自定义数据集(论文中未提及具体名称)
收藏资源简介:
该自定义数据集由康奈尔大学和堪萨斯州立大学的研究人员开发,包含在复杂果园环境下拍摄的857张真实场图像。图像采集于华盛顿州普罗瑟的一个商业果园,这些图像的特点是未成熟的绿色果实与绿色树冠背景颜色相似,导致检测中存在显著的遮挡和视觉混淆问题。数据集用于评估RF-DETR和YOLOv12模型在单类和多类绿色果实检测任务中的性能,这些任务对于精确农业中的自动化监测和采摘具有重要意义。
This custom dataset was developed by researchers from Cornell University and Kansas State University, consisting of 857 real-field images captured in complex orchard environments. The images were collected in a commercial orchard in Prosser, Washington State. A notable feature of these images is that unripe green fruits exhibit similar color to the green canopy background, which causes severe occlusion and visual confusion challenges during detection. This dataset is utilized to assess the performance of RF-DETR and YOLOv12 models on single-class and multi-class green fruit detection tasks, which are critical for automated monitoring and harvesting in precision agriculture.

- 1RF-DETR Object Detection vs YOLOv12 : A Study of Transformer-based and CNN-based Architectures for Single-Class and Multi-Class Greenfruit Detection in Complex Orchard Environments Under Label Ambiguity康奈尔大学生物与环境工程系, 堪萨斯州立大学生物与农业工程系 · 2025年



