Voxel51/GrowliFlower
收藏资源简介:
GrowliFlower是一个地理参考的、基于图像的无人机时间序列数据集,涵盖德国科隆附近两个监测的花椰菜田(田1,2020年,0.39公顷;田2,2021年,0.60公顷),使用47.4 MP索尼A7rIII RGB相机和MicaSense 5通道多光谱相机拍摄。此FiftyOne数据集将源发布的三个任务特定子集整合为一个分组数据集:独立图像补丁上的像素级植物/叶子/茎实例分割、具有真实测量性状(高度、直径、BBCH生长阶段、收获状态)的表型植物生长时间序列,以及具有配对前/后脱叶捕获的脱叶植物时间序列,以暴露花椰菜头部。
GrowliFlower is a georeferenced, image-based drone time-series dataset covering two monitored cauliflower fields near Cologne, Germany (Field 1, 2020, 0.39 hectares; Field 2, 2021, 0.60 hectares), captured using a 47.4 MP Sony A7rIII RGB camera and a MicaSense 5-channel multispectral camera. This FiftyOne dataset integrates three source-released task-specific subsets into a grouped dataset: pixel-level plant/leaf/stem instance segmentation on independent image patches, phenotypic plant growth time series with ground-truth measured traits including height, diameter, BBCH growth stage, and harvest status, as well as defoliated plant time series with paired pre- and post-defoliation captures to expose cauliflower heads.




