Classification of objects with complex geometric shape by means of low frequency electro-magnetic response

A study of the effectiveness of low frequency electromagnetic responses for identifying objects of complex shape is presented. The linear separability of a large variety of objects such as cubes, cylinders and aircraft were examined. An efficient computational method for modeling the scattering returns from an arbitrarily shaped object was developed. This technique was used to simulate the required radar returns from aircraft without resorting to time consuming and costly experimental measurements. Two classification algorithms, a linear discriminant and a nearest neighbor rule, were used to classify a set of four aircraft models. The performance of classification is presented in terms of the probability of misclassification versus noise level. The effectiveness of the various combinations of electromagnetic features was evaluated. (Modified author abstract)