Most publicly accessible remote sensing data suffer from low resolution, limiting their practical applications. To address this, we propose a diffusion model guided by neural operators for continuous remote sensing image super-resolution (NeurOp-Diff). Neural operators are used to learn resolution representations at arbitrary scales, encoding low-resolution (LR) images into high-dimensional features, which are then used as prior conditions to guide the diffusion model for denoising. This effectively addresses the artifacts and excessive smoothing issues present in existing super-resolution (SR) methods, enabling the generation of high-quality, continuous super-resolution images. Specifically, we adjust the super-resolution scale by a scaling factor s, allowing the model to adapt to different super-resolution magnifications. Furthermore, experiments on multiple datasets demonstrate the effectiveness of NeurOp-Diff. Our code is available at https://github.com/zerono000/NeurOp-Diff.

翻译：大多数公开可用的遥感数据分辨率较低，限制了其实际应用。为解决此问题，我们提出了一种由神经算子引导的扩散模型，用于实现连续遥感图像超分辨率（NeurOp-Diff）。该模型利用神经算子学习任意尺度的分辨率表示，将低分辨率（LR）图像编码为高维特征，并以此作为先验条件引导扩散模型进行去噪。这有效解决了现有超分辨率（SR）方法中存在的伪影和过度平滑问题，能够生成高质量的连续超分辨率图像。具体而言，我们通过缩放因子 s 调整超分辨率尺度，使模型能够适应不同的超分辨率放大倍数。此外，在多个数据集上的实验验证了NeurOp-Diff的有效性。我们的代码发布于 https://github.com/zerono000/NeurOp-Diff。