The sources of information available are:

• a multispectral satellite image and
• images called out-image data describing the geographical context of the area in terms of:
• altitude,
• slope orientation,
• localization of the urban areas,
• localization of the roads...

The result of a classification carried out by a neural network is also used. This classification provides the ``ideal'' theoretical map of distribution of the classes to be recognized according to the most favourable geographical context ([Mascarilla, 1996]).

The classifications are supervised. The classes of vegetation to be recognized are defined using samples characteristic of them.

• The Information Sources
• LANDSAT Images of Palni Area
• Spectral Bands
• Dimensions and Grey Levels
• Vegetation
• Out-Image Data
• Geographical Context
• Premises
• Example
• The Map of Areas Favourable to each Class
• Samples of each Class
• The Learning Step
• How to Compute a Histogram from Samples of a Class.
• Calculation of the Possibility Distributions from the Histograms of Grey Levels.
• Classification Accuracy
• Classification Process
• Estimation of Classification Accuracy
• A more Realistic Estimation