Automatic Target Recognition Using Discrimination Based on Optimal Transport
This work addresses automatic target recognition for civilian vehicles using SAR images, presenting an incremental improvement by applying known optimal transport distances to a new classification task.
The paper tackled the problem of classifying civilian vehicles from SAR images by comparing the Monge-Kantorovich optimal transport distance to the standard l2 distance for automatic target recognition, showing that the transportation-based approach provides a robust geometric framework for discrimination.
The use of distances based on optimal transportation has recently shown promise for discrimination of power spectra. In particular, spectral estimation methods based on l1 regularization as well as covariance based methods can be shown to be robust with respect to such distances. These transportation distances provide a geometric framework where geodesics corresponds to smooth transition of spectral mass, and have been useful for tracking. In this paper, we investigate the use of these distances for automatic target recognition. We study the use of the Monge-Kantorovich distance compared to the standard l2 distance for classifying civilian vehicles based on SAR images. We use a version of the Monge-Kantorovich distance that applies also for the case where the spectra may have different total mass, and we formulate the optimization problem as a minimum flow problem that can be computed using efficient algorithms.