JMuBEN2

Datasets were taken from Arabica coffee plantation using a camera and with the help of a plant pathologist. The images were then cropped to focus on the region of interest . Image augmentation was done with the aim of increasing the dataset size and preventing over-fitting problems during model training and validation. The images are in different folders containing annotated images of healthy and Miner

本数据集采集自阿拉比卡咖啡（Arabica coffee）种植园，数据采集工作借助相机完成，并得到了植物病理学家的协助。随后对所有图像进行裁剪，以聚焦于感兴趣区域。为扩充数据集规模、避免模型训练与验证过程中出现过拟合问题，研究人员对图像实施了数据增强处理。图像被分类存储于不同文件夹中，内含健康样本与Miner（原文未完整表述）的标注图像。