Skip to content

Image batch datasets for training and testing ag-net to recognize crop from Landsat imagery

License

Notifications You must be signed in to change notification settings

ZihengSun/Ag-Net-Dataset

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

12 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Ag-Net-Dataset

Example datasets for training and testing ag-net for crop recognition

This is related to ESIP Google Summer of Code project idea - AgNet

The input folder contains the 8bit version of seven bands of Landsat 8. Each file name is constructed in this pattern:

lc8_{observation_date}_{row number}_{column number}_{band number}.tif

The input_32bit contains the original values of Landsat 8. The valid value range is 0-10000. The file name structure is the same.

The input_rgb contains a RGB composite of Landsat 8 band 2, 3, 4. The file name structure is:

lc8_{observation_date}_{row number}_{column number}_rgb.tif

The target folder contains the corresponding tiles. The file name is composed of:

cdl_{landsat observation date}_{row number}_{column number}.tif

The value of CDL ranges from 0 to 255. Each value represents a crop type or non-crop land cover. The CDL value explanation table is here.

Please cite this dataset by:

Ziheng Sun. (2019). Ag-Net Sample Training Dataset [data files and tutorial notebook]. https://github.com/ZihengSun/Ag-Net-Dataset.

About

Image batch datasets for training and testing ag-net to recognize crop from Landsat imagery

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages