Military Ground Targets Database (MGTD) for SAR ATR BelloniCarole diane lucie BalleriAlessio AoufNabil Le CaillecJean-Marc MerletThomas 2019 <div><b>Access to the database:</b></div><div>In order to access the MGTD, please contact the corresponding author:</div><div>Dr. Alessio Balleri (a.balleri@cranfield.ac.uk)</div><div><br></div><div>Backup contact:</div><div>Cranfield University research data support (researchdata@cranfield.ac.uk)</div><div><br></div><div><br></div><div><b>Abstract:</b></div><div>We present a high resolution SAR dataset consisting of images of a set of ground military target models taken at various aspect angles, The dataset can be used for a fair evaluation and comparison of SAR ATR algorithms.We applied the Inverse Synthetic Aperture Radar (ISAR) technique to echoes from targets rotating on a turntable and illuminated with a stepped frequency waveform. The targets in the database consist of three 1.5m-1.7m-long models of T-64, T-72, BMP-1 tanks. The gun, the turret position and the depression angle are varied to form 24 different sequences of images. The emitted signal spanned the frequency range from 13 GHz to 18 GHz to achieve a bandwidth of 5 GHz sampled with 4001 frequency points. Single polarised images (Horizontal-Horizontal) are generated using the backprojection algorithm. Images are produced using a 20° integration angle. The 1728 images in the dataset are organised in a suggested training and testing set to facilitate a standard evaluation of SAR ATR algorithms. More images are though available. The training and testing sets have different depression angles, different backgrounds and the target are mostly in different configurations in both sets to maximise the environment variability and minimise possible correlation in the database.</div><div><br></div>