MODEL-BASED ADAPTIVE HISTOGRAM EQUALIZATION

Adaptive histogram equilization techniques have been used successfully in a variety of applications for improving local contrast in images. In this paper we take a model-based approach to the analysis of histog ram equalization algorithms. On this basis statistical properties can be derived and used to find objective and optimal filters. We also pre sent a fast approximate algorithm which generalizes to a wider class o f grey-level mappings that are functions of local statistics.