The main objectives of this chapter are to evaluate the performances of different image fusion techniques for the enhancement of spectral and textural variations of different forest types and toapply arefined maximum likelihood
classifier for the extraction of forest class informationfrom the fused images in order to update a forest geographical information system (GIS). For the data fusion, modified intensity-hue-saturation (IHS) transformation, principal
components analysis (PCA) method, Gram-Schmidt fusion, color normalization spectral sharpening, wavelet-based method, and Ehlers fusion are used and the results are compared. Of these methods, the better results are obtained through the use of the modifiedIHS transformation, PCA and wavelet-based fusion. The refined classification method uses spatial thresholds defined from contextual knowledge and different features obtained through a feature derivation process. The result of the refined classification is compared with the results of a standard method and it demonstrates a higher accuracy. Overall, the research indicates that multisource data fusion can significantly improve the interpretation and classification of forest types and the elaborated refined classification method is a powerful tool to increase classification accuracy.
