ROSE-X: an annotated data set for evaluation of 3D plant organ segmentation methods

CVPPPA @ ICCV 2021 Workshop Challenge Dataset. Given RGB stereo images of plots of Sorghum, estimate the end-of-season biomass. This dataset contains 277,327 images of 176 sorghum species with biomass yields from 349 plots.

CVPR 2020 Agri-Vision Workshop Challenge. This challenge dataset contains 21,061 aerial farmland images captured throughout 2019 across the US. Each image consists of four 512x512 color channels, which are RGB and Near Infra-red (NIR). Each image also has a boundary map and a mask. The boundary map indicates the region of the farmland, and the mask indicates valid pixels in the image.

The dataset contains six types of annotations: Cloud shadow, Double plant, Planter skip, Standing Water, Waterway and Weed cluster.

A field-level fruit localization dataset containing 1,700 annotated images of mango plants.

This dataset contains multimodal images of arabidopsis bean plants. The represented modalities are fluorescence, infrared(IR), RGB color, and depth. There are 16 arabidopsis plants and 5 bean plants tracked across several days, with each image having a corresponding segmentation mask for each leaf.

For related papers, click here.

This dataset contains the ground truth data for the CVPPP leaf segmentation challenge and the leaf counting challenge. It contains several images of arabidopsis plants with associated masks for each leaf.

For related papers, click here.

This dataset contains images of flowering pear, apple, and peach plants under different lighting conditions. The dataset contains instance segmentation masks for the flowers, and there are 190 images. The intention for this dataset is that the network is trained on 100 images of apple flowers and the network parameters are transferred to the task of instance segmentation for the flowers of other plant species.

For related papers, click here.

3D scans of shoots and roots for different plant species

Vitis International Variety Catalogue (VIVC) contains an inventory of Vitis species, varieties and other Vitis genotypes existing in the grapevine collections world-wide

The dataset consists of synthetic images of grass and clover mixtures with pixelwise species labels in a hierarchy, collected images of grass and clover mixtures with biomass composition labels (grass, clover, weeds), and unlabeled images of grass and clover mixtures.

An open-access reference dataset for the study of plant sensing.

Sensor data were generated by a field scanner sensing platform that captures color, thermal, hyperspectral, and active flourescence imagery as well as three dimensional structure and associated environmental measurements. This dataset is provided alongside data collected using traditional field methods.

This dataset contains expert-annotated, manually-captured, high-quality, real-life symptom images of multiple apple foliar diseases, with variable illumination, angles, surfaces, and noise. It is a pilot dataset for apple scab, cedar apple rust, and healthy leaves, which was made available to the Kaggle community for 'Plant Pathology Challenge'; part of the Fine-Grained Visual Categorization (FGVC) workshop at CVPR 2020 (Computer Vision and Pattern Recognition).

List of aerial and satellite imagery datasets with annotations for computer vision and deep learning.

Several resources for synthetic plant generation in virtual scenes are listed in this section.

Procedural generation algorithms: A survey containing popular systems that implement procedural techniques to model plants and natural phenomena. Some of the resources mentioned are listed below.

• FractTree

• L-System4

• LStudio

• An Ivy Generator

• TreeGenerator

• TreeMagik G3

• Mesh Tree Studio

Software Packages: A table comparing different software tools for plant generation can be found here. Examples of said tools are listed below.

• XFrog

• SpeedTree

• Lenné3D

• ngPlant

• OpenAlea