资源共享

	EnglishVersion

本站首页

小组动态

Identifying cropland non-agriculturalization with high representational

consistency from bi-temporal high-resolution remote sensing images:

From benchmark datasets to real-world application

Abstract

The Hi-CNA is a high-resolution remote sensing dataset dedicated to the cropland non-agriculturalization (CNA) tasks, featuring high-quality semantic and change annotations for cropland.The study area covers parts of Hebei, Shanxi, Shandong, and Hubei provinces in China, with a total area exceeding 1100 km². These regions exhibit significant variations in crop planting, ensuring the diversity of cropland morphologies. The first temporal phase spans from 2015 to 2017,while the second phase ranges from 2020 to 2022, covering multiple phenological periods of crops.These characteristics provide a rich variety of samples for CNA tasks.

1.The Hi-CNA dataset

The dataset is sourced from multispectral GF-2 fusion images with a spatial resolution of 0.8m,encompassing four bands including visible light and near-infrared. All images are cropped to 512*512, resulting in a total of 6797 pairs of dual-temporal images with corresponding annotations. An overview of this dataset is provided in Fig. 1.

Fig. 1. The overview of the Hi-CNA dataset.(a) Study area. (b) Primary change types. (c) Images and annotations of selected scenes in Hi-CNA

2.Experiment

Table 1 presents the testing accuracy of multiple change detection methods on the Hi-CNA dataset.

Table. 1. Test results of multiple change detection methods on the Hi-CNA dataset.

Method	OA(%)	UA(%)	PA(%)	IoU(%)	F1(%)
FC-EF[1]	91.41	72.2	69.89	55.07	71.03
FC-Siam-conc[1]	92.06	73.5	73.95	58.39	73.73
FC-Siam-diff[1]	87.86	58.23	68.68	46.01	63.03
SNUNet[2]	93.56	81.16	74.55	63.55	77.72
MSCANet[3]	92.71	75.85	75.66	60.97	75.76
BiT[4]	92.87	77.69	73.38	60.95	75.74
HANet[5]	92.79	77.98	73.63	60.27	75.21
DTCDSCN[6]	93.17	77.12	77.74	63.17	77.43
HRSCD Str.4[7]	91.91	71.32	77.38	59.02	77.23
ChangeMask[8]	92.97	80.11	75.17	62.72	77.2
CNANet[9](proposed)	93.81	81.1	76.83	65.16	78.9

3.Download

We hope that the release of the Hi-CNA dataset will promote the development of CNA detection. You can click the link below to download the data:
A large-scale high-resolution cropland non-agriculturalization (Hi-CNA) dataset
● Baidu Drive: download
● Google Drive: download

4.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-CNA 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.

Sun Z, Zhong Y, Wang X, et al. Identifying cropland non-agriculturalization with high representational consistency from bi-temporal high-resolution remote sensing images: From benchmark datasets to real-world application[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2024, 212: 454-474.
● 论文链接

5.Contact

If you have any the problem or feedback in using Hi-CNA dataset, please contact:
Mr. Zhendong Sun: sunzhendong@whu.edu.cn
Dr. Xinyu Wang: wangxinyu@whu.edu.cn
Prof. Yanfei Zhong: zhongyanfei@whu.edu.cn

Reference:

[1] Daudt, R.C., Le Saux, B., Boulch, A., 2018. Fully convolutional siamese networks for change detection, 2018 25th IEEE International Conference on Image Processing (ICIP). IEEE, pp. 4063-4067.
[2] Fang, S., Li, K., Shao, J., Li, Z., 2021. SNUNet-CD: A densely connected Siamese network for change detection of VHR images. IEEE Geoscience and Remote Sensing Letters 19, 1-5.
[3] Liu, M., Chai, Z., Deng, H., Liu, R., 2022. A CNN-transformer network with multiscale context aggregation for fine-grained cropland change detection. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 15, 4297-4306.
[4] Chen, H., Qi, Z., Shi, Z., 2021. Remote sensing image change detection with transformers. IEEE Transactions on Geoscience and Remote Sensing 60, 1-14.
[5] Han, C., Wu, C., Guo, H., Hu, M., Chen, H., 2023. HANet: A hierarchical attention network for change detection with bi-temporal very-high-resolution remote sensing images. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.
[6] Liu, Y., Pang, C., Zhan, Z., Zhang, X., Yang, X., 2020. Building change detection for remote sensing images using a dual-task constrained deep siamese convolutional network model. IEEE Geoscience and Remote Sensing Letters 18, 811-815.
[7] Daudt, R.C., Le Saux, B., Boulch, A., Gousseau, Y., 2019. Multitask learning for large-scale semantic change detection. Computer Vision and Image Understanding 187, 102783.
[8] Zheng, Z., Zhong, Y., Tian, S., Ma, A., Zhang, L., 2022. ChangeMask: Deep multi-task encoder-transformer-decoder architecture for semantic change detection. ISPRS Journal of Photogrammetry and Remote Sensing 183, 228-239.
[9] Sun Z, Zhong Y, Wang X, et al. 2023. Identifying cropland non-agriculturalization with deep representational consistency from bi-temporal high-resolution remote sensing images: From benchmark datasets to real-world application. ISPRS Journal of Photogrammetry and Remote Sensing [J], under review.