Machine-learning reprogrammable metasurface imager

Conventional microwave imagers usually require either time-consuming data acquisition, or complicated reconstruction algorithms for data post-processing, making them largely ineffective for complex in-situ sensing and monitoring. Here, we experimentally report a real-time digital-metasurface imager that can be trained in-situ to generate the radiation patterns required by machine-learning optimized measurement modes. This imager is electronically reprogrammed in real time to access the optimized solution for an entire data set, realizing storage and transfer of full-resolution raw data in dynamically varying scenes. High-accuracy image coding and recognition are demonstrated in situ for various image sets,…

• NS
  National Science Foundation

  Award: 111-2-05
• NN
  National Natural Science Foundation of China

  Awards: 2017YFA0700202, 61731010, 111-2-05, 2017YFA0700203, 61471006, 61631007, 2017YFA0700201, 61571117
• HE
  Higher Education Discipline Innovation Project

  Awards: 2017YFA0700201, 111-2, 61571117, 61631007, 111-2-05
• NK
  National Key Research and Development Program of China

  Awards: 2017YFA0700201, 2017YFA0700203, 61631007, 111-2-05, 2017YFA0700202, 61571117