“…With the successful application of SVMs to a wide variety of different pattern recognition areas, such as 3D object recognition, image classification, character recognition, etc, SVMs have recently attracted increasing attention in remote-sensed multi/hyper-spectral communities [7,8,9,10,11,12,13,14]. Previous literature applying SVMs to hyperspectral data classification [10,11,12] has shown competitive performance with the best available classification algorithms.…”