Datasets for Agricultural research

This is a collection of datasets for either image or 3D object detection or segmentation, that can be be download from the internet and are already found in FOLA (Agroscope internal network) For the case of 3D data, the dataset may or may not be annotated. PLEASE let me know if you know about another dataset :)

2D

| Name | Identification | Class | Type | Paper | Authors | Year | URL | URL2 | Description | Modifications | Status | URL: | |:----------------------------------------|:-------------------|:---------------------------------------------|:----------------------------------------------|:--------------------------------------------------------------------------------------------------|:-------------------------------------------|:-----------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------|:----------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------|:-----------------------------------------------------------------------------------| | RumexWeeds | RumexWeeds | ['Plant'] | ['object detection', 'instance segmentation'] | RumexWeeds: A grassland dataset for agricultural robotics | Güldenring et al | 2023 | https://dtu-pas.github.io/RumexWeeds/ | https://data.dtu.dk/ndownloader/files/39268307 | Rumex detection and segmentation | | nan | nan | | Flowers | WSUA | ['flowers', 'apple', 'peach', 'pear'] | ['object detection'] | | Dias et al | 2018 | | | | | nan | nan | | ACFR Orchard Fruit Dataset | acfr-fruit-almonds | ['fruit', 'almond'] | ['object detection'] | Image Based Mango Fruit Detection, Localisation and Yield Estimation Using Multiple View Geometry | Underwood and Bargoti | 2016 | https://data.acfr.usyd.edu.au/ag/treecrops/2016-multifruit/ | | Small tilles of fruits in trees | 20240411: Was divided in val, train and test using the txt provides by authors | DONE | nan | | AppleScabFDs | ASF | ['fruit', 'apple'] | ['classification'] | | Kodors et al | 2021 | https://www.kaggle.com/datasets/projectlzp201910094/applescabfds | | | | Behind | nan | | WSUapple | WSUA | ['fruit', 'apple'] | Object detection | | Santosh Bhusal, Manoj Karkee and Qin Zhang | 2019 | https://rex.libraries.wsu.edu/esploro/outputs/dataset/Apple-Dataset-Benchmark-from-Orchard-Environment/99900502619401842#details | | Apple Dataset Benchmark from Orchard Environment in Modern Fruiting Wall | This dataset contain multiple datasets | nan | nan | | Agroscope apple | Agroscopeapple | ['fruit', 'apple'] | ['object detection'] | | Chiang | 2024 | | | Agroscope dataset for apple fruit counting | Labels from ppt to txt | DONE | nan | | acfr-multifruit-2016 | AM | ['fruit', 'apple'] | ['object detection'] | | Bargoti et al. | 2016 | http://data.acfr.usyd.edu.au/ag/treecrops/2016-multifruit/ | | The dataset was gathered by the agriculture team at the Australian Centre for Field Robotics | Only apples were keep. The original dataset have circles as annotations | nan | nan | | ACFR Orchard Fruit Dataset | acfr-fruit-apples | ['fruit', 'apple'] | ['object detection', 'instance segmentation'] | Image Based Mango Fruit Detection, Localisation and Yield Estimation Using Multiple View Geometry | Underwood and Bargoti | 2016 | https://data.acfr.usyd.edu.au/ag/treecrops/2016-multifruit/ | | Small tilles of fruits in trees | 20240411: Was divided in val, train and test using the txt provides by authors | DONE | nan | | KFuji RGB-DS database | KFS | ['fruit', 'apple'] | ['object detection'] | | Gené-Mola et al. | 2019 | http://www.grap.udl.cat/en/publications/KFujiRGBDSdatabase.html | https://zenodo.org/record/3715991#.YrQqfNJByV4 | | annotations were changed from x1-y1-widh-height to x1y1x2y2 format | nan | nan | | Minneapple | MA | ['fruit', 'apple'] | ['Instance segmentation'] | | Häni et al | 2019 | https://github.com/nicolaihaeni/MinneApple | | 1000 apples tree images with more than 41k manual annotations | DONE, USED | nan | nan | | WSUappledepth | WSUAD | ['fruit', 'apple'] | ['object detection'] | | Longsheng Fu, Manoj Karkee and Qin Zhang | 2020 | https://rex.libraries.wsu.edu/esploro/outputs/dataset/Scifresh-Apple-Orignial-and-DepthFilter-RGB/99900501726801842 | | Scifresh Apple Orignial and DepthFilter RGB Images | | nan | nan | | PAppleRGB-D-Size dataset | PAS | ['fruit', 'apple'] | ['instance segmentation'] | | Ferrer Ferrer M et al 2022 | 2022 | http://www.grap.udl.cat/en/publications/PAppleRGB-D-Size.html | https://gofile-36514d3739.fr3.quickconnect.to/sharing/brPZduTyi | | annotations were changed from instance segmentation to object segmentation x1y1x2y2 format | nan | nan | | deepFruits | DF | ['fruit', 'apple', 'capsicum', 'strawberry'] | ['object detection'] | | Inkyu et al | 2016 | hhttp://enddl22.net/wordpress/datasets/deepcrops-datasets-and-annotation-tool | | Object detection on 7 species. Only three were downloaded (apple, capsicum and strawberry) | Different size. | nan | nan | | Deepblueberry | DB | ['fruit', 'blueberry'] | ['object detection', 'instance segmentation'] | | Gonzalez et al. | 2019 | https://ieeexplore.ieee.org/document/8787818 | 10.1109/ACCESS.2019.2933062 | Blue berry detection on 293 images and instance segmentation on 7 images | Annotation files were modified as were not done in VIA. Additionaly the instance segmentation pictures (7) were rotated as they dont match the provided pictures | nan | nan | | WGNbroccoliRGBD | WGNB | ['fruit', 'broccoli'] | ['instance segmentation'] | | Blok 2021 | 2021 | https://data.4tu.nl/articles/dataset/DataunderlyingthepublicationImage-basedsizeestimationofbroccoliheadsundervaryingdegreesofocclusion/13603787 | | RGBD for annodal network | | nan | nan | | ACFR Orchard Fruit Dataset | acfr-fruit-mangoes | ['fruit', 'mango'] | ['object detection'] | Image Based Mango Fruit Detection, Localisation and Yield Estimation Using Multiple View Geometry | Underwood and Bargoti | 2016 | https://data.acfr.usyd.edu.au/ag/treecrops/2016-multifruit/ | | Small tilles of fruits in trees | 20240411: Was divided in val, train and test using the txt provides by authors | DONE | nan | | MOrangeT | MOR | ['fruit', 'orange'] | ['object detection'] | | Santos et al | 2024 | nan | | Oranges in trees. include green oranges | | nan | https://www.redape.dados.embrapa.br/dataset.xhtml?persistentId=doi:10.48432/OI7BFG | | strawberry Digital Images | SDI | ['fruit', 'strawberry'] | ['instance segmentation'] | | Perez-Borrero et al | 2020 | https://strawdi.github.io/ | | 3.1K images of strawberries on field | DONE, USED | nan | nan | | strawberry-semantic-segmentation | SSS | ['fruit', 'strawberry'] | ['instance segmentation'] | | NaN | NaN | https://www.kaggle.com/datasets/woodiedudy/strawberry-segmentation-dataset | | 141 strawberries images with masks for berries, leaves, stems and flowers | | nan | nan | | strawberry-disease-detection-dataset | SD | ['fruit', 'strawberry'] | ['object detection'] | | Afzaal et al | 2021 | https://www.kaggle.com/usmanafzaal/strawberry-disease-detection-dataset | | 2500 images for 7 diferent diseases in strawberry | | nan | nan | | strawberry-dataset-for-object-detection | SDO | ['fruit', 'strawberry'] | ['object detection'] | | Pastell et al | 2022 | https://zenodo.org/record/6126677#.YrQMe9JByV5 | https://www.luke.fi/en/projects/poimintarobottieip-01 | 813 images in two classes | Reclassfied to 4 different classes | nan | nan | | strawberry-skripsie | SSK | ['fruit', 'strawberry'] | ['object detection'] | | NaN | 2021 | https://universe.roboflow.com/skripsie/strawberry.00/15 | | 450 images in one class | Reclassfied to 4 different classes | nan | nan | | strawberry detection | STL | ['fruit', 'strawberry'] | ['instance segmentation'] | | Perez-Borrero et al | 2020 | https://strawdi.github.io/ | | 3.1K images of strawberries on field | DONE, USED | nan | nan | | laborotomato | LT | ['fruit', 'tomato'] | ['object detection', 'instance segmentation'] | | Laboro AI | 2019 | https://github.com/laboroai/LaboroTomato | | Tomatos (Normal and cherry) in three maturity status | Added object detection in YOLOv5 Format | DONE | nan | | tomatOD | TD | ['fruit', 'tomato'] | ['object detection'] | | Tsironis et al. 2020 | 2021 | https://github.com/up2metric/tomatOD | | 277 images for 2418 annotated tomato fruits in three categories (1592 unripe, 395 semi-ripe, 431 fully ripe | | nan | nan | | Rob2Pheno | RP | ['fruit', 'tomato'] | ['instance segmentation'] | | Afonso et al | 2021 | https://research.wur.nl/en/datasets/rob2pheno-annotated-tomato-image-dataset | | 123 RGBD pictures took with Realsense D435 | | nan | nan | | tomatodetection | TE | ['fruit', 'tomato'] | ['object detection'] | | ?? | 2020 | https://makeml.app/datasets/tomato | | 895 images for tomatoes without categories (class =1, tomatoes) | | nan | nan | | Plantphenotyping | PP | ['leaf'] | ['Instance segmentation'] | | Plant phenotyping | 2014, 2015, 2017 | https://www.plant-phenotyping.org | | 3 Datasets for leaf instance segmentation on Arabidopsis and Tobacco | | nan | nan | | PlantVillage | PV | ['leaf'] | ['classification'] | | Hugues and Salathe | 2016 | https://data.mendeley.com/datasets/tywbtsjrjv/1 | https://www.tensorflow.org/datasets/catalog/plantvillage | PlantVillage dataset consists of 54303 healthy and unhealthy leaf images divided into 38 categories by species and disease | | nan | nan | | PlantDoc | PD | ['leaf'] | ['object detection'] | | Singh et al. | 2020 | https://github.com/pratikkayal/PlantDoc-Dataset | https://public.roboflow.com/object-detection/plantdoc | Disease detection on 13 plant species | resized to 416 x 416 thanks to roboflow | nan | nan | | AppleScabLDs | ASL | ['leaf', 'apple'] | ['classification'] | | Kodors et al | 2021 | https://www.kaggle.com/datasets/projectlzp201910094/applescablds | | | | Behind | nan | | Plant-pathology-2020-fgvc7 | PPA | ['leaf', 'apple'] | ['instance segmentation'] | | Thapa et al. 2020 | 2020 | https://www.kaggle.com/c/plant-pathology-2020-fgvc7 | | Disease detection in leaves of apple trees. | | nan | nan | | downlymildewimages | DM | ['leaf', 'grape'] | ['instance segmentation'] | | Abdelghafour et al | 2021 | https://pubmed.ncbi.nlm.nih.gov/34258341/ | | Instance segmentation in grape, specialy focus on downly mildew | WAITING TO BE PROCESSED | nan | nan | | strawberry tipburn detection | STD | ['leaf', 'strawberry'] | ['classification'] | | Hairi and Avsar | 2022 | https://www.kaggle.com/datasets/ercanavsar/images-of-strawberry-leaves-for-tipburn-detection | | | | nan | nan | | tomatoleaves_disease | TLD | ['leaf', 'tomato'] | ['classification'] | | NaN | 2021 | https://www.kaggle.com/datasets/kaustubhb999/tomatoleaf | | | | nan | nan |

3D

| Name | Identification | Class | Paper | Authors | Year | URL | URL2 | Equipment | Description | Modifications | Status | |:---------------------------|:---------------------------|:--------------------|:-------------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------------------------------------------------|-------:|:--------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------|:-----------------------------------------|:---------------------------------------------------------------------------------------------------------|:----------------|:---------| | Grapevineprunningdata | Grapevineprunningdata | Bush - No leaves | 3D Skeletonization of Complex Grapevines for Robotic Pruning | Schneider et al | 2023 | https://labs.ri.cmu.edu/aiira/resources/ | https://drive.google.com/drive/folders/1Oi01eBknf8hUb0zXSWHcSAw2sB1OJcz | RGBD - PointGrey CM3 | Wine plants | None | Raw | | Blueberries | Blueberries | Bush - With leaves | 3D point cloud data to quantitatively characterize size and shape of shrub crops | Jiang | 2019 | https://doi.org/10.1038/s41438-019-0123-9 | https://figshare.com/s/2abb4eeadfda4103545b | ZEB1 scanner | 47 bushes of raspberry plants with leaves | None | Raw | | ROSE-X | ROSE-X | Bush - With leaves | ROSE-X: an annotated data set for evaluation of 3D plant organ segmentation methods | Dutagaci et al | 2020 | https://plantmethods.biomedcentral.com/articles/10.1186/s13007-020-00573-w | | Siemens X-ray | Rose plants on 3D | None | Raw | | VineLIDAR | VineLIDAR | Bush - With leaves | High resolution LiDAR dataset acquired using UAV (unmanned aerial vehicle) over two vineyards and two years located in 'Tomiño', Pontevedra, Spain | Vélez, S., Ariza-Sentís, M., & Valente, J. | 2023 | https://zenodo.org/records/8113105 | | DJI Zenmuse L1 | High-resolution UAV-LiDAR vineyard dataset acquired over two years in northern Spain | None | Raw | | Brocoli | Brocoli | Organ | Image-based size estimation of broccoli heads under varying degrees of occlusion | Blok, P., van Henten, E., van Evert, F. and Kootstra, G. | 2021 | https://doi.org/10.1016/j.biosystemseng.2021.06.001 | https://git.wur.nl/blok012/sizecnn | RGBD - Realsense D435 | Brocoli heads for occlusion studies | None | Raw | | CVPPA@ECCV2024bellpeper | CVPPA@ECCV2024bellpeper | Organ | Efficient and Accurate Transformer-Based 3D Shape Completion and Reconstruction of Fruits for Agricultural Robots | Magistri et al | 2024 | https://www.ipb.uni-bonn.de/data/shapecompletion/index.html | | RGBD - Realsense D435 | Sweet peper rgb frames for reconstruction | None | Raw | | BiomassevaluationLIDAR | BiomassevaluationLIDAR | Trees - No leaves | Advancing Fine Branch Biomass Estimation with Lidar and Structural Models | | 2024 | https://github.com/VEZY/BiomassevaluationLiDAR | | Riegl VZ-400 | Walnut trees without leaves | None | Raw | | cacaocameroon | cacaocameroon | Trees - With leaves | Terrestrial LiDAR point cloud dataset of cocoa trees grown in agroforestry systems in Cameroon | Peynaud, E. and Momo, S. | 2024 | https://doi.org/10.1016/j.dib.2024.110108 | https://dataverse.cirad.fr/dataset.xhtml?persistentId=doi:10.18167/DVN1/5HZB1F | Leica C10 | Cocoa tree point clouds obtained by terrestrial Lidar scanning (TLS) in agroforestry systems in Cameroon | None | Raw | | FOR-instance | FOR-instance | Trees - With leaves | FOR-instance (FOR-instance: a UAV laser scanning benchmark dataset for semantic and instance segmentation of individual trees) | Puliti et al | 2023 | https://arxiv.org/abs/2309.01279 | | Riegl - Multiple sensors | Trees datasets | None | Raw | | TUMBA | | Trees - With leaves | Tumbarumba Wet Eucalypt Terrestrial LiDAR, 2022 | Shaun et al | 2022 | https://researchdata.edu.au/tumbarumba-wet-eucalypt-lidar-2022/2766669 | | Riegl VZ-2000i Terrestrial Laser Scanner | Australian eucalyptus | None | Raw | | Weisser2024 | Weisser2024 | Trees - With leaves | Manually labeled terrestrial laser scanning point clouds of individual trees for leaf-wood separation | Weisser et al | 2024 | https://heidata.uni-heidelberg.de/dataset.xhtml?persistentId=doi:10.11588/data/UUMEDI | https://doi.org/10.11588/data/UUMEDI | RIEGL VZ-400 TLSRIEGL VZ-400 TLS | Forestry trees | None | Raw | | Vicari2018a | Vicari2018a | Trees - With leaves | Leaf and wood classification framework for terrestrial LiDAR point clouds: Simulated data validation dataset | Vicari et al | 2018 | https://zenodo.org/records/1324158 | | Simulation | Forestry trees | None | Raw | | Vicari2018b | Vicari2018b | Trees - With leaves | Leaf and wood classification framework for terrestrial LiDAR point clouds: Field data validation dataset | Vicari et al | 2018 | https://zenodo.org/records/1324156 | | Riegl VZ-400 | Forestry trees | None | Raw | | Westling2021 | Westling2021 | Trees - With leaves | Graph-based methods for analyzing orchard tree structure using noisy point cloud data | Westling et al | 2021 | https://data.mendeley.com/datasets/d6k5v2rmyx/1 | | Multiple | Agricultural trees | None | Raw | | Momotakoudjou2018 | Momotakoudjou2018 | Trees - With leaves | Using terrestrial laser scanning data to estimate large tropical trees biomass and calibrate allometric models: a comparison with traditional destructive approach | Momo takoudjou et al | 2018 | https://datadryad.org/stash/dataset/doi:10.5061/dryad.10hq7 | | Leica C10 Scanstation | tropical trees | None | Raw | | Wytham woods | Wytham woods | Trees - With leaves | Virtual forest for radiative transfer modelling: realistic stand reconstruction from terrestrial LiDAR | Calders et al | 2018 | https://bitbucket.org/treeresearch/wythamwoods3dmodel/src/master/ | | RIEGL VZ-400 | forestry trees | None | Raw |