OBJECT CLASSIFICATION AND ACOUSTIC IMAGING WITH ACTIVE SONAR

The theoretical underpinnings of underwater acoustic classification and imaging using high-frequency active sonar are studied. All essential components of practical classification systems are incorporated in a Bayesian theoretic framework. The optimum decision rules and array processing are presented and evaluated. A systematic performance evaluation methodology is derived. New results quantify the relationship between classifier performance and object geometry, acoustic imaging, and the accuracy of a priori knowledge infused into the processor.