SUES-200
收藏资源简介:
SUES-200是由上海工程技术大学电子电气工程学院创建的多高度多场景跨视角图像基准数据集,包含24120张无人机在四个不同高度拍摄的图像及其对应的卫星图像。该数据集旨在解决现有公共数据集在评估模型适应复杂多变场景能力方面的不足。SUES-200涵盖了公园、学校、湖泊和公共建筑等多种场景,每种场景的数据均在150m、200m、250m和300m四个高度收集。数据集的应用领域包括无人机定位和导航,旨在通过学习高度判别性的特征,提高模型在不同视角和高度下的匹配准确性。
SUES-200 is a multi-height, multi-scenario cross-view image benchmark dataset created by the School of Electronic and Electrical Engineering, Shanghai University of Engineering Science. It contains 24,120 images captured by UAVs at four different altitudes and their corresponding satellite images. This dataset is designed to address the shortcomings of existing public datasets in evaluating the ability of models to adapt to complex and variable scenarios. SUES-200 covers various scenarios such as parks, schools, lakes, and public buildings, and data for each scenario is collected at four altitudes: 150m, 200m, 250m, and 300m. The application fields of this dataset include UAV positioning and navigation, aiming to improve the matching accuracy of models across different viewpoints and altitudes by learning highly discriminative features.




