Hyperspectral image feature classification using stationary wavelet transform

Hyperspectral Images are a set of narrow spectrum band images used in the recognition and mapping of surface materials such as minerals and vegetation. Usually these Hyperspectral Image datasets are of high dimensional which makes its classification process a complex task and of low accuracy by using conventional classification approaches. Image dimensionality reduction and feature classification have become necessary steps in multi-dimensional hyperspectral image processing. This study investigates an effective algorithm for extracting spatial features using stationary wavelet transform (SWT) and reducing spectral dimensionality using principal component analysis (PCA). K-nearest neighbor classifier is used in the classification step for the features. Experimental results show that the proposed SWT-PCA algorithm outperforms the other two methods.