|
|
|
||||||||||||
|
|||||||||||||||
|
|
|
WHU-RuR+: A Large-Scale Dataset for Rural Road extraction
in High-resolution Remote Sensing Imagery |
||||
|
|
||||
|
Abstract
To meet the application needs of large-scale rural road mapping, a large-scale high-resolution rural road extraction dataset based on deep learning, namely the WHU-RuR+ dataset, has been constructed. It contains 36,098 pairs of 1024×1024 images and annotations, covering diverse and representative rural areas around the world, providing a comprehensive resource for large-scale rural road research.The WHU-RuR+ dataset is collected and shared by the RSIDEA research group of Wuhan University and can be used as a benchmark dataset for rural road extraction worldwide. 1.The WHU-RuR+ dataset 1.1 Study areas and data sources The research area of the dataset covers China, the United States, Russia, Canada, Argentina, India, Kenya, and Australia, including regions of different scales, structures, development levels, and historical cultures. Fig. 1 shows the study area covered by the dataset. The images in the WHU-RuR+ dataset are from Google Earth (Google Inc.), and all images are annotated by image interpretation experts.
1.2 Dataset processing The dataset consists of aerial images with a spatial resolution of 0.3 m-0.8 m. The coverage area is about 6866.35 square kilometers. We cropped the images to 1024×1024, which is more suitable for deep learning. Finally, 36098 pairs of images with semantic labels were obtained. In order to better verify the generalization ability of the model, we divided the training set and the test set in a ratio close to 1:1, with 18,103 pairs forming the training set and 17,995 pairs forming the test set. To mitigate potential biases introduced by data partitioning, we divided the dataset based on geographical regions. This ensures that the training and testing datasets come from different geographical areas, helping to validate the model's robustness and practicality under various conditions. Fig.2 shows the visualization of some annotations in the WHU-RuR+ dataset. According to the different road surface materials, the roads in the WHU-RuR+ dataset can be divided into cement roads, asphalt roads, and dirt roads; according to the diversity of road types, the dataset includes field roads, forest roads, mountain roads, village roads, and highways.
1.3 Download We provide download links of the WHU-RuR+ dataset on Baidu Drive and MEGA. We hope you can fill in a simple questionnaire before downloading, which will appear after clicking the following link: ● Baidu Drive: download ● MEGA : download 2.Copyright The copyright belongs to Intelligent Data Extraction, Analysis and Applications of Remote Sensing(RSIDEA) academic research group, State Key Laboratory of Information Engineering in Surveying, Mapping, and Remote Sensing (LIESMARS), Wuhan University. The WHU-RuR+ dataset can be used for academic purposes only and need to cite the following paper, but any commercial use is prohibited. Otherwise, RSIDEA of Wuhan University reserves the right to pursue legal responsibility. Reference: [1] Ningjing Wang, Xinyu Wang, Yang Pan, Wanqiang Yao, Yanfei Zhong (2024) Large-Scale Rural Road Mapping with Reverse Refinement Network: From Benchmark Dataset to Method[J]. ISPRS Journal of Photogrammetry and Remote Sensing,2024 (Under Review). 3.Contact If you have any the problem or feedback in using WHU-RuR+ dataset, please contact: Ms. Ningjing Wang: 1121906691@qq.com Dr. Xinyu Wang: wangxinyu@whu.edu.cn Prof. Yanfei Zhong: zhongyanfei@whu.edu.cn |
|
