Cloud-ImVN 1.0

This paper introduces a new dataset for solving the ground-based cloud images classification task. We name it ‘Cloud-ImVN 1.0’ which is an extension of SWIMCAT database. This dataset contains 6 categories of clouds images which consists of 2,100 color images (150 × 150 pixels). Several task can be applied on this dataset including classification, clustering and segmentation in both supervised and unsupervised learning context.本文提出一款面向地面云图像分类任务的新型数据集，将其命名为“Cloud-ImVN 1.0”，该数据集系SWIMCAT数据库的扩展版本。本数据集涵盖6类云图像样本，总计2100张分辨率为150×150像素的彩色图像。该数据集可适配多类计算机视觉任务，涵盖监督学习与无监督学习场景下的分类、聚类及图像分割任务。