LAUSR

Hyperspectral remote sensing sensors can provide plenty of valuable information with hundreds of spectral bands at each pixel. Feature selection and spectral-spatial information play an important role in the field of hyperspectral image (HSI) classification. In this paper, a novel two-stage spectral-spatial HSI classification method is proposed. In first stage, the standard particle swarm optimization (PSO) is adopted to optimize the parameters, and a novel binary PSO with mutation mechanism is used for feature selection simultaneously. Then, the support vector machine classifier is performed. In second stage, in order to reduce salt and pepper phenomenon, mathematical morphology post-processing is used to further refine the obtained results of the above stage. Experiments are conducted on two real hyperspectral data sets. The evaluation results show that the proposed approach achieves better accuracy than several state-of-the-art methods.