Near-Infrared High-Resolution Imaging via Deep Learning Based on Broadband Achromatic Metalens

Abstract

The development of large-aperture broadband achromatic metalenses is crucial for advancing compact optical systems but remains challenging due to complex phase compensation requirements and the inherent trade-off between numerical aperture and aperture size. Therefore, a high-resolution near-infrared imaging framework that integrates a large-aperture polarization-independent achromatic metalens with a deep learning-based super-resolution algorithm is proposed. The metalens, operating from 1.34 to 1.54 μm with a diameter of 210 μm, is designed using a discrete multi-wavelength methodology combined with a global optimization algorithm, breaking the limitations of conventional continuous achromatic design. Experimental results show the fabricated metalens achieves a resolution of 19.69 μm. To further transcend the hardware limitation, a super-resolution residual network (SRResNet) is employed for post-processing, enhancing the imaging resolution to 17.54 μm. This work establishes a novel paradigm by synergistically combining meta-optics with deep learning, which exhibits promising potential in the applications of virtual/augmented reality and integrated photonics.