CropAndWeedAndLeaf
收藏资源简介:
CropAndWeedAndLeaf是由奥地利理工学院和维也纳应用技术大学联合创建的植物叶片分割数据集,旨在解决精准农业中叶片级分析的不足。该数据集包含345个样本,覆盖23种植物物种,每个样本均经过半自动化标注流程生成高质量的叶片级掩码。数据来源于真实农田环境中的CropAndWeed图像,通过模型辅助预标注与专家人工修正相结合的方式创建。该数据集主要用于评估深度学习模型在跨物种叶片分割任务中的泛化能力,为精准农业中的植物健康监测、生长阶段评估等应用提供数据支持。
CropAndWeedAndLeaf is a plant leaf segmentation dataset jointly created by the Austrian Institute of Technology and the University of Applied Sciences Vienna, aiming to address the gaps in leaf-level analysis in precision agriculture. This dataset includes 345 samples spanning 23 plant species, with each sample paired with high-quality leaf-level masks generated via a semi-automated annotation workflow. The data is sourced from CropAndWeed images captured in real-world farmland environments, and was compiled through a combination of model-assisted pre-annotation and expert manual correction. This dataset is primarily utilized to evaluate the generalization ability of deep learning models in cross-species leaf segmentation tasks, offering data support for applications such as plant health monitoring and growth stage assessment in precision agriculture.
根据提供的README文件内容,目前该数据集详情页面仅包含以下信息:
数据集概述
- 名称:该页面涉及ReLeaf框架及其对应的CropAndWeedAndLeaf数据集。
- 当前状态:数据集详情页面标注为“Coming soon...”,表示数据集尚未正式发布,相关内容即将上线。
许可协议
- 该工作采用 Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License(CC BY-NC-SA 4.0)国际许可协议进行授权。
引用信息
若在研究中使用了ReLeaf框架或CropAndWeedAndLeaf数据集,请引用以下文献:
-
论文标题:ReLeaf: Benchmarking Leaf Segmentation across Domains and Species
-
会议:Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops
-
发表时间:2026年6月
-
作者:Robert Martinko, Daniel Steininger, Julia Simon, Andreas Trondl, Matthias Blaickner
-
BibTeX引用格式:
@InProceedings{Martinko_2026_CVPR, author = {Martinko, Robert and Steininger, Daniel and Simon, Julia and Trondl, Andreas and Blaickner, Matthias}, title = {ReLeaf: Benchmarking Leaf Segmentation across Domains and Species}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2026} }
注意事项
- 目前页面内容极简,未提供数据集的具体描述、样本数量、标注格式、任务类型等详细信息,后续发布后请以实际页面内容为准。




