Our repository consists of several data for SAR change detection.
1. Ottawa [Link]
2. Farmland [Link]
The Farmland dataset was acquired by RADARSAT-2 at the region of Yellow River Estuary in China in June 2008 and June 2009, respectively. The two images are single-look and four-look image, respectively. Hence, the influence of speckle noise on the image captured in 2008 is much greater than that of the one acquired in 2009.
3. Yellow River [Link]
The Yellow River dataset was acquired by RADARSAT-2 at the region of Yellow River Estuary in China in June 2008 and June 2009, respectively. The two images are single-look and four-look image, respectively. Hence, the influence of speckle noise on the image captured in 2008 is much greater than that of the one acquired in 2009.
4. San Francisco [Link]
The San Francisco dataset presents multitemporal SAR images acquired by the ERS-2 sensor. Both images are captured in August 2003 and May 2004, respectively. Please cite the following paper if this dataset helps your research: Feng Gao, Junyu Dong, Bo Li, Qizhi Xu, and Cui Xie, "Change detection from synthetic aperture radar images based on neighborhood-based ratio and extreme learning machine," Journal of Applied Remote Sensing, 10(4), 046019, Dec. 2016. doi: 10.1117/1.JRS.10.046019.
5. Sulzberger [Link]
The dataset are captured by the Envisat satellite on March 11 and 16, 2011. Both the images show the progression of the ice breakup. When the Tohoku Tsunami was triggered in the Pacific Ocean on March 11, 2011, massive ocean waves caused the ice shelf to flex and break. Please cite the following paper if this dataset helps your research: F. Gao, X. Wang, Y. Gao, J. Dong and S. Wang, "Sea Ice Change Detection in SAR Images Based on Convolutional-Wavelet Neural Networks," IEEE Geoscience and Remote Sensing Letters, vol. 16, no. 8, pp. 1240-1244, Aug. 2019, doi: 10.1109/LGRS.2019.2895656.
Please contact Feng Gao (gaofeng@ouc.edu.cn) for questions about the dataset.