The implications of Boltzmann-type machines for SAR data processing: A preliminary survey

It is proposed that Markov random field models (MRFs) be used as a framework within which to construct models of synthetic aperture radar (SAR) images. The relationship between this class of models and the Boltzmann machine (BM) of artificial intelligence is clarified. The BM training procedure is generalized and used to train MRF models. This technique is used to investigate the ability of a simple MRF texture model to learn a texture by maximizing a relative entropy objective function. The marriage of MRF models with the BM training procedure is fruitful.