Sugar Beets 2016

该数据集包含了位于德国波恩附近的一块甜菜田在整个两月农业生产周期内的记录。为其中的12,196张图像提供了地面真实语义标签，并通过在语义标签周围绘制轮廓生成了实例地面真实标签。该数据集共包含123,062个样本，适用于对象检测、语义分割和实例分割等任务。

This dataset comprises records acquired over a two-month agricultural production cycle from a sugar beet field near Bonn, Germany. Semantic ground truth labels are provided for 12,196 of its images, and instance ground truth labels are generated by drawing contours around the semantic labels. This dataset contains a total of 123,062 samples, and is applicable to tasks including object detection, semantic segmentation, and instance segmentation.

搜集汇总

背景与挑战

背景概述

Sugar Beets 2016数据集是一个大规模农业机器人数据集，主要用于植物分类、定位和地图构建。数据集包含多光谱相机、RGB-D传感器、激光雷达和GPS等多种传感器采集的数据，覆盖了糖甜菜从出苗到无法机械进入的生长阶段，总数据量约为5 TB。