Hyperspectral band selection by multitask sparsity pursuit

Abstract—Hyperspectral images have been proved to be effec-tive for a wide range of applications; however, the large volume and redundant information also bring a lot of inconvenience at the same time. To cope with this problem, hyperspectral band selection is a pertinent technique, which takes advantage of re-moving redundant components without compromising the original contents from the raw image cubes. Because of its usefulness, hy-perspectral band selection has been successfully applied to many practical applications of hyperspectral remote sensing, such as land cover map generation and color visualization. This paper fo-cuses on groupwise band selection and proposes a new framework, including the following contributions: 1) a smart yet intrinsic descriptor for efficient band representation; 2) an evolutionary strategy to handle the high computational burden associated with groupwise-selection-based methods; and 3) a novel MTSP-based criterion to evaluate the performance of each candidate band combination. To verify the superiority of the proposed framework, experiments have been conducted on both hyperspectral classi-fication and color visualization. Experimental results on three real-world hyperspectral images demonstrate that the proposed framework can lead to a significant advancement in these two applications compared with other competitors. Index Terms—Band selection, compressive sensing (CS), hyper-spectral image, immune clonal strategy (ICS), machine learning, multitask learning (MTL). I