Image-based yield prediction for tall fescue using random forests and convolutional neural networks

This record contains all data used in the research paper 'Image-based yield prediction for tall fescue using random forests and convolutional neural networks' by Ghysels, S., De Baets, B., Reheul, D. and Maenhout, S. 'Train_dataset.zip' and 'Test_dataset.zip' contain the RGB images of individual tall fescue plants, split into a training set and test set respectively. 'Multigras_data.csv' contains the dry matter yield measurements ('DMY (kg/ha)'), the breeder's evaluation scores ('Score MG') and the location of each individual plant on the field ('Blok_Rij_Plantnr', meaning Block-row-column).