Fusion of Quickbird MS and Pan data for urban studies

摘要

Most of the satellite sensors, presently operating in the optical domain, are providing a data set comprising multispectral images at a low spatial resolution and images at a higher spatial resolution but with a lower spectral content. The trend of satellite sensors is similar to the present situation. The idea of fusing multispectral images with a highest spatial resolution enables the creation of useful products for urban planning and management. This paper aims at evaluating two methods for construction of synthetic multispectral images having a highest spatial resolution available within the data set, in the objective of studying urban areas. The first one is derived from the ARSIS concept and the second one is based on a correlation technique. The two methods are described and tested over the urban area of Strasbourg (France). The resulting images are evaluated through visual, qualitative and quantitative criteria. Some conclusions are drawn on the difference between the two algorithms and on the benefits of their use in urban studies.