Agriculture-Vision Database
Download
To download this year's challenge datasets, you will need to install AWS CLI.
Supervised dataset
Each field image in the supervised dataset has a file name in the format of (field id)_(x1)-(y1)-(x2)-(y2).(jpg/png). Each field id uniquely identifies the farmland that the image is cropped from, and (x1, y1, x2, y2) is a 4-tuple indicating the position in which the image is cropped. Please refer to our paper for more details regarding how we construct the dataset.
The supervised dataset can be downloaded with the following command:
aws s3 cp s3://intelinair-data-releases/agriculture-vision/cvpr_challenge_2021/supervised supervised --no-sign-request --recursive
Raw image dataset
The raw image dataset contains a sequence of 2-7 flights from 54 fields from 2017-2020 for a total of 261 full field, 10cm resolution images. Additionally, some of these fields contain sentinel imagery at 10m resolution. Each image can be identified by raw/(field id)/(flight id)_(color channel).tif. A boundary map is also provided for each field.
The raw field images are not pre-processed as those in the supervised dataset. Users can refer to our paper on how to convert them into NRGB images.
The raw image dataset can be downloaded with the following command:
aws s3 cp s3://intelinair-data-releases/agriculture-vision/cvpr_challenge_2021/raw raw --no-sign-request --recursive
Visualizations (annotated images)
Double plant
Drydown
Endrow
Nutrient Deficiency
Planter Skip
Water
Waterway
Weed Cluster
Mang Tik Chiu*, Xingqian Xu*, Yunchao Wei, Zilong Huang, Alexander Schwing, Robert Brunner, Hrant Khachatrian, Hovnatan Karapetyan, Ivan Dozier, Greg Rose, David Wilson, Adrian Tudor, Naira Hovakimyan, Thomas S. Huang, Honghui Shi
UIUC, IntelinAir, University of Oregon
Mang Tik Chiu*, Xingqian Xu*, Kai Wang, Jennifer Hobbs, Naira Hovakimyan, Thomas S. Huang, Honghui Shi
UIUC, IntelinAir, University of Oregon