用双层优化学习适合显微镜的二值光照模式，提升欠采样下的成像质量。
Learning Binary Sampling Patterns for Single-Pixel Imaging using Bilevel Optimisation
- 通过双层优化自动学习任务相关的二值光照模式，替代人工设计。
- 在低采样率和小数据量下，重建精度优于传统方法和端到端深度学习。
- 结合可学习的变分正则化，提升图像质量和抗噪能力，适合显微成像场景。
单像素成像（SPI）通过一系列结构光照射与单个探测器配合，实现物体重构。光照模式的选择至关重要，尤其在高度欠采样条件下，直接影响重建质量和采集速度。本文提出一种基于双层优化的方法，直接从数据中学习针对特定任务的二值光照模式，适用于单像素荧光显微成像。针对二值优化不可导的问题，采用直通估计器（Straight-Through Estimator）。同时引入可学习的变分正则化，提升重建质量与鲁棒性。在CytoImageNet显微数据集上验证，所学模式在高度欠采样和数据稀缺条件下，性能优于基线方法及端到端深度学习模型。
原文摘要 · Abstract (English)
Single-Pixel Imaging (SPI) enables the reconstruction of objects using a single detector through sequential illuminations with structured light patterns. The choice of illumination patterns is critical, particularly in highly undersampled regimes, where it directly determines reconstruction quality and acquisition speed. Instead of relying on handcrafted or fixed patterns, we propose to learn task-specific patterns directly from data. Practical SPI hardware only supports binary patterns, making binary pattern design a necessary consideration. We propose a bilevel optimisation method for learning task-specific binary illumination patterns optimised for applications such as single-pixel fluorescence microscopy. We address the non-differentiable nature of binary optimisation using the Straight-Through Estimator. In addition, we incorporate learned variational regularisation, improving reconstruction quality and robustness. We demonstrate our method on the CytoImageNet microscopy dataset. We show that our learned patterns achieve superior reconstruction performance compared to baseline methods and end-to-end deep learning, particularly in highly undersampled regimes and in scarce-data settings.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。