Automatic target recognition for hyperfine SAR sensors

The challenge of automatically detecting and classifying targets in SAR imagery has been actively researched for many years. ARFL's MSTAR target database has been used extensively to develop and demonstrate methods addressing this problem. However, MSTAR is over 20 years old and no longer reflects the capabilities of modern SAR systems

Modern SAR systems such as DSTL's BRIGHT SPARK system (developed by Thales UK) can achieve resolutions of the order of a few centimetres. At such fine resolutions, many of the classical assumptions made about targets must be reconsidered in order to extract the greatest amount of information from the scene. This project aims to investigate these challenges and demonstrate appropriate solutions to the problems identified in the work.