Robust automatic target recognition using extra-trees

Jonathan Pisane (1, 2) , Raphaël Marée (2) , Louis Wehenkel (2) , Jacques Verly (2)

1 SONDRA - Sondra, CentraleSupélec, Université Paris-Saclay (COmUE)
2 Institut Montefiore - Department of Electrical Engineering and Computer Science

In this paper, we describe a new automatic target recognition algorithm for classifying SAR images based on the PiXiT image classifier. It 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 an extended version of the public standard MSTAR database, that includes targets of interest, false targets, and background clutter. A misclassification rate of about three percent has been achieved. In this paper, we describe a new automatic target recognition algorithm for classifying SAR images based on the PiXiT image classifier. It 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 an extended version of the public standard MSTAR database, that includes targets of interest, false targets, and background clutter. A misclassification rate of about three percent has been achieved.

Domains: Engineering Sciences [physics]; Signal and Image processing; Computer Science [cs]; Signal and Image Processing

Jonathan Pisane, Raphaël Marée, Louis Wehenkel, Jacques Verly. Robust automatic target recognition using extra-trees. 2010 IEEE Radar Conference, May 2010, Washington United States. ⟨10.1109/RADAR.2010.5494683⟩. ⟨hal-00556229⟩