Using a Kernel Adatron for Object Classification with RCS Data

TL;DR

This paper demonstrates that support vector machines can rapidly and accurately classify objects from radar cross section data, outperforming Bayesian methods in speed, with high identification accuracy across various object shapes.

Contribution

The paper introduces the use of support vector machines for object classification from RCS data and shows significant speed advantages over Bayesian approaches.

Findings

Achieved over 95% accuracy in object classification.

SVM approach is three orders of magnitude faster than Bayesian methods.

Best results obtained with data fusion of X-band and S-band signals.

Abstract

Rapid identification of object from radar cross section (RCS) signals is important for many space and military applications. This identification is a problem in pattern recognition which either neural networks or support vector machines should prove to be high-speed. Bayesian networks would also provide value but require significant preprocessing of the signals. In this paper, we describe the use of a support vector machine for object identification from synthesized RCS data. Our best results are from data fusion of X-band and S-band signals, where we obtained 99.4%, 95.3%, 100% and 95.6% correct identification for cylinders, frusta, spheres, and polygons, respectively. We also compare our results with a Bayesian approach and show that the SVM is three orders of magnitude faster, as measured by the number of floating point operations.