Dataset

SwissCrop 2021, 2022, 2023, a three-year crop dataset covering the entirety of Switzerland from 2021 to 2023. SwissCrop combines Sentinel-2 Level-2A bottom-of-atmosphere multi-spectral imagery, averaging 110 timestamps per season, with annual crop-type labels for each field. The dataset includes 50 distinct crop types, each representing the primary crop grown in a field during a given season.

The ZueriCrop is a dataset for time-series instance segmentation of crops. 

It is a benchmark in the Pytorch TorchGeo.

Dataset features:

• Sentinel-2 multispectral imagery

• instance masks of 48 crop categories

• nine multispectral bands

• 116k images with 10 m per pixel resolution (24x24 px)

• ~28k time-series containing 142 images each