EVIDENTIAL REASONING APPROACH TO MULTISOURCE-DATA CLASSIFICATION IN REMOTE-SENSING

In the evidential reasoning approach to the classification of remotely sensed multisource data, each data source is considered as providing a body of evidence with a certain degree of belief, The degrees of bel ief are represented by ''interval-valued probabilities'' rather than b y conventional point-valued probabilities so that uncertainty can be e mbedded in the measures, The proposed method is applied to the ground- cover classification of simulated 201-band High Resolution Imaging Spe ctrometer (HIRIS) data, from which a set of multiple sources is obtain ed by dividing the dimensionally huge data into smaller pieces based o n the global statistical correlation information, By a divide-and-comb ine process, the method is able to utilize more features than conventi onal maximum likelihood methods.