GF2Hyper dataset for training and test

GF2Hyper dataset involves Panchromatic/Multispectral imager carried on Gaofen-1 and Hyperion sensor on EO-1 satellite. Gaofen-1 is a Chinese satellite employed two imagers, which can obtain 8 m resolution multispectral images with four bands and a 2 m resolution PAN image that covered the full spectral ranges of corresponding multispectral images, from 450 to 900 nm. Images obtained by Hyperion are with a spatial resolution of 30 meters and obtain 242 bands from 357 to 2567 nm. In this dataset, we only select 148 useful bands. GF2Hyper dataset is used to verify the one situation of generalized spectral super-resolution, i.e., PansSR. PansSR is to achieve the joint enhancement of spatial resolution and spectral resolution using auxiliary higher-resolution spectral channels, usually, panchromatic images.

The file named "GF2Hyper_train.h5" contains 1152 training samples where the high-resolution PAN images are with size of 128×128×1 while the size of multispectral images is 32×32×4. Moreover, we also give some test images in the "GF2Hyper_test.mat". Details about generalized spectral super-resolution can be found in the following paper.

Paper: J. He, Q. Yuan, J. Li, and L. Zhang, "PoNet: A universal physical optimization-based spectral super-resolution network for arbitrary multispectral images," Information Fusion, vol. 80, pp. 205–225, 2022.

More information about the author can be found at https://jianghe96.github.io/

If this dataset is helpful please cite as:

@article{He2022PoNet,
title={PoNet: A universal physical optimization-based spectral super-resolution network for arbitrary multispectral images},
author={He, Jiang and Yuan, Qiangqiang and Li, Jie and Zhang, Liangpei},
journal={Information Fusion},
volume={80},
pages={205--225},
year={2022},
}

GF2Hyper数据集涉及高分一号（Gaofen-1）搭载的全色/多光谱成像仪，以及EO-1卫星上的Hyperion传感器。高分一号是中国卫星，搭载两台成像仪，可获取4波段、分辨率为8米的多光谱图像，以及分辨率2米的全色图像，该全色图像覆盖对应多光谱图像的全光谱范围（450~900 nm）。Hyperion传感器获取的图像空间分辨率为30米，涵盖357~2567 nm的242个光谱波段，本数据集仅选取其中148个有效波段。
GF2Hyper数据集用于验证广义光谱超分辨率（generalized spectral super-resolution）的典型场景——全色驱动光谱超分辨率（PansSR）。PansSR指借助辅助的高分辨率光谱通道（通常为全色图像），实现空间分辨率与光谱分辨率的联合增强。
名为"GF2Hyper_train.h5"的文件包含1152个训练样本，其中高分辨率全色图像尺寸为128×128×1，多光谱图像尺寸为32×32×4。此外，"GF2Hyper_test.mat"中提供了部分测试图像。
关于广义光谱超分辨率的详细内容可参阅以下论文：
论文：何姜、袁强强、李杰、张亮培，《PoNet: 一种面向任意多光谱图像的通用物理优化型光谱超分辨率网络》，载于《信息融合（Information Fusion）》，第80卷，第205–225页，2022年。
作者更多研究信息可访问：https://jianghe96.github.io/
若该数据集对您的研究有所帮助，请引用如下文献：
@article{He2022PoNet,
title={PoNet: A universal physical optimization-based spectral super-resolution network for arbitrary multispectral images},
author={He, Jiang and Yuan, Qiangqiang and Li, Jie and Zhang, Liangpei},
journal={Information Fusion},
volume={80},
pages={205--225},
year={2022},
}