REMOTE OBJECT RECOGNITION BY ANALYSIS OF SURFACE-STRUCTURE

We present a new algorithm for the discrimination of remote objects by their surface structure. Starting from a range-azimuth profile functi on, we formulate a range-azimuth matrix whose largest eigenvalues are used as discriminating features to separate object classes. A simpler, competing algorithm uses the number of sign changes in the range-azim uth profile function to discriminate among classes. Whereas both algor ithms work. well on noiseless data, an experiment involving real data shows that the eigenvalue method is far more robust with respect to no ise than is the sign-change method. Two well-known methods based on su rface structure, variance, and fractal dimension were also tested on r eal data. Neither method furnished the aspect invariance and the discr iminability of the eigenvalue method.