CAS滑坡数据集

在本研究中，我们推出了CAS滑坡数据集——一个由中国科学院·水利部成都山地灾害与环境研究所人工智能团队开发的大规模、多传感器数据集，用于基于深度学习的滑坡检测。该数据集旨在解决滑坡识别中遇到的挑战。随着气候变化和地震导致的滑坡事件日益频繁，迫切需要精确而全面的数据集来支持快速高效的滑坡识别。与现有数据集在规模、覆盖范围、传感器类型和分辨率方面的局限性相比，CAS滑坡数据集包含20,865张图像，整合了来自九个区域的卫星和无人机数据。为确保可靠性和适用性，我们建立了一套稳健的方法来评估数据集质量。我们建议将该滑坡数据集作为构建滑坡识别模型的基准，并推动深度学习技术的发展。研究人员可利用该数据集获得更强的预测、监测和分析能力，从而推进自动化滑坡检测进程。
若使用本数据，请引用我们发表于《Scientific Data》的论文。
Xu, Y., Ouyang, C., Xu, Q. et al. CAS Landslide Dataset: A Large-Scale and Multisensor Dataset for Deep Learning-Based Landslide Detection. Sci Data 11, 12 (2024). https://doi.org/10.1038/s41597-023-02847-z

In this study, we present the CAS Landslide Dataset—a large-scale, multisensor dataset developed by the AI team of the Chengdu Institute of Mountain Hazards and Environment, Chinese Academy of Sciences and Ministry of Water Resources, for deep learning-based landslide detection. This dataset aims to address the challenges encountered in landslide identification. With the increasing frequency of landslide events caused by climate change and earthquakes, there is an urgent need for accurate and comprehensive datasets to support rapid and efficient landslide identification. Compared to the limitations of existing datasets in terms of scale, coverage, sensor types, and resolution, the CAS Landslide Dataset contains 20,865 images and integrates satellite and unmanned aerial vehicle (UAV) data from nine regions. To ensure reliability and applicability, we established a robust methodology to evaluate the dataset’s quality. We propose this landslide dataset as a benchmark for constructing landslide identification models, thereby advancing the development of deep learning technologies. Researchers can utilize this dataset to achieve stronger prediction, monitoring, and analysis capabilities, thus facilitating the automation of landslide detection.
If you use this dataset, please cite our paper published in *Scientific Data*:
Xu, Y., Ouyang, C., Xu, Q. et al. CAS Landslide Dataset: A Large-Scale and Multisensor Dataset for Deep Learning-Based Landslide Detection. Sci Data 11, 12 (2024). https://doi.org/10.1038/s41597-023-02847-z

搜集汇总

背景与挑战

背景概述

CAS滑坡数据集是一个用于深度学习滑坡检测的大规模多传感器数据集，包含20865张图像，整合了卫星和无人机数据，覆盖9个区域。该数据集通过严格的质量控制流程，具有高质量、高可靠性和广泛适用性，可作为滑坡识别模型的基准数据集。