一种高分辨率SAR图像变化检测数据集制作方法
A Method for Generating High-Resolution SAR Image Change Detection Datasets
DOI: 10.12677/GST.2023.112013, PDF,    国家自然科学基金支持
作者: 李泽宇：北京建筑大学测绘与城市空间信息学院，北京；建筑遗产精细重构与健康监测北京市重点实验室，北京；赵江洪：北京建筑大学测绘与城市空间信息学院，北京；建筑遗产精细重构与健康监测北京市重点实验室，北京；自然资源部城市空间信息重点实验室，北京；地理信息工程国家重点实验室，陕西 西安 ；武汉大学测绘遥感信息工程国家重点实验室，湖北 武汉
关键词: 变化检测；合成孔径雷达图像；数据集；训练样本；Change Detection； SAR Image； Dataset； Training Sample
摘要: 针对可用于变化检测模型训练的高分辨率SAR图像数据不足、SAR图像处理与样本标注复杂的问题，本文开展SAR图像变化检测数据集制作方法研究，简化SAR图像处理与样本标注流程，对多幅高分三号SAR影像依次进行预处理、精配准、样本标注与数据增强等操作，得到一套高分辨率SAR图像变化检测数据集，为深度学习变化检测网络的训练与测试提供数据基础。通过多种网络模型训练的实验测试与结果分析，验证了该数据集能够有效支持多场景变化检测任务，对于不同变化检测模型训练精度与效率的提升具有较好的参考价值。
Abstract: Aiming at the problems of insufficient high-resolution SAR image data that can be used for change detection model training and the complexity of SAR image processing and sample labeling, this pa-per carried out research on the production method of SAR image change detection data set to sim-plify the SAR image processing and sample labeling process. A set of high-resolution SAR image change detection data set is obtained by preprocessing, fine registration, sample labeling and data enhancement of several high-resolution third-level SAR images successively, which provides data basis for the training and testing of deep learning change detection network. Through the experi-mental test and analysis of STANet, DSAMNet and SNUNet network model training, the validity of the SAR image change detection data set obtained by the data set making method proposed in this paper is verified in the multi-scene change detection task.
文章引用：李泽宇, 赵江洪. 一种高分辨率SAR图像变化检测数据集制作方法[J]. 测绘科学技术, 2023, 11(2): 115-121. https://doi.org/10.12677/GST.2023.112013

参考文献

[1] Ronneberger, O., Fischer, P. and Brox, T. (2015) U-Net: Convolutional Networks for Biomedical Image Segmentation. International Conference on Medical Image Computing and Computer-Assisted Intervention, 9351, 234-241. [Google Scholar] [CrossRef] 
[2] Daudt, R.C., Le Saux, B. and Boulch, A. (2018) Fully Convolutional Sia-mese Networks for Change Detection. 2018 25th IEEE International Conference on Image Processing (ICIP), Athens, 7-10 October 2018, 4063-4067. [Google Scholar] [CrossRef] 
[3] Peng, D., Zhang, Y. and Guan, H. (2019) End-to-End Change Detection for High Resolution Satellite Images Using Improved UNet++. Remote Sensing, 11, 1382. [Google Scholar] [CrossRef] 
[4] Mou, L., Bruzzone, L. and Zhu, X.X. (2018) Learning Spectral-Spatial-Temporal Fea-tures via a Recurrent Convolutional Neural Network for Change Detection in Multispectral Imagery. IEEE Transactions on Geoscience and Remote Sensing, 57, 924-935. [Google Scholar] [CrossRef] 
[5] Qu, X., Gao, F., Dong, J., et al. (2021) Change Detection in Synthetic Aperture Radar Images Using a Dual-Domain Network. IEEE Geoscience and Remote Sensing Letters, 19, 1-5. [Google Scholar] [CrossRef] 
[6] Chen, H. and Shi, Z. (2020) A Spatial-Temporal Attention-Based Method and a New Dataset for Remote Sensing Image Change Detection. Remote Sensing, 12, 1662. [Google Scholar] [CrossRef] 
[7] Shi, Q., Liu, M., Li, S., et al. (2021) A Deeply Supervised Attention Metric-Based Network and an Open Aerial Image Dataset for Remote Sensing Change Detection. IEEE Transactions on Geoscience and Remote Sensing, 60, 1-16. [Google Scholar] [CrossRef] 
[8] Huang, G., Liu, Z., Van Der Maaten, L., et al. (2017) Densely Connected Convolutional Networks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, 21-26 July 2017, 4700-4708. [Google Scholar] [CrossRef] 
[9] Zhou, Z., Siddiquee, M.M.R., Tajbakhsh, N., et al. (2019) Unet++: Redesigning Skip Connections to Exploit Multiscale Features in Image Segmentation. IEEE Transactions on Medical Imaging, 39, 1856-1867. [Google Scholar] [CrossRef] 
[10] Fang, S., Li, K., Shao, J., et al. (2021) SNUNet-CD: A Densely Connected Si-amese Network for Change Detection of VHR Images. IEEE Geoscience and Remote Sensing Letters, 19, 1-5. [Google Scholar] [CrossRef] 

为你推荐