{"success":true,"data":{"author":"André Silva Aguiar and Sandro Magalhães (INESC TEC); Hugging Face mirror by Project-AgML","bucketAccessType":"public","bucketName":"Project-AgML/vineyard_grape_trunk_detection","bucketPrefix":"","category":"Agriculture & Environment","cloudProvider":"huggingface","createdAt":"2026-11-22T01:09:12.547Z","description":"Ground-level RGB photographs of vineyards in the Douro wine region of Portugal, taken with several cameras under varied illumination, with bounding boxes for three classes: vine trunks, small grape bunches just after bloom, and medium-sized grape bunches. The release accompanies work on detecting grape clusters with quantized deep-learning detectors on low-power hardware for agricultural robots. The raw configuration holds 2,846 unaugmented images with 13,218 boxes; the augmented configuration holds 18,515 copies (rotation, flipping, hue and saturation shifts, noise, scaling and translation) with 85,722 boxes, each labeled with its augmentation and capture set. Thermal images in the original Zenodo record are not included. Stored as Parquet shards with the image bytes and annotations in columns; load with datasets.load_dataset(\"Project-AgML/vineyard_grape_trunk_detection\") or through the AgML Python library (Project-AgML, UC Davis), which reformatted the original release. License: CC BY 4.0, as stated by the original publisher; cite André Silva Aguiar and Sandro Magalhães et al. and the DOI 10.5281/zenodo.5139598.","displayTitle":"Grape Bunch and Vine Trunk Object Detection Images: 2,846 RGB Vineyard Photos from the Douro Region, Portugal (INESC TEC, AgML Mirror)","doi":"10.5281/zenodo.5139598","huggingFaceUrl":"https://huggingface.co/datasets/Project-AgML/vineyard_grape_trunk_detection","id":"ef672dd5-bc79-5e97-bfd2-ac2eba379711","license":"CC-BY-4.0","licenseEvidenceUrl":"https://zenodo.org/records/5139598","objectCountEstimate":12,"probeStatus":"ok","referenceUrl":"https://huggingface.co/datasets/Project-AgML/vineyard_grape_trunk_detection","region":"us","schemaCard":{"accessHints":[{"language":"python-datasets","snippet":"from datasets import load_dataset\nds = load_dataset(\"Project-AgML/vineyard_grape_trunk_detection\")"},{"language":"python-hub","snippet":"from huggingface_hub import snapshot_download\nsnapshot_download(repo_id=\"Project-AgML/vineyard_grape_trunk_detection\", repo_type=\"dataset\")"}],"caveats":["HuggingFace metadata: author=Project-AgML, license:cc-by-4.0, size_categories:10K<n<100K, format:parquet, downloads=120"],"format":"mixed","generatedAt":"2026-11-22T01:10:00.860Z","layout":{"example":"augmented/train-00000-of-00006.parquet","pattern":"{repoId}/{filename}"},"mountPath":"/mnt/archive","sampleHeaders":[{"bytes":"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","encoding":"hex","key":"augmented/train-00000-of-00006.parquet","truncatedAt":512}],"sampleObjects":["augmented/train-00000-of-00006.parquet","augmented/train-00001-of-00006.parquet","augmented/train-00002-of-00006.parquet","augmented/train-00003-of-00006.parquet","augmented/train-00004-of-00006.parquet"],"sidecars":[{"bytes":"---\ndataset_info:\n- config_name: augmented\n  features:\n  - name: image\n    dtype: image\n  - name: objects\n    struct:\n    - name: bbox\n      list:\n        list: float64\n    - name: categories\n      list:\n        class_label:\n          names:\n            '0': trunk\n            '1': tiny_grape_bunch\n            '2': medium_grape_bunch\n  - name: capture_set\n    dtype: string\n  - name: augmentation\n    dtype: string\n  splits:\n  - name: train\n    num_bytes: 2696503818\n    num_examples: 18515\n  download_size: 2694490286\n  dataset_size: 2696503818\n- config_name: raw\n  features:\n  - name: image\n    dtype: image\n  - name: objects\n    struct:\n    - name: bbox\n      list:\n        list: float64\n    - name: categories\n      list:\n        class_label:\n          names:\n            '0': trunk\n            '1': tiny_grape_bunch\n            '2': medium_grape_bunch\n  - name: capture_set\n    dtype: string\n  splits:\n  - name: train\n    num_bytes: 870691241\n    num_examples: 2846\n  download_size: 870301684\n  dataset_size: 870691241\nconfigs:\n- config_name: augmented\n  data_files:\n  - split: train\n    path: augmented/train-*\n- config_name: raw\n  data_files:\n  - split: train\n    path: raw/train-*\n  default: true\nlicense: cc-by-4.0\ntask_categories:\n- object-detection\nsize_categories:\n- 1K<n<10K\n---\n# Vineyard Grape Trunk Detection\n\nImages of vineyards in the Douro region of Portugal, taken by several ground-level cameras in varied illumination, with bounding boxes for vine trunks and for grape bunches at two growth stages (tiny, after bloom, and medium). The dataset accompanies a study on detecting grape bunches with quantized deep-learning models on low-power hardware; the current Zenodo version also adds trunk annotations. The raw config holds the 2,846 unaugmented RGB images; the augmented config holds 18,515 augmented copies (rotation, flipping, hue/saturation shifts, noise, scaling and translation), with the augmentation named per image. Each image records its capture set, and the record's thermal images are left out. The dataset contains raw and augmented versions.\nThe raw split contains 2,846 images with 13,218 bounding box annotations across 3 categories.  \n\nThe augmented split contains 18,515 images with 85,722 bounding box annotations across 3 categories.  \n\n\nThis dataset is indexed on https://project-agml.github.io/ as part of the AgML python library.\n\n## Citation\n\n```bibtex\n@article{aguiar2021grape,\n  title={Grape Bunch Detection at Different Growth Stages Using Deep Learning Quantized Models},\n  author={Aguiar, André Silva and Magalhães, Sandro Augusto and dos Santos, Filipe Neves and Castro, Luis and Pinho, Tatiana and Valente, João and Martins, Rui and Boaventura-Cunha, José},\n  journal={Agronomy},\n  volume={11},\n  pages={1890},\n  year={2021},\n  publisher={MDPI}\n}\n```\n\n\nThe dataset itself can be cited as:\n\nAguiar, A. S., &amp; Magalhães, S. (2021). *Grape bunch and vine trunk dataset for Deep Learning object detection.* [Dataset]. Zenodo. https://doi.org/10.5281/ZENODO.5139598\n\n\n*This dataset was reformatted from its original format to match HuggingFace standards.*\n","encoding":"utf8","key":"README.md","name":"README.md","size":3115,"truncated":false}],"sourceObjectKey":"augmented/train-00000-of-00006.parquet","starterCell":"from datasets import load_dataset\nds = load_dataset(\"Project-AgML/vineyard_grape_trunk_detection\")\nprint(ds)","starterSamplePath":"augmented/train-00000-of-00006.parquet","totalObjects":12},"summary":"Ground-level RGB photographs of vineyards in the Douro wine region of Portugal, taken with several cameras under varied illumination, with bounding boxes for three classes: vine…","title":"Grape Bunch and Vine Trunk Object Detection Images: 2,846 RGB Vineyard Photos from the Douro Region, Portugal (INESC TEC, AgML Mirror)","updatedAt":"2026-11-22T04:00:36.574Z","visibility":"public","websiteUrl":"https://project-agml.github.io/"}}