SGDE：面向编码孔径压缩光谱成像的自监督几何退化估计框架
SGDE: Self-supervised Geometry Degradation Estimation Framework for Coded Aperture Compressive Spectral Imaging
将掩码错位建模为仿射变换并嵌入成像模型，以自监督方式联合估计退化参数与重建高光谱图像，该思路可迁移到其他对硬件对准敏感的编码成像系统（如编码曝光、结构光）中做在线退化校正。
原文依据 we propose a Self-supervised Geometry Degradation Estimation (SGDE) framework that explicitly models mask misalignments as an affine transformation and embeds it into the imaging model.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论编码孔径快照光谱成像（CASSI）已成为高效高光谱成像的重要技术。然而，物理编码与计算解码之间的紧密耦合使CASSI对轻微的硬件失准高度敏感，这会显著降低重建质量。现有方法要么假设理想成像条件，要么依赖离线校准，因而容易受到动态扰动（如热膨胀和机械振动）的影响，这些扰动会导致掩码偏移。为解决这些局限，我们提出了一种自监督几何退化估计（SGDE）框架，该框架将掩码失准显式建模为仿射变换并将其嵌入成像模型。SGDE以自监督方式联合估计仿射参数并重建高光谱图像，无需参考目标或设备特定的训练数据。此外，我们引入了一种多核估计策略，以增强大扰动下退化估计的鲁棒性。在模拟和真实数据集上的大量实验表明，SGDE对掩码失准具有优越的鲁棒性。此外，估计的仿射参数可直接集成到现有重建算法中，为实际CASSI系统实现即插即用的退化估计。代码可在 https://github.com/heyuqiao/SGDE 获取。
英文原文摘要 1468 字符
Coded Aperture Snapshot Spectral Imaging (CASSI) has emerged as a prominent technique for efficient hyperspectral imaging. However, the tight coupling between physical encoding and computational decoding makes CASSI highly sensitive to slight hardware misalignments, which can significantly degrade reconstruction quality. Existing methods either assume ideal imaging conditions or rely on offline calibration, making them vulnerable to dynamic perturbations, such as thermal expansion and mechanical vibrations, which cause mask shifts. To address these limitations, we propose a Self-supervised Geometry Degradation Estimation (SGDE) framework that explicitly models mask misalignments as an affine transformation and embeds it into the imaging model. SGDE jointly estimates affine parameters and reconstructs the hyperspectral image in a self-supervised manner, eliminating the need for reference targets or device-specific training data. Furthermore, we introduce a multi-kernel estimation strategy to enhance the robustness of degradation estimation under large perturbations. Extensive experiments on both simulated and real-world datasets demonstrate that SGDE achieves superior robustness against mask misalignments. Moreover, the estimated affine parameters can be directly integrated into existing reconstruction algorithms, enabling plug-and-play degradation estimation for practical CASSI systems. The code is available at https://github.com/heyuqiao/SGDE.