Context and objectives
The INS (Institut National de Statistique) and other governmental or regional administrations which have a heavy need in ground occupation data for economic or land planning and management purposes more and more rely on remotely sensed image data to acquire the necessary information.
With the launch in early 1998 of the SPOT 4 satellite, Spot Image not only ensures continuity in the acquisition of its traditional image data but also provides an improvement of its multispectral image products by adding a new 20 metre resolution mid infrared (MIR) channel to the previous three channels.
The principal objective of this feasibility study is to test the general quality of these new four band multispectral images of SPOT 4 and to find out how the new MIR channel contributes to an improved image classification of landcover, with a special aim at the delineation of built areas.
The new MIR channel is highly sensitive to soil and leaf moisture, making it especially useful for improving analysis of vegetation cover and soil cover types. The aim of this study is to analyse how this channel also contributes to the improvement of the discrimination of other features especially in urban areas.
The general methodology which will be used in this study is to compare the results of three-band versus four-band analysis of the new SPOT 4 image over some test areas, which have already been the object of previous studies and for which other ancillary data are available, in order to be able to effective improvement bound to the new channel.