Anomaly targets detection of hyperspectral imagery based on sparse representation

摘要

Anomaly target detection of hyperspectral image has become a hot in remote sensing research field, the paper is studied on the popular sparse representation method of anomaly target detection, and which are compared with traditional algorithm, such as the generalized likelihood ratio detection KRX and RX algorithm. The results show very good detection performance for sparse representation method of anomaly target detection. At last, the simulation results demonstrate that the proposed sparse representation algorithm outperforms the other algorithm, it is higher precision and lower false alarm rate.