Spectral-Routed Mixture of Experts for High-Fidelity Complex-Field Measurement in Single-Frame Lens-Free On-Chip Microscopy
- 26 March 2026
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 75 (00189456) , 1-13
Abstract
Single-frame lens-free on-chip microscopy (LFOM) offers a compact and cost-effective solution for biomedical diagnostics and micro/nanoscale inspection but suffers from severe twin-image artifacts and frequency-dependent degradation due to numerical aperture (NA) limitations. Existing reconstruction methods often neglect the heterogeneous frequency sensitivity of LFOM and fail to explicitly suppress artifacts, leading to suboptimal complex-field measurement quality. To address these challenges, we propose SEMNet, a spectral-expert measurement network that introduces frequency-aware modeling into single-frame LFOM reconstruction. Specifically, SEMNet transforms captured holograms into polar Fourier space, partitions them into directional subbands, and adaptively routes each subband via a soft top-k gating mechanism to lightweight expert networks specialized for low-, mid-, and high-frequency restoration. A physics-consistent loss jointly enforces complex-field fidelity, Fresnel-likelihood consistency, twin-image suppression, and spectral load balancing, thereby aligning the reconstruction objective with physical imaging principles. Extensive experiments on 404 high-resolution hologram–complex field pairs from seven types of biological tissues demonstrate that SEMNet improves peak signal-to-noise ratio by up to 1.7 dB over strong single-frame baselines, with statistically significant gains in structural similarity indices (p < 0.001). SEMNet restores high-frequency textures while suppressing NA-boundary artifacts. It delivers 47.1 FPS end-to-end inference with 100× fewer parameters than a representative CNN baseline and 5× fewer MACs than a representative Transformer baseline, highlighting a frequency-structured, physics-aligned design for efficient LFOM.Keywords
Funding Information
- National Natural Science Foundation of China (12374271)
- National Natural Science Foundation of China (62375065)
- National Natural Science Foundation of China (62405079)
- National Natural Science Foundation of China (52375328)
- National Natural Science Foundation of China (62575091)
- Fund of National Laboratory on Adaptive Optics of China (FNLAO-24-MS-O07)
- Fundamental Research Funds for the Central Universities (AUGN5750300325)
- Natural Science Foundation of Heilongjiang Province (LH2024A009)
- Key Laboratory of Image Processing and Intelligent Control (Huazhong University of Science and Technology), Ministry of Education, Wuhan, China (IPIC2025-05)
- Key Laboratory of Advanced Medical Imaging and Intelligent Computing of Guizhou Province (AMIIC250204)
- China Postdoctoral Science Foundation-China Coal Technology and Engineering Group (CCTEG) Joint Support Program (2025T011ZGMK)
- State Key Laboratory of Ultrafast Optical Science and Technology, Xi’an Institute of Optics and Precision Mechanics, Chinese Academy of Sciences (2025SKLuFAST-KF04)
- Open Fund of IPOC, BUPT (IPOC2025A13)
This publication has 46 references indexed in Scilit:
- FViT: A focal vision transformer with gabor filterApplied Soft Computing, 2025
- Angular Momentum Superposition Beams‐Based InterferometerLaser & Photonics Reviews, 2025
- Adaptive Structured‐Light 3D Surface Imaging with Cross‐Domain LearningLaser & Photonics Reviews, 2025
- Speckle noise reduction for digital holographic images using Swin TransformerOptics and Lasers in Engineering, 2024
- Flexible and universal autofocus based on amplitude difference of fractional Fourier transformOptics and Lasers in Engineering, 2023
- Fast autofocusing of recorded planes by salient feature region for coherent diffraction imagingJournal of Biophotonics, 2023
- Phase-based motion estimation in complex environments using the illumination-invariant log-Gabor filterMechanical Systems and Signal Processing, 2022
- Fourier Imager Network (FIN): A deep neural network for hologram reconstruction with superior external generalizationLight: Science & Applications, 2022
- Phase recovery and holographic image reconstruction using deep learning in neural networksLight: Science & Applications, 2017
- Pixel super-resolution using wavelength scanningLight: Science & Applications, 2015