A study on estimation of high resolution component for image enhancement using wavelet

An estimation method of high resolution component, based on multi-scale wavelet representation, is discussed. We have newly proposed an edge classification rule and edge model. In our method, each edge in an image is analyzed by multi-resolution analysis and is classified according to the pattern of neighboring right and left edges together. A high deflnition image and its band-limited image are used for learning the edge transition from low resolution to high resolution. Finally, high resolution component is estimated by switching the transition rule that is adapted to an input image. We have confirmed the validity of our estimation method by computer simulation.