A Super-Resolution Imaging Method Based on a Left-Handed Material and a Diffusion Model
Year: 2024
Type: article
Abstract: This article proposes a super-resolution imaging method based on a left-handed material cylindrical lens (LHCL) and a diffusion model (DM). First, to recover the super-resolution information loss in the received scattered field (SF), we design an LHCL with a solid immersion mechanism (SIM). With the evanescent wave components recovered by the LHCL’s physical characteristics, the imaging resolution can reach$0.04\lambda $and$0.0157\lambda $in the radial and tangential directions, respectively. Due to the narrow focus range inside the LHCL, a DM is used to overcome the limitations of the imaging region. Combining the physical characteristics of the LHCL with the nonlinear fitting abilities of the DM, our method can realize high-quality inverse scattering imaging for different targets and performs better both in generalization and noise resistance compared with traditional iterative methods and the U-Net method.
Authors: Qianqian Zhang, Chengyou Yin, Anqi Li
Institution: National University of Defense Technology
Language: English
Subfield: Biomedical Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: hybrid
Estimated APC paid: $2,495
OpenAlex ID: W4405632103