最初，图像是从国家农业图像计划（NAIP）数据集中提取的。NAIP数据集由遍布整个美国大陆（CONUS）的共计33万个场景组成。作者使用了未压缩的数字正交四分之一四边形（DOQQ），这是GeoTIFF图像，该区域对应于美国地质调查局（USGS）地形四边形。平均图像平铺宽度约为6000像素，高度约为7000像素，每个平铺大小约为200 MB。CONUS的整个NAIP数据集约为65 TB。该图像是在1米地面采样距离（GSD）处采集的，其水平精度在距离照片可识别地面控制点6米以内。

图像由4个波段组成——红色、绿色、蓝色和近红外（NIR）。为了保持整个NAIP数据集固有的高方差，我们从覆盖整个加利福尼亚州的不同景观（如农村地区、城市地区、茂密的森林、山区、小到大的水体、农业地区等）的大量场景（总共1500个图像块）中采样图像块。作为本研究的一部分开发的图像标记工具用于手动标记属于特定土地覆盖类别的统一图像块。

一旦标记，从均匀图像块中提取28x28个非重叠滑动窗口块，并将其与相应的标签一起保存到数据集。我们选择28x28作为窗口大小，以保持显著更大的上下文，同时不使其大到足以在上下文窗口中删除目标类条件分布的相对统计属性。注意避免在选定和标记的图像块内的类间重叠。

Content

• Each sample image is 28x28 pixels and consists of 4 bands - red, green, blue and near infrared.

• The training and test labels are one-hot encoded 1x4 vectors

• The four classes represent the four broad land covers which include barren land, trees, grassland and a class that consists of all land cover classes other than the above three.

• Training and test datasets belong to disjoint set of image tiles.

• Each image patch is size normalized to 28x28 pixels.

• once generated, both the training and testing datasets were randomized using a pseudo-random number generator.

CSV files

• X_train_sat4.csv: 400,000 training images, 28x28 images each with 4 channels
  

• y_train_sat4.csv: 400,000 training labels, 1x4 one-hot encoded vectors
  

• X_test_sat4.csv: 100,000 training images, 28x28 images each with 4 channels
  

• y_test_sat4.csv: 100,000 training labels, 1x4 one-hot encoded vectors
  

The original MAT file

• train_x:    28x28x4x400000 uint8 (containing 400000 training samples of 28x28 images each with 4 channels)
  

• train_y:    400000x4 uint8 (containing 4x1 vectors having labels for the 400000 training samples)
  

• test_x:    28x28x4x100000 uint8 (containing 100000 test samples of 28x28 images each with 4 channels)
  

• test_y:    100000x4 uint8 (containing 4x1 vectors having labels for the 100000 test samples)
  

Acknowledgements

The original MATLAB file was converted to multiple CSV files

The original SAT-4 and SAT-6 airborne datasets can be found here:

http://csc.lsu.edu/~saikat/deepsat/