Nuclear norm joint sparse representation for hyperspectral image classification

Joint sparse representation (JSR) models have been widely applied into the field of hyperspectral image (HSI) classification. However, most of JSR-based models adopt the Frobenius norm to measure the reconstruction error, which ignores the structural information of the small patch. In this paper, we propose a nuclear-norm joint sparse representation (NuJSR) model for hyperspectral image classification. The main motivation of NuJSR is to utilize the nuclear-norm to measure the reconstruction error, so as to reflect the low-rank structural information of the small patch. To optimize our proposed NuJSR, an efficient algorithm is proposed. Experiments results confirm the effectiveness of NuJSR.