Due to the inevitable presence of quality problems, remote sensing image quality inspection is indeed an indispensable step between the acquisition and the application of remote sensing images. However, traditional manual inspection suffers from low efficiency. Hence, we propose a novel deep learning-based two-step intelligent system consisting of multiple advanced computer vision models, which first performs image classification and then accordingly adopts the most appropriate method, such as semantic segmentation, to localize the quality problems. Results demonstrate that the proposed method exhibits excellent performance and efficiency, surpassing traditional methods. Furthermore, we conduct an initial exploration of applying multimodal models to remote sensing image quality inspection.

翻译：由于质量问题难以完全避免，遥感图像质量检测是连接遥感图像获取与应用之间不可或缺的关键环节。然而，传统的人工检测存在效率低下的问题。为此，我们提出一种基于深度学习的新型两步智能系统，该系统集成了多种先进的计算机视觉模型：首先进行图像分类，随后根据分类结果采用最适宜的方法（如语义分割）对质量问题进行定位。实验结果表明，该方法在性能与效率上均显著优于传统检测手段。此外，本文还初步探索了多模态模型在遥感图像质量检测中的应用。