CI-dataset

CI-dataset是一个专门为检测卫星图像中过小和过大的关键基础设施而设计的高质量数据集，由安达卢西亚数据科学与计算智能研究所创建。该数据集分为两个子集：CI-SS（小规模关键基础设施）和CI-LS（大规模关键基础设施），分别用于训练识别如变电站和机场等不同规模的基础设施。CI-dataset的构建过程遵循动态方法，通过分析Faster R-CNN模型的性能来指导数据集的扩展和优化。该数据集的应用领域包括安全、土地使用规划和变化检测，旨在解决在卫星图像中准确检测不同规模关键基础设施的挑战。

CI-dataset is a high-quality dataset specifically designed for detecting excessively small and large critical infrastructure in satellite imagery, created by the Andalusian Institute of Data Science and Computational Intelligence. This dataset is divided into two subsets: CI-SS (Small-scale Critical Infrastructure) and CI-LS (Large-scale Critical Infrastructure), which are respectively used to train models for identifying infrastructure of different scales such as substations and airports. The construction of CI-dataset follows a dynamic methodology, where the performance of the Faster R-CNN model is analyzed to guide the expansion and optimization of the dataset. Its application fields include security, land use planning and change detection, aiming to address the challenge of accurately detecting critical infrastructure of varying scales in satellite imagery.