IMAGE CLASSIFICATION BY EXTENDED CERTAINTY FACTORS

Two supervised classifiers based on Certainty Factors (CFs) are descri bed; in particular, a new algorithm for the generation of the base of classification rules is proposed. Such an algorithm specifies what ''e vents'' should be used as conditions and with what CFs rules may assig n samples to classes. The main novelty lies in the definition of event s as one-dimensional adaptively computed functions. Interesting featur es of the proposed classifiers are the use of comprehensible classific ation criteria and the automatic ''knowledge acquisition'' from traini ng patterns. Experimental results obtained in the case of multisensori al remote-sensing images are reported.