Image classification in complex spaces

摘要

This paper addresses issues related to classification of images in complex spaces. The image is represented in terms of a phase and amplitude components. The classifier optimizes functions of joint real and imaginary conditional probability density functions. Bound on the total probability of errors in terms of Rayliegh quotient is derived and compared to the cases where non-complex amplitude-only signal is used. Examples of application of the proposed approach on polarimetric radar imagery indicate several orders of magnitude improvement in performance