In the field of remote sensing, the challenge of comparing images captured by disparate sensors is a common obstacle. This requires image translation -- converting imagery from one sensor domain to another while preserving the original content. Denoising Diffusion Implicit Models (DDIM) are potential state-of-the-art solutions for such domain translation due to their proven superiority in multiple image-to-image translation tasks in classic computer vision. However, these models struggle with large-scale multi-patch imagery, often focusing solely on small patches and resulting in inconsistencies across the full image. To overcome these limitations, we propose a novel method that leverages DDIM for effective optical image translation over large areas. Our approach is tailored to super-resolve large-scale low spatial resolution images into high-resolution equivalents from disparate optical sensors, ensuring uniformity across hundreds of patches. Extensive experiments with a dataset of paired Sentinel-II and Planet Dove images show that our approach provides precise domain adaptation and artifact reduction. Our technique preserves the image content while also improving radiometric (color) accuracy and feature representations. The outcome is a high-resolution large-scale image with consistent patches, vital for applications such as heterogeneous change detection (HCD). We present a unique training and testing algorithm rooted in DDIMs, a thorough image quality assessment, and a comparative study against the standard classifier-free guided DDIM framework and five other leading methods. The efficacy of our approach is further demonstrated by substantial enhancements in HCD tasks performed in the urban settings of Beirut, Lebanon, and Austin, USA.

翻译：在遥感领域，对比不同传感器获取的图像是一个常见挑战。这需要图像转换——将图像从一个传感器域转换到另一个传感器域，同时保持原始内容。去噪扩散隐式模型（DDIM）因其在经典计算机视觉中多个图像到图像转换任务上已证实的优越性，成为此类域转换的潜在前沿解决方案。然而，这些模型在处理大规模多图块图像时存在困难，往往仅关注小图块，导致整幅图像出现不一致性。为克服这些限制，我们提出一种新方法，利用DDIM实现大范围光学图像的有效转换。我们的方法专门用于将大规模低空间分辨率图像超分辨为来自不同光学传感器的高分辨率对应图像，确保数百个图块间的一致性。通过对Sentinel-II与Planet Dove配对图像数据集的广泛实验表明，我们的方法能提供精确的域适应并减少伪影。该技术在保持图像内容的同时，还提升了辐射（色彩）精度与特征表示能力。最终生成具有一致图块的高分辨率大规模图像，这对于异构变化检测（HCD）等应用至关重要。我们提出了一种基于DDIM的独特训练与测试算法、全面的图像质量评估，以及与标准无分类器引导DDIM框架及其他五种主流方法的对比研究。在黎巴嫩贝鲁特和美国奥斯汀城市环境中进行的HCD任务中，该方法带来的显著性能提升进一步证明了其有效性。