NEURAL-NETWORK CLASSIFIERS FOR AUTOMATIC REAL-WORLD AERIAL IMAGE RECOGNITION

We describe the application of the multilayer perceptron (MLP) network and a version of the adaptive resonance theory version 2-A (ART 2-A) network to the problem of automatic aerial image recognition (AAIR). T he classification of aerial images; independent of their positions and orientations, is required for automatic tracking and target recogniti on. Invariance is achieved by the use of different invariant feature s paces in combination with supervised and unsupervised neural networks. The performance of neural-network-based classifiers in conjunction wi th several types of invariant AAIR global features, such as the Fourie r-transform space, Zernike moments, central moments, and polar transfo rms, are examined. The advantages of this approach are discussed. The performance of the MLP network is compared with that of a classical co rrelator. The MLP neural-network correlator outperformed the binary ph ase-only filter (BPOF) correlator. It was found that the ART 2-A disti nguished itself with its speed and its low number of required training vectors. However, only the MLP classifier was able to deal with a com bination of shift and rotation geometric distortions. (C) 1996 Optical Society of America