A sample dataset for erosion gullies remote sensing recognition in typical Northeast China’s black soil region

The agricultural activities are considerably affected by soil erosion, with erosion gullies representing a prominent manifestation of this issue in Northeast China’s black soil region. Therefore, accurate identification and detection of erosion gullies based on remote sensing are pivotal for safeguarding arable land. Here we provide a sample dataset for erosion gullies remote sensing recognition in typical Northeast China’s black soil region, i.e., Hailun City of Heilongjiang Province. The very high resolution remote sensing image is derived from Google Earth with a spatial resolution of 0.6 m/pixel. Each sample is 224*224 pixels in size with two categories, i.e., erosion gullies and else. A total of 10,000 samples are provided, with positive and negative samples in a ratio of 1:1. Besides, the sample dataset is split into training, validation and testing dataset with a ratio of 6:2:2. The sample dataset is from the paper in National Remote Sensing Bulletin, please refer to this paper for details.