On the potential for land cover mapping from multiple-view-angle (MVA) remotely-sensed images

This Letter explores the use of multiple-view-angle (MVA) remotely-sensed images for land cover classification. It is shown that, for a simple agricultural scene, the classification accuracy obtained using single-waveband MVA data is comparable to that derived from conventional multi-spectral data. The study highlights the need to separate the spectral and directional components of information contained in MVA data, so that an objective assessment can be made of these data for land cover mapping.