An Approach to Automatic Target Recognition in Radar Images Using SVM

Abstract

This paper introduces an Automatic Target Recognition (ATR) method based on X Band Radar image processing. A software which implements this method was developed following four principal stages: digital image formation, image preprocessing, feature selection through a combination of C4.5 Decision Tree and PCA and classification using SVM. The automatic process was validated using two images sets, one of them containing real images with natural noise levels and the other with different degrees of impulsive noise contamination. The method achieves a very nice computation behavior and effectiveness, high accuracy and robustness in noise environments with a low storage memory and high decision speed.

A Low-RCS, High-GBP Fabry–Perot Antenna With Embedded Chessboard Polarization Conversion Metasurface

“…The X-band was widely used in stealth radars, such as airborne radar [50], [51], shipborne radar [52], [53], and air defense radar [54]. Its relative short wavelength enables higher resolution imaging for target identification and discrimination [55], [56]. For better stealth, improving the in-band RCS reduction of the designed FP antenna is an important issue.…”