调控而非约束：全色锐化中复杂光谱重建的自适应引导
Regulating Rather than Constraining: Adaptive Guidance for Complex Spectral Reconstruction in Pansharpening
提出一种与架构无关的正则化引导机制，通过MixShuffle数据增强和分层注意力加权损失，可迁移到其他需要处理复杂光谱混合的低层视觉任务中。
原文依据 we propose an architecture-agnostic regularization-guided mechanism that adaptively directs the model to focus on learning reliable reconstruction priors for challenging regions.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论在遥感全色锐化中，光谱混合区域——相邻地物间的光谱相互作用导致重建模式高度不一致——仍然是最具挑战性的区域。由于地物空间分布复杂且光谱特性异质，现有依赖刚性架构和物理约束的方法难以从有限的光谱混合样本中学习泛化的重建模式，导致泛化不稳定。为解决这一局限，我们提出一种与架构无关的正则化引导机制，自适应地引导模型专注于学习困难区域的可靠重建先验。具体而言，我们引入一种简单的数据级变换MixShuffle，通过在空间位置和光谱通道上进行随机凸组合，生成具有更丰富空间结构和更强光谱混合的训练数据。同时，我们提出一种分层注意力加权机制，一种在样本、通道和像素级别的损失级梯度重分配策略，使模型能够强调结构复杂的区域。在多个基准数据集（WV3、GF2、QB）和多种网络架构上的大量实验证明了所提策略的强大通用性和有效性，当集成到DANet中时达到了最优性能。我们的代码可在https://github.com/Geo-Tell/DANet获取。
英文原文摘要 1502 字符
In remote sensing pansharpening, spectrally mixed regions, where the spectral interactions among adjacent land covers lead to highly inconsistent reconstruction patterns, remain the most challenging areas. Due to the complex spatial distribution and heterogeneous spectral characteristics of ground objects, existing methods relying on rigid architectures and physical constraints struggle to learn generalized reconstruction patterns from limited spectral mixing samples, resulting in unstable generalization. To address this limitation, we propose an architecture-agnostic regularization-guided mechanism that adaptively directs the model to focus on learning reliable reconstruction priors for challenging regions. Specifically, we introduce a simple data-level transformation, MixShuffle, which performs random convex combinations across spatial positions and spectral channels to generate training data with richer spatial structures and stronger spectral mixing. In parallel, we propose a hierarchical attention weighting mechanism, a loss-level gradient reallocation strategy at the sample, channel, and pixel levels, enabling the model to emphasize structurally complex regions. Extensive experiments on multiple benchmark datasets (WV3, GF2, QB) and across various network architectures demonstrate the strong generality and effectiveness of the proposed strategies, achieving state-of-the-art performance when integrated into DANet. Our code is available at https://github.com/Geo-Tell/DANet.