Multisensor microwave sea-ice classification

摘要: A knowledge of polar sea-ice extent, type, and location is a valuable tool in understanding many geophysical processes. This paper presents an iterative statistical technique for ice classification multisensor remote sensing imagery. NSCAT, ERS-2, SSM/I data are used to construct enhanced resolution images. Preprocessing consists standardization principal component analysis dimensionality reduction. The algorithm developed based on maximum posteriori (MAP) method with the assumption Gaussian conditional probability density each type. applied collected during six day imaging interval September 1996. For this application, performs better than nearest neighbor, likelihood, or K-means approaches.