{"success":true,"data":{"author":"Lalit Pun Magar, Matthew Thompson and Eugene P. Law (The Ohio State University)","bucketAccessType":"public","bucketName":"punlalit19/Waterhemp_GreenHouse_Images","bucketPrefix":"","category":"Agriculture & Environment","cloudProvider":"huggingface","createdAt":"2026-11-21T22:15:35.976Z","description":"Top-down photographs of potted waterhemp (Amaranthus tuberculatus), a herbicide-resistant weed of Ohio corn and soybean, grown in the Kottman Hall greenhouse at The Ohio State University and treated with 2,4-D (WSSA Group 4), glyphosate (Group 9), glufosinate (Group 10) or fomesafen (Group 14), with untreated controls. Collected by Lalit Pun Magar in the Weed Ecology and Management Lab to support computer vision rating of herbicide injury, which is otherwise scored visually by eye, and used at FloraPalooza 2026 to evaluate BioCLIP, BioCLIP 2 and DINOv3 image embeddings. There are 622 RGB JPEG images at 4032 x 3040 taken from 30 cm and 50 cm above the pot, sorted into folders by symptom class: healthy_untreated, chlorosis_only, necrosis_only, chlorosis_necrosis, epinasty and foliar_distortion. File names encode date, trial, camera height and pot id, so the same plant can be followed over time. About 4.3 GB; load as an image folder with datasets or torchvision. Useful for weed phenotyping, herbicide injury classification and plant stress detection. Dedicated to the public domain under CC0 1.0 by the curators; please cite the dataset.","displayTitle":"Herbicide-Damaged Waterhemp (Amaranthus tuberculatus) Greenhouse Images: Healthy and 2,4-D, Glyphosate, Glufosinate and Fomesafen Injury Symptoms (Ohio State University)","huggingFaceUrl":"https://huggingface.co/datasets/punlalit19/Waterhemp_GreenHouse_Images","id":"9ce55192-53da-58c1-8548-83e49b9b6a9b","license":"CC0-1.0","licenseEvidenceUrl":"https://huggingface.co/datasets/punlalit19/Waterhemp_GreenHouse_Images","objectCountEstimate":624,"probeStatus":"ok","referenceUrl":"https://huggingface.co/datasets/punlalit19/Waterhemp_GreenHouse_Images","region":"us","schemaCard":{"accessHints":[{"language":"python-datasets","snippet":"from datasets import load_dataset\nds = load_dataset(\"punlalit19/Waterhemp_GreenHouse_Images\")"},{"language":"python-hub","snippet":"from huggingface_hub import snapshot_download\nsnapshot_download(repo_id=\"punlalit19/Waterhemp_GreenHouse_Images\", repo_type=\"dataset\")"}],"caveats":["HuggingFace metadata: author=punlalit19, license:cc0-1.0, size_categories:n<1K, format:imagefolder, downloads=678"],"format":"mixed","generatedAt":"2026-11-21T22:31:48.655Z","layout":{"example":"chlorosis_necrosis/2026-05-11_IGPTRIAL1H30_110_1_rgb_32effab8-e969-4b6b-ae19-f9168d37996a.jpg","pattern":"{repoId}/{filename}"},"mountPath":"/mnt/archive","sampleHeaders":[{"bytes":"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","encoding":"hex","key":"chlorosis_necrosis/2026-05-11_IGPTRIAL1H30_110_1_rgb_32effab8-e969-4b6b-ae19-f9168d37996a.jpg","truncatedAt":512}],"sampleObjects":["chlorosis_necrosis/2026-05-11_IGPTRIAL1H30_110_1_rgb_32effab8-e969-4b6b-ae19-f9168d37996a.jpg","chlorosis_necrosis/2026-05-11_IGPTRIAL1H30_209_1_rgb_9a91209e-769c-4cbe-865a-e8a19b0d7e29.jpg","chlorosis_necrosis/2026-05-11_IGPTRIAL1H30_304_1_rgb_1a6f0420-ec74-4b4f-b0b9-2c0c7f0ef767.jpg","chlorosis_necrosis/2026-05-11_IGPTRIAL1H30_309_1_rgb_76728578-03d4-413c-a5e6-44b924ebca0d.jpg","chlorosis_necrosis/2026-05-11_IGPTRIAL1H30_507_1_rgb_bd2c96b1-46e6-44b2-849d-88a3db335525.jpg"],"sidecars":[{"bytes":"---\nlicense: cc0-1.0\nlanguage:\n- en\npretty_name: Herbicide-Damaged Waterhemp Greenhouse Images\ntask_categories:\n- image-classification\ntags:\n- biology\n- image\n- plants\n- CV\n- waterhemp\n- herbicide_damage\n- chlorosis\n- necrosis\n- epinasty\nsize_categories:\n- n<1K\ndescription: Greenhouse images of waterhemp (Amaranthus tuberculatus) plants, untreated and treated with 2,4-D, glyphosate, glufosinate, and fomesafen, labeled by visual injury symptom and captured at 30 cm and 50 cm from 1 day before to 21 days after herbicide application.\n---\n\n# Dataset Card for Herbicide-Damaged Waterhemp Greenhouse Images\n\nImages of healthy and herbicide-injured waterhemp plants, collected to support automated recognition of herbicide injury symptoms. This dataset was used during [FloraPalooza 2026](https://github.com/Imageomics/FloraPalooza-2026).\n\n## Dataset Details\n\n### Dataset Description\n\n- **Curated by:** Lalit Pun Magar, Matthew Thompson, Eugene P. Law\n- **Language(s) (NLP):** English\n- **Repository:** https://github.com/Imageomics/AgDamage\n\nThe dataset consists of images of waterhemp (*Amaranthus tuberculatus*) plants grown under greenhouse conditions. Waterhemp is one of the major weeds in Ohio corn and soybean production and has evolved resistance to herbicides from several sites of action. The dataset includes images of both untreated (control) plants and plants treated with multiple herbicides. Untreated plants are healthy, whereas treated plants show injury symptoms. Images are categorized into six classes: chlorosis, necrosis, epinasty, foliar distortion, chlorosis + necrosis, and healthy. Images were captured from two heights (30 cm and 50 cm above the pots) at 1 day before herbicide application and at 3, 5, 7, 10, 14, and 21 days after application (DAA).\n\n### Supported Tasks and Leaderboards\n\nImage classification in two stages: (1) binary classification of healthy vs. damaged plants, and (2) classification of damaged plants into injury symptom classes. During FloraPalooza 2026, the dataset was used to evaluate foundation-model image embeddings (BioCLIP, BioCLIP 2, BioCLIP 2.5, and DINOv3 ViT-7B/16) for both stages.\n\n## Dataset Structure\n\n```\n/Waterhemp_GreenHouse_Images/\n    chlorosis_necrosis/\n        <file_name 1>.jpg\n        ...\n    chlorosis_only/\n        <file_name 1>.jpg\n        ...\n    epinasty/\n        <file_name 1>.jpg\n        ...\n    foliar_distortion/\n        <file_name 1>.jpg\n        ...\n    healthy_untreated/\n        <file_name 1>.jpg\n        ...\n    necrosis_only/\n        <file_name 1>.jpg\n        ...\n```\n\n### Data Instances\n\nAll images are RGB JPEG files (4032 × 3040 pixels), each stored in a folder named for its symptom class. Each image shows a single potted waterhemp plant photographed from above. There is no separate metadata file; other information is encoded in the file name (see Data Fields).\n\n### Data Fields\n\nFile names follow this pattern:\n\n```\n50 cm: <date>_<trial>_<potID>_1_rgb_<number>.jpg\n30 cm: <date>_<trial>H30_<potID>_1_rgb_<number>.jpg\n```\n\n- `date`: date the image was collected (YYYY-MM-DD).\n- `trial`: name of the trial.\n- `H30`: camera height tag, appended directly to the trial name. Images taken 30 cm above the pot carry `H30`; images taken at 50 cm have no height tag.\n- `potID`: identifier of the pot. Use this to group images of the same plant across dates and heights.\n- `1`: fixed value; not used for analysis.\n- `rgb`: image type (RGB).\n- `number`: image number; not used for analysis.\n\n### Data Splits\n\nNo predefined train/test split is provided. Because the same plants are imaged at multiple time points and heights, we recommend splitting by plant (pot) rather than by image.\n\n## Dataset Creation\n\n### Curation Rationale\n\nHerbicide injury in weeds and crops is assessed visually in weed management research. Manual visual rating is the standard procedure but is time-consuming and subjective. This dataset was collected to support the development of computer vision methods for weed phenotyping, specifically the recognition of herbicide injury symptoms on waterhemp.\n\n### Source Data\n\n#### Data Col","encoding":"utf8","key":"README.md","name":"README.md","size":8163,"truncated":true}],"sourceObjectKey":"chlorosis_necrosis/2026-05-11_IGPTRIAL1H30_110_1_rgb_32effab8-e969-4b6b-ae19-f9168d37996a.jpg","starterCell":"from huggingface_hub import hf_hub_download\npath = hf_hub_download(repo_id=\"punlalit19/Waterhemp_GreenHouse_Images\", filename=\"chlorosis_necrosis/2026-05-11_IGPTRIAL1H30_110_1_rgb_32effab8-e969-4b6b-ae19-f9168d37996a.jpg\", repo_type=\"dataset\")\nprint(path)","starterSamplePath":"chlorosis_necrosis/2026-05-11_IGPTRIAL1H30_110_1_rgb_32effab8-e969-4b6b-ae19-f9168d37996a.jpg","totalObjects":624},"summary":"Top-down photographs of potted waterhemp (Amaranthus tuberculatus), a herbicide-resistant weed of Ohio corn and soybean, grown in the Kottman Hall greenhouse at The Ohio State…","title":"Herbicide-Damaged Waterhemp (Amaranthus tuberculatus) Greenhouse Images: Healthy and 2,4-D, Glyphosate, Glufosinate and Fomesafen Injury Symptoms (Ohio State University)","updatedAt":"2026-11-22T00:55:33.199Z","visibility":"public","websiteUrl":"https://github.com/Imageomics/AgDamage"}}