资源共享

	EnglishVersion

本站首页

小组动态

Large-scale deep learning based binary and semantic change detection

in ultra high resolution remote sensing imagery:

from benchmark datasets to urban application

1. Highlight of Hi-UCD dataset

1. 40800 pairs ultra-high resolution images (0.1 m) for Tallinn, Estonia.
2. Focus on refined urban semantic change detection.
3. Include 2 years of images, 9 land cover classes and 48 semantic change classes.
4. Tasks that can be performed on the dataset: semantic segmentation, binary change detection, semantic change detection.

Table 1. The public datasets for urban change detection with remote sensing imagery

2. Annotation format

The land cover labels of T1 and T2 are combined with their change labels to save as a PNG image with three channels. The first channel is T1 land cover labels, the second is T2 land cover labels and the last is change labels.

The numbers of land cover labels are listed as follows:

Table 2. The numbers of land cover labels

No.	Class name	Legend
0	unlabeld
1	water
2	grass
3	building
4	green house
5	road
6	bridge
7	others
8	bare land
9	woodland

The numbers of change labels are listed as follows:

Table 3. The numbers of change labels.

No.	Class name	Legend
0	unlabeled
1	no-change
2	change

3. Download

We hope that the release of the Hi-UCD dataset will promote the development of urban change detection. You can click the link below to download the data. You can click the link below to download the data:

● Baidu Drive: download

4. Evaluation Server

If you want to get the test scores, please join our hosted benchmark platform: semantic change detection

5. 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, China. The Hi-UCD dataset only can be used for academic purposes and need to cite the following paper, but any commercial use is prohibited. Any form of secondary development, including annotation work, is strictly prohibited for this dataset. Otherwise, RSIDEA of Wuhan University reserves the right to pursue legal responsibility.

[1] Tian, S., Zhong, Y., Zheng, Z., Ma, A., Tan, X., Zhang, L., 2022. Large-scale deep learning based binary and semantic change detection in ultra high resolution remote sensing imagery: From benchmark datasets to urban application. ISPRS Journal of Photogrammetry and Remote Sensing 193, 164–186.
[2] Tian, S., Tan, X., Ma, A., Zheng, Z., Zhang, L., Zhong, Y., 2023. Temporal-agnostic change region proposal for semantic change detection. ISPRS Journal of Photogrammetry and Remote Sensing 204, 306–320.

6.Contact

If you have any the problem or feedback in using Hi-UCD dataset, please contact:
Dr. Shiqi Tian: shiqitian@whu.edu.cn
Prof. Ailong Ma: maailong007@whu.edu.cn
Prof. Yanfei Zhong: zhongyanfei@whu.edu.cn