http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
AN AUTOMATIC METHOD FOR IMAGE CONTRAST ENHANCEMENT BY HISTOGRAM WEIGHT CLUSTERING
Nyamlkhagva Sengee,Byambaragchaa Bazarragchaa,Choi Heung Kook 한국멀티미디어학회 2008 한국멀티미디어학회 학술발표논문집 Vol.2008 No.2
Histogram equalization (GHE) is a simple and well known method in contrast enhancement techniques. There are extensions of GHE to preserve image brightness. Although they can preserve a brightness of original image more than GHE, they couldn't enhance a visualization of original image on some images. Therefore, we propose a new method which not only can preserve the brightness of original image but also can enhance the visualization of original image. The proposed method is called 'Brightness Preserving Weight Clustering Histogram Equalization' (BPWCHE). BPWCHE assigns each non zero bins of original image's histogram to one cluster, and computes every cluster's weight. Then, in order to reduce cluster number, we use three criterions (cluster weight, weight ratio and width of two neighbor clusters) to merge two neighbor clusters. The clusters acquire same partitions of result image histogram. Finally, transformation functions of each cluster's sub-histogram are calculated based on traditional GHE method in their new acquired partitions of result image histogram, and the sub-histogram's gray levels are mapped to the result image by the transformation functions of sub-histograms, correspondingly. As experimental results, BPWCHE can preserve image brightness and enhance visualization of image more effective than GHE and other brightness preserving method.