Related Experiment Video
Updated: Jun 5, 2025
09:33
Demonstration of Equal-Intensity Beam Generation by Dielectric Metasurfaces
Published on: June 7, 2019
6.2K
Intelligent electromagnetic metasurface camera: system design and experimental results
Zhuo Wang1, Hongrui Zhang1, Hanting Zhao1
1Department of Electronics, State Key Laboratory of Advanced Optical Communication Systems and Networks, Peking University, Beijing 100871, China.
Nanophotonics (Berlin, Germany)
|December 5, 2024
Summary
This study introduces an intelligent electromagnetic metasurface camera that uses machine learning for high-quality, efficient sensing. It can image through walls, overcoming limitations of current electromagnetic (EM) sensing technologies.
Area of Science:
- Electromagnetic (EM) sensing and advanced imaging technologies.
Background:
- Traditional EM sensing faces challenges in energy consumption, cost, efficiency, and portability.
- Growing demand for high-quality, 3D, high-frame-rate sensing requires innovative solutions.
Purpose of the Study:
- To propose an intelligent EM metasurface camera integrating programmable metasurfaces and machine learning.
- To establish a Bayesian inference framework for enhanced EM sensing capabilities.
Main Methods:
- Synergetic integration of inexpensive programmable metasurfaces with modern machine learning techniques.
- Development of a Bayesian inference framework for intelligent data acquisition and processing.
- Deployment in real-world settings for practical validation.
Main Results:
- Demonstrated high-quality imaging with improved efficiency, even in noisy environments.
- Successfully imaged human behaviors behind a 60 cm thick reinforced concrete wall.
- Achieved frame rates in the order of tens of Hz for real-time monitoring.
Conclusions:
- The intelligent EM metasurface camera offers a promising solution to current EM sensing limitations.
- This technology enables clear visualization of targets previously invisible to conventional methods.
- The strategy has potential impacts on sensing, smart communities, and beyond.
