A comparison of feature reduction techniques for classification of hyperspectral remote-sensing data

The task of the analysis of hyperspectral data, due to their high spectral resolution, requires dealing with the problem of the curse of dimensionality. Many feature selection/extraction techniques have been developed, which map the hyperdimensional feature space in a lower-dimensional space, based on the optimization of a suitable criterion function. This paper studies the impact of several such techniques and of the criterion chosen on the accuracy of different supervised classifiers (the "minimal-distance-to-means", the k-NN, and the Bayes classifier with Gaussian distributions). The compared methods are the "Sequential Forward Selection" (SFS), the "Steepest Ascent" (SA), the "Fast Constrained Search" (FCS), the "Projection Pursuif' (PP) and the "Decision Boundary Feature Extraction" (DBFE), while the considered criterion functions are standard interclass distance measures (Bhattacharyya, Jeffries-Matusita and divergence distances). SFS is well known for its conceptual and computational simplicity. SA provides more effective subsets of selected features at the price of a higher computational cost. DBFE is an effective transformation technique, usually applied after a preliminary feature-space reduction through PP. The experimental comparison is performed on an AVIRIS hyperspectral data set characterized by 220 spectral bands and nine ground cover classes. The computational time of each algorithm is also reported.