Example dataset of EOPatches for Slovenia 2019

Official Bucket Link:


Foreword

This example dataset will help you get started with Remote Sensing data and analysis in the open-source framework of eo-learn.

To promote the use of eo-learn, we have decided to share the dataset of EOPatches for the whole region of Slovenia for the year 2019. This data can be used in remote sensing applications, such as land cover classification.

See our Land Cover Classification blog posts!

Bucket Content

The contents of this AWS S3 bucket are the following:

Content Description

The full dataset contains 940 EOPatches of the size 5000 x 5000 pixels at 10 m resolution.

Each EOPatch is a container of EO and non-EO data. You can load an EOPatch in eo-learn with the following command:
EOPatch.load('download_path/eopatch_id_0_col_0_row_19')
Or directly from the AWS S3 bucket
EOPatch.load('s3://eo-learn.sentinel-hub.com/eopatches_slovenia_2019/eopatch_id_0_col_0_row_19')
You will obtain the EOPatch of the following structure:
EOPatch(
     data: {
      BANDS: numpy.ndarray(shape=(96, 500, 500, 6), dtype=float32)
      CLP: numpy.ndarray(shape=(96, 500, 500, 1), dtype=uint8)
      NDBI: numpy.ndarray(shape=(96, 500, 500, 1), dtype=float32)
      NDVI: numpy.ndarray(shape=(96, 500, 500, 1), dtype=float32)
      NDWI: numpy.ndarray(shape=(96, 500, 500, 1), dtype=float32)
     }
     mask: {
      CLM: numpy.ndarray(shape=(96, 500, 500, 1), dtype=uint8)
      IS_DATA: numpy.ndarray(shape=(96, 500, 500, 1), dtype=uint8)
      IS_VALID: numpy.ndarray(shape=(96, 500, 500, 1), dtype=bool)
     }
     mask_timeless: {
      LULC: numpy.ndarray(shape=(500, 500, 1), dtype=uint8)
      VALID_COUNT: numpy.ndarray(shape=(500, 500, 1), dtype=int64)
     }
     meta_info: {
      maxcc: 0.8
      service_type: 'processing'
      size_x: 500
      size_y: 500
      time_difference: datetime.timedelta(seconds=7200)
      time_interval: ('2019-01-01T00:00:00', '2019-12-31T23:59:59')
     }
     bbox: BBox(((370000.0, 5120000.0), (375000.0, 5125000.0)), crs=CRS('32633'))
     timestamp: [datetime.datetime(2019, 1, 1, 10, 7, 50), ..., datetime.datetime(2019, 12, 30, 10, 17, 47)], length=96
    )
It is possible to then access various EOPatch content via calls like:
eopatch.timestamp
eopatch.mask['LULC']
eopatch.data['CLM'][0]
eopatch.data['BANDS'][5][..., [3, 2, 1]]