*Update 19/01/26*:
As the challenge has ended and a number of people have now requested ground truth labels for hidden test set for the Public and Private leaderboards, I have uploaded them under the data tab in the solutions.csv for those who are interested in making full use of this dataset.

The HyperLeaf2024 challenge is part of the FGVC11 Workshop at CVPR 2024. It consists of 2410 hyperspectral images of wheat flag leaves, encompassing multiple strains and fertilizer levels and taken at different exposures and under different sunlight conditions. The challenge is aimed at developing models for fine-grained prediction of wheat field properties, plant health metrics, and yield.

Although there already exist publicly available drone and satellite hyperspectral imaging datasets. These typically have the goal of per-pixel image classification (one target per pixel). However, the power of hyperspectral imaging in full image classification and regression (one target per image) is less studied. We hope to promote this use case of hyperspectral imaging in wheat agriculture through this dataset and challenge.

![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F1859215%2Ff0a3e47521ccee9d259a6917058b183b%2Fleaves.png?generation=1709892667975111&alt=media)

# Citation
If you use this dataset in your research, please cite this [paper](https://openaccess.thecvf.com/content/CVPR2024W/FGVC11/papers/Laprade_HyperLeaf2024_-_A_Hyperspectral_Imaging_Dataset_for_Classification_and_Regression_CVPRW_2024_paper.pdf):

`William Michael Laprade, Pawel Pieta, Svetlana Kutuzova, Jesper Cairo Westergaard, Mads Nielsen, Svend Christensen, Anders Bjorholm Dahl; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2024, pp. 1234-1243`