 21(3):280 - 289. DOI:10.1002/ima.20295 · 0.77 Impact Factor