datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
early-crop-weed_classification
Early Crop Weed Classification
This dataset contains real-world RGB images capturing early-stage crop and weed plants in agricultural field environments. The images were collected under natural outdoor conditions, providing visual data for distinguishing between different plant species during critical growth phases. The dataset contains 508 images across 4 classes: black_nightsade, cotton, tomato, velvet_leaf .Images per class:
black_nightsade: 123
cotton: 54
tomato: 201… See the full description on the dataset page: https://huggingface.co/datasets/Project-AgML/early-crop-weed_classification.Crop_Weed_classification
Crop & Weed Classification Dataset
© 2026 Rishi. All rights reserved.
This dataset is provided under the Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) license.
Usage
You are free to share and adapt this dataset under the following terms:
Attribution: You must give appropriate credit to the author.
Non-Commercial: You may not use this material for commercial purposes.
Citation & Paper
A research paper utilizing this… See the full description on the dataset page: https://huggingface.co/datasets/Rishi210904/Crop_Weed_classification.greenhouse_crop_weed_classification
Greenhouse Crop Weed Classification
A dataset for classification of crops and weeds in a greenhouse setting. The dataset contains 28,000 images across 14 classes:Images per class:
Blackbean: 2,000
Canola: 2,000
Corn: 2,000
Field Pea: 2,000
Flax: 2,000
Horseweed: 2,000
Kochia: 2,000
Lentil: 2,000
Palmer Amaranth: 2,000
Ragweed: 2,000
Redroot Pigweed: 2,000
Soybean: 2,000
Sugar beet: 2,000
Waterhemp: 2,000
This dataset is indexed on https://project-agml.github.io/ as part of the… See the full description on the dataset page: https://huggingface.co/datasets/Project-AgML/greenhouse_crop_weed_classification.weedsense
WeedSense Dataset
Multi-task temporal dataset of 16 weed species for semantic segmentation, height regression, and growth stage classification from the WeedSense paper:
WeedSense: Multi-Task Learning for Weed Segmentation, Height Estimation, and Growth Stage Classification
Toqi Tahamid Sarker, Khaled R Ahmed, Taminul Islam, Cristiana Bernardi Rankrape, Karla Gage
Southern Illinois University Carbondale, USA
ICCV 2025
Paper (arXiv) | Project Page | Code
Overview… See the full description on the dataset page: https://huggingface.co/datasets/baselab/weedsense.crop_weeds_greece
Crop Weeds Greece
A dataset for image classification of Crop Weeds Greece. The dataset contains 508 images across 4 classes: black_nightshade, cotton, tomato, velvet_leaf.Images per class:
black_nightshade: 123
cotton: 54
tomato: 201
velvet_leaf: 130
This dataset is indexed on https://project-agml.github.io/ as part of the AgML python library.
Citation
@article{ESPEJOGARCIA2020105306,
title = {Towards weeds identification assistance through transfer learning}… See the full description on the dataset page: https://huggingface.co/datasets/Project-AgML/crop_weeds_greece.maize_tomato_weed_classification
Maize Tomato Weed Classification
A dataset for classification of weeds and crops in maize and tomato fields. The dataset contains 55,828 images across 12 classes:Images per class:
atriplex: 2,459
chenopodium: 3,375
convolvulus: 2,302
cyperus: 3,215
datura: 1,271
lolium: 1,063
maize: 25,273
portulaca: 2,052
salsola: 2,409
solanum: 4,084
sorghum: 1,703
tomato: 6,622
This dataset is indexed on https://project-agml.github.io/ as part of the AgML python library.… See the full description on the dataset page: https://huggingface.co/datasets/Project-AgML/maize_tomato_weed_classification.soybean_weed_uav_brazil
Soybean Weed Uav Brazil
A dataset for image classification of Soybean Weed Uav Brazil. The dataset contains 15,336 images across 4 classes: broadleaf, grass, soil, soybean.Images per class:
broadleaf: 1,191
grass: 3,520
soil: 3,249
soybean: 7,376
This dataset is indexed on https://project-agml.github.io/ as part of the AgML python library.
Citation
dos Santos Ferreira, Alessandro; Pistori, Hemerson; Matte Freitas, Daniel; Gonçalves da Silva, Gercina (2017)… See the full description on the dataset page: https://huggingface.co/datasets/Project-AgML/soybean_weed_uav_brazil.MH_Weed16_weed_variety_classification
MH Weed16 Weed Variety Classification
A dataset for variety classification of weeds. The dataset contains 19,141 images across 16 classes:Images per class:
Asian_Pigeonwings_(Clitoria Ternatea): 678
Bilayat_(Mexicana_Argemone): 859
Choti_dudhi_(Euphorbia_hirta): 580
Digitaria_SP_(Digitaria Sanguinalis ): 1,561
Dwarf_cassia_(Chamaecrista pumila): 1,306
Gajar_gavat_(Parthenium hysterophorus): 1,543
Graceful_Sandmart_(Euphorbia hypericifolia): 262
Harali_(Cynodon_dactylon): 709… See the full description on the dataset page: https://huggingface.co/datasets/Project-AgML/MH_Weed16_weed_variety_classification.rice_field_weed_classification
Rice Field Weed Classification
A dataset for weed classification in rice fields. The dataset contains 3,632 images across 11 classes:
Images per class:
Alternanthera philoxeroide: 106
Centella asiatica: 404
Commelina benghalensis: 316
Cyperus ochraceus: 190
Fimbristylis littoralis: 168
Ipomoea aquatic: 897
Marsilea minuta: 346
Panicum repens: 294
Paspalum scrobiculatum: 345
Pteris vittata: 221
Synedrella nodiflora: 345
This dataset is indexed on… See the full description on the dataset page: https://huggingface.co/datasets/Project-AgML/rice_field_weed_classification.rangeland_weeds_australia
Rangeland Weeds Australia
A dataset for image classification of rangeland weeds. The dataset contains 17,509 images across 9 classes: chinee_apple, lantana, negative, parkinsonia, parthenium, prickly_acacia, rubber_vine, siam_weed, snake_weed.Images per class:
chinee_apple: 1,125
lantana: 1,064
negative: 9,106
parkinsonia: 1,031
parthenium: 1,022
prickly_acacia: 1,062
rubber_vine: 1,009
siam_weed: 1,074
snake_weed: 1,016
This dataset is indexed on… See the full description on the dataset page: https://huggingface.co/datasets/Project-AgML/rangeland_weeds_australia.sorghum_weed_classification
Sorghum Weed Classification
A dataset for weed classification in sorghum fields. The dataset contains 4,312 images across 3 classes: BroadLeafWeed, Grass, Sorghum.Images per class:
BroadLeafWeed: 1,441
Grass: 1,467
Sorghum: 1,404
This dataset is indexed on https://project-agml.github.io/ as part of the AgML python library.
Citation
@article{justina2024sorghumweeddataset_classification,
title={SorghumWeedDataset\_Classification and… See the full description on the dataset page: https://huggingface.co/datasets/Project-AgML/sorghum_weed_classification.weedsense
WeedSense Dataset
Multi-task temporal dataset of 16 weed species for semantic segmentation, height regression, and growth stage classification from the WeedSense paper:
WeedSense: Multi-Task Learning for Weed Segmentation, Height Estimation, and Growth Stage Classification
Toqi Tahamid Sarker, Khaled R Ahmed, Taminul Islam, Cristiana Bernardi Rankrape, Karla Gage
Southern Illinois University Carbondale, USA
ICCV 2025
Paper (arXiv) | Project Page | Code
Overview… See the full description on the dataset page: https://huggingface.co/datasets/w2836460946/weedsense.weedsense
WeedSense Dataset
Multi-task temporal dataset of 16 weed species for semantic segmentation, height regression, and growth stage classification from the WeedSense paper:
WeedSense: Multi-Task Learning for Weed Segmentation, Height Estimation, and Growth Stage Classification
Toqi Tahamid Sarker, Khaled R Ahmed, Taminul Islam, Cristiana Bernardi Rankrape, Karla Gage
Southern Illinois University Carbondale, USA
ICCV 2025
Paper (arXiv) | Project Page | Code
Overview… See the full description on the dataset page: https://huggingface.co/datasets/Shuoyang111/weedsense.