**7. Conclusions**

In this paper, an optimized algorithm , referred to as DPBS-CEM for hyperspectral target detection, is proposed. A deep pipelined architecture of DPBS-CEM on FPGA is developed by using HLS as well. The experimental results show that the proposed FPGA implementation of DPBS-CEM has an extraordinary performance improvement in terms of data throughput without compromising for detection accuracy. Under the same test conditions, the detection speed of our proposed DPBS-CEM is about 7.3 times faster than that of SBS-CEM.

**Acknowledgments:** This work was partially supported by the National Natural Science Foundation of China (Nos. 61571345, 91538101, 61501346, 61502367, and 61701360) and the 111 project (B08038).

**Author Contributions:** Jie Lei and Yunsong Li conceived and designed the experiments; Dongsheng Zhao and Jing Xie performed the experiments; Jie Lei and Jing Xie analyzed the result data; Chein-I Chang provided suggestions on algorithm optimization and paper revision; Lingyun Wu, Xuepeng Li, Jintao Zhang, and Wenguang Li contributed reagents/materials/analysis tools; Jie Lei wrote the paper.

**Conflicts of Interest:** The authors declare no conflict of interest.
