Rob2Pheno Annotated Tomato Image Dataset



Dataset of RGBD images of tomato plants in a production greenhouse (Enza Zaden BV) obtained using Realsense cameras, and object instance level ground truth annotations of the fruit for using object detectors such as MaskRCNN or YOLACT. These data were used to obtain the results reported in our paper Tomato Fruit Detection and Counting in Greenhouses using Deep Learning, Frontiers in Plant Science, 2020.
Date made available25 Feb 2021
PublisherWageningen University & Research


  • tomato
  • greenhouse
  • computer vision in agriculture
  • deep learning
  • phenotyping

Cite this