Local Histogram Equalization with Brightness Preservation

摘要: As the traditional algorithms for local histogram equalization is used to enhance contrast of an image,the output brightness always regardless input image.However,this not desirable property in many applications where preserving original essential,so a new algorithm proposed.Based on partially overlapped sub-block equalization(POSHE),we change way equalizing sub-block:based mean,each separated recursively into series sub-images with different gray ranges,then histograms are equalized over their respective ranges,and finally composed resulting sub-images.The simulation results show that,with algorithm,the details image can be enhanced effectively,and also preserved meantime.Besides,scalable degree preservation may achieved by choosing number recursive separation.The proposed more natural manner and hence,is very useful commercial purposes such as consumer electronics required.