TACSTechnology and Application of Computer Science2998-89262998-8934Art and Technology10.61369/TACS.2026040026Article基于暗场稀疏先验的QDPC 相位重建与缺陷检测https://artdesignp.com/journal/TACS/3/4/10.61369/TACS.2026040026曾子铭,夏宇轩,刘玄玉,王艳华,李好,涂雨轩2026342026-02-28定量差分相衬（Quantitative Differential Phase Contrast, QDPC）成像作为一种非干涉定量相位恢复技术，在无标记生物样本成像中具有重要应用。然而，传统QDPC 重建算法未充分考虑成像系统的内在特性，导致重建质量易受噪声和背景波动影响。近年来，研究者提出利用暗场稀疏先验（Dark-field Sparse Prior, DSP）改进QDPC 相位重建，通过L0范数优化实现高保真度相位恢复。本文系统阐述了暗场图像与QDPC 相位重建的基本原理，分析了DSP 方法的核心思想与数学实现，并探讨了其在缺陷检测领域的应用潜力。定量差分相衬,暗场稀疏先验,相位重建,缺陷检测,L0范数

[1]Shao M. Dark-field Sparse Prior for Quantitative DPC Imaging. SPIE Digital Library, 2020.[2]Zhang S, Peng T, Ke Z, et al. High-fidelity quantitative differential phase contrast deconvolution using dark-field sparse prior. In the processing of the SixteenthInternational Conference on Photonics and Imaging in Biology and Medicine (PIBM 2023). SPIE, 234-242(12745), 2023.[3]Peng T, Ke Z, Zhang S, et al. Quantitative differential phase contrast phase reconstruction for sparse samples. Optics and Lasers in Engineering, 74-78(163), 2023.[4]Zhang S, Wu H, Peng T, et al. Pupil-driven quantitative differential phase contrast imaging. arXiv preprint, 2306-2310, 2023.[5]Fan Y, Sun J, Chen Q, et al. Optimal illumination scheme for isotropic quantitative differential phase contrast microscopy. Photonics Research, 890-905(8), 2019.