COMPOSITE FOURIER-PLANE NONLINEAR FILTER FOR DISTORTION-INVARIANT PATTERN-RECOGNITION

We investigate composite Fourier-plane nonlinear filters for distortio n-invariant pattern recognition. Different types of noise such as spat ially nonoverlapping color background noise, other nontarget objects, real scene noise, and additive white noise are used to investigate the performance of the filler. Computer simulation results are presented to show the, performance enhancement using Fourier-plane nonlinearitie s in the composite filter designs. The training image sets and the tes t image sets are generated by in-plane and out-plane rotation of diffe rent aircraft. (C) 1997 Society of Photo-Optical Instrumentation Engin eers.