GFM-Bench/SegMunich

SegMunich是一个基于Sentinel-2卫星图像的分割任务数据集，包含慕尼黑城市景观在三年时间跨度内的光谱图像。该数据集共有10个Sentinel-2波段，图像分辨率为128x128像素，空间分辨率为10米，包含13个类别。数据集分为训练集、验证集和测试集，分别包含3000、403和403个样本。数据集特征包括Sentinel-2图像、分割标签、各波段的中心波长以及图像的空间分辨率。

SegMunich is a segmentation task dataset based on Sentinel-2 satellite imagery, containing spectral images of Munichs urban landscape over a span of three years. The dataset consists of 10 Sentinel-2 bands, with an image resolution of 128x128 pixels and a spatial resolution of 10 meters, including 13 classes. The dataset is split into training, validation, and test sets with 3,000, 403, and 403 samples respectively. The dataset features include the Sentinel-2 image, segmentation labels, the central wavelength of each band, and the spatial resolution of the images.

搜集汇总

背景与挑战

背景概述

SegMunich是一个基于Sentinel-2卫星图像的语义分割数据集，包含慕尼黑城市景观三年的光谱影像，具有10个波段、128x128像素分辨率和10米空间分辨率，适用于遥感图像分析任务。