Combining ALOS AVNIR-2 and PALSAR for land cover classification

Recent lunched ALOS satellite carries both optical AVNIR-2 and full polarimetric PALSAR sensors that provide both surface spectral and scattering information of the Earth ground surface. The full polarimetric PALSAR is particularly important in relation to target identification and recognition, since it contains additional real physical target information. Therefore, combination of optical and polarimetric data has the potential to improve the land cover classification accuracy. As a consequence, the development of techniques for efficiently and conveniently handling combined multi-source data for land cover classification is becoming increasingly important. The objective of this paper is to assess the potential synergies and complementary nature of optical, full polarimetry PALSAR intensity data, and polarimetric features (indicators) for land cover classification and evaluate the recent developed subspace method for the combined optical and SAR data classification.