An extra-trees-based automatic target recognition algorithm

In this article, we describe a new automatic target recognition algorithm for classifying SAR images. It is based on the PiXiT image classifier, already used in several applications, which uses randomized sub-windows extraction and extremely randomized trees (extra-trees). This approach requires very little pre-processing of the images, thereby limiting the computational load. It was successfully tested on the public standard MSTAR database, achieving a misclassification rate of about one percent.