An algorithm for the fusion of images based on Jaynes' maximum entropy method
This work offers a new algorithm for the fusion of images for improving the spatial resolution of multispectral images, using the concepts of entropy of information and a priori probability. The algorithm is based on Jaynes' maximum entropy method and starts from a previous modelling of the fusion of data as an operation in which a series of probabilities are designated at a micro-scale in terms of average statistical values at a macro-scale, which are those recorded by the image with the lowest spatial resolution. The theoretical base of the model is presented, its practical implementation is explained, a complete description allowing reproduction of the analysis is offered and the results of fusing Landsat Thematic Mapper (TM) images with IRS-1D Pan are shown.
No Reference information available - sign in for access.
No Citation information available - sign in for access.
No Supplementary Data.
No Article Media