多尺度空谱特征协同的高光谱图像全色锐化网络

doi:10.3785/j.issn.1008-973X.2026.12.004

浙江大学学报(工学版)

2026, Vol. 60

Issue (12): 2581-2591 DOI: 10.3785/j.issn.1008-973X.2026.12.004

计算机技术

多尺度空谱特征协同的高光谱图像全色锐化网络

张剑鑫1(

),李小军1,2,3,*(

),徐欣钰4,李轶鲲1,2,3

1. 兰州交通大学测绘与地理信息学院，甘肃兰州 730070
2. 地理国情监测技术应用国家地方联合工程研究中心，甘肃兰州 730070
3. 甘肃省测绘科学与技术重点实验室，甘肃兰州 730070
4. 中国科学院西北生态环境资源研究院国家冰川冻土沙漠科学数据中心，甘肃兰州 730000

Multi-scale spatial-spectral feature collaborative network for hyperspectral image pansharpening

Jianxin ZHANG1(

),Xiaojun LI1,2,3,*(

),Xinyu XU4,Yikun LI1,2,3

1. Faculty of Geomatics, Lanzhou Jiaotong University, Lanzhou 730070, China
2. National-Local Joint Engineering Research Center of Technologies and Applications for National Geographic State Monitoring, Lanzhou 730070, China
3. Key Laboratory of Science and Technology in Surveying and Mapping, Gansu Province, Lanzhou 730070, China
4. National Cryosphere Desert Data Center, Northwest Institute of Eco-Environment and Resources, Chinese Academy of Sciences, Lanzhou 730000, China

全文: PDF(7825 KB) HTML

摘要：

为了提高高光谱图像的空间表达能力并保持光谱一致性，构建多尺度空谱特征协同的高光谱图像全色锐化方法. 采用双向金字塔多尺度对齐结构，通过对高光谱图像逐级上采样和对全色图像逐级下采样生成多尺度输入对，实现输入端的空谱一致性对齐. 构建空谱特征挖掘增强模块，通过多尺度交叉注意增强空谱特征提取能力，结合空间-光谱特征挖掘与滑动窗口融合实现异构特征的深层挖掘与协同融合. 引入光谱注意力机制增强通道一致性，形成“提取-挖掘-融合-增强”的空谱协同特征流. 设计渐进式多尺度空谱重建模块，通过逐级重建与特征补偿整合多尺度信息，实现层次化的空谱一致性重建. 实验在多个高光谱数据集上进行验证，结果表明，该方法在空间细节恢复和光谱保真度方面均优于多种现有的全色锐化技术. 利用提出的模型，能够有效提升高光谱图像锐化融合的质量，在不同场景下均表现出更优的空间细节和光谱保真度.

关键词： 全色锐化; 高光谱图像; 多尺度特征; 空谱融合; 特征提取

Abstract:

A multi-scale spatial-spectral feature collaborative hyperspectral image pansharpening method was constructed in order to enhance the spatial representation capability of hyperspectral images while preserving spectral consistency. A bidirectional pyramid multi-scale alignment structure was adopted to progressively upsample the hyperspectral image and downsample the panchromatic image, generating multi-scale input pairs and achieving spatial-spectral consistency alignment at the input end. A spatial-spectral feature mining and enhancement module was constructed, and multi-scale cross-attention was employed to strengthen spatial-spectral feature extraction. Spatial-spectral feature mining together with sliding-window fusion were combined to realize deep mining and collaborative fusion of heterogeneous features. A spectral attention mechanism was introduced to enhance channel consistency, forming an extraction-mining-fusion-enhancement spatial-spectral collaborative feature flow. A progressive multi-scale spatial-spectral reconstruction module was designed to integrate multi-scale information through stage-wise reconstruction and feature compensation, achieving hierarchical spatial-spectral consistency reconstruction. Experiments were conducted on multiple hyperspectral datasets. Results demonstrate that the proposed method outperforms several existing pansharpening techniques in both spatial detail recovery and spectral fidelity. The quality of hyperspectral image pansharpening can be effectively improved by the proposed method, which exhibits superior spatial detail and spectral fidelity across different scenes.

Key words: pansharpening hyperspectral image multi-scale feature spatial-spectral fusion feature extraction

收稿日期: 2025-12-07 出版日期: 2026-10-21

CLC:

TP 751

基金资助: 国家重点研发计划资助项目（2022YFB3903604）；国家自然科学基金资助项目（41861055）；中国博士后基金资助项目（2019M653795）.

通讯作者: 李小军 E-mail: 11230866@stu.lzjtu.edu.cn;xjli@mail.lzjtu.cn

作者简介: 张剑鑫（2002—），男，硕士生，从事遥感图像融合研究. orcid.org/0009-0003-2779-6184.E-mail：11230866@stu.lzjtu.edu.cn

	服务
	把本文推荐给朋友
	加入引用管理器
	E-mail Alert
	作者相关文章
	张剑鑫
	李小军
	徐欣钰
	李轶鲲

引用本文:

张剑鑫,李小军,徐欣钰,李轶鲲. 多尺度空谱特征协同的高光谱图像全色锐化网络[J]. 浙江大学学报(工学版), 2026, 60(12): 2581-2591.

Jianxin ZHANG,Xiaojun LI,Xinyu XU,Yikun LI. Multi-scale spatial-spectral feature collaborative network for hyperspectral image pansharpening. Journal of ZheJiang University (Engineering Science), 2026, 60(12): 2581-2591.

链接本文:

https://www.zjujournals.com/eng/CN/10.3785/j.issn.1008-973X.2026.12.004 或 https://www.zjujournals.com/eng/CN/Y2026/V60/I12/2581

图 1 多尺度空谱特征协同的高光谱图像全色锐化网络框架

图 2 多尺度交叉注意与残差增强块结构

图 3 多尺度交叉注意力的结构

图 4 空间-光谱特征挖掘块的结构

图 5 滑动窗口融合块的结构

图 6 不同方法在Pavia Center数据集上的对比结果

表 1 不同方法在Pavia Center数据集和Houston数据集上的定量评价结果

图 7 不同方法在Houston数据集上的对比结果

图 8 不同数据集随机位置上不同方法锐化结果与参考图像光谱曲线差异

表 2 不同尺度组合的定量评价结果

表 3 不同架构的定量评价结果

1	陈江浩, 杨军结合深度可分离卷积的多源遥感融合影像目标检测[J]. 浙江大学学报(工学版), 2025, 59 (12): 2545- 2555 Chen Jianghao, Yang Jun Object detection for multi-source remote sensing fused images based on depthwise separable convolution[J]. Journal of Zhejiang University (Engineering Science), 2025, 59 (12): 2545- 2555
2	Rajabi R, Zehtabian A, Singh K D, et al Editorial: hyperspectral imaging in environmental monitoring and analysis[J]. Frontiers in Environmental Science, 2024, 11: 1353447 doi: 10.3389/fenvs.2023.1353447
3	李明婉, 房胜, 李哲非对称结构的高光谱与激光雷达图像分类模型[J]. 浙江大学学报(工学版), 2025, 59 (12): 2539- 2544, 565 Li Mingwan, Fang Sheng, Li Zhe Asymmetric structure based hyperspectral and LiDAR image classification model[J]. Journal of Zhejiang University (Engineering Science), 2025, 59 (12): 2539- 2544, 565
4	李强强, 李小军, 李轶鲲, 等自适应膨胀和结构嵌入的非对称哈希遥感图像检索算法[J]. 地球信息科学学报, 2024, 26 (8): 1926- 1940 Li Qiangqiang, Li Xiaojun, Li Yikun, et al An adaptive dilated and structural embedding asymmetric hashing algorithm for remote sensing image retrieval[J]. Journal of Geo-Information Science, 2024, 26 (8): 1926- 1940 doi: 10.12082/dqxxkx.2024.240168
5	Tu T M, Su S C, Shyu H C, et al A new look at IHS-like image fusion methods[J]. Information Fusion, 2001, 2 (3): 177- 186 doi: 10.1016/S1566-2535(01)00036-7
6	Kwarteng P, Chavez A Extracting spectral contrast in Landsat Thematic Mapper image data using selective principal component analysis[J]. Photogrammetric Engineering and Remote Sensing, 1989, 55 (1): 339- 348
7	Vivone G Robust band-dependent spatial-detail approaches for panchromatic sharpening[J]. IEEE Transactions on Geoscience and Remote Sensing, 2019, 57 (9): 6421- 6433 doi: 10.1109/TGRS.2019.2906073
8	Vivone G, Restaino R, Dalla Mura M, et al Contrast and error-based fusion schemes for multispectral image pansharpening[J]. IEEE Geoscience and Remote Sensing Letters, 2014, 11 (5): 930- 934 doi: 10.1109/LGRS.2013.2281996
9	Aiazzi B, Alparone L, Baronti S, et al Context-driven fusion of high spatial and spectral resolution images based on oversampled multiresolution analysis[J]. IEEE Transactions on Geoscience and Remote Sensing, 2002, 40 (10): 2300- 2312 doi: 10.1109/TGRS.2002.803623
10	Vivone G, Restaino R, Chanussot J Full scale regression-based injection coefficients for panchromatic sharpening[J]. IEEE Transactions on Image Processing, 2018, 27 (7): 3418- 3431 doi: 10.1109/TIP.2018.2819501
11	Wei Q, Dobigeon N, Tourneret J Y Fast fusion of multi-band images based on solving a sylvester equation[J]. IEEE Transactions on Image Processing, 2015, 24 (11): 4109- 4121 doi: 10.1109/TIP.2015.2458572
12	Masi G, Cozzolino D, Verdoliva L, et al Pansharpening by convolutional neural networks[J]. Remote Sensing, 2016, 8 (7): 594 doi: 10.3390/rs8070594
13	He L, Zhu J W, Li J, et al HyperPNN: hyperspectral pansharpening via spectrally predictive convolutional neural networks[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2019, 12 (8): 3092- 3100 doi: 10.1109/JSTARS.2019.2917584
14	Zhuo Y W, Zhang T J, Hu J F, et al A deep-shallow fusion network with multidetail extractor and spectral attention for hyperspectral pansharpening[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2022, 15: 7539- 7555 doi: 10.1109/JSTARS.2022.3202866
15	Zheng Y, Li J, Li Y, et al Hyperspectral pansharpening using deep prior and dual attention residual network[J]. IEEE Transactions on Geoscience and Remote Sensing, 2020, 58 (11): 8059- 8076 doi: 10.1109/TGRS.2020.2986313
16	Bandara W G C, Valanarasu J M J, Patel V M Hyperspectral pansharpening based on improved deep image prior and residual reconstruction[J]. IEEE Transactions on Geoscience and Remote Sensing, 2022, 60: 5520816 doi: 10.36227/techrxiv.14912976
17	Liu J, Wu Z, Xiao L A spectral diffusion prior for unsupervised hyperspectral image super-resolution[J]. IEEE Transactions on Geoscience and Remote Sensing, 2024, 62: 5528613
18	Meng Q, Shi W, Li S, et al PanDiff: a novel pansharpening method based on denoising diffusion probabilistic model[J]. IEEE Transactions on Geoscience and Remote Sensing, 2023, 61: 5611317 doi: 10.1109/tgrs.2023.3279864
19	Li S, Li S, Zhang L Hyperspectral and panchromatic images fusion based on the dual conditional diffusion models[J]. IEEE Transactions on Geoscience and Remote Sensing, 2023, 61: 5526315 doi: 10.1109/tgrs.2023.3321318
20	Jiang H, Chen Z Transformer-based diffusion and spectral priors model for hyperspectral pansharpening[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2025, 18: 18962- 18977 doi: 10.1109/JSTARS.2025.3590685
21	Guan P, Lam E Y Multistage dual-attention guided fusion network for hyperspectral pansharpening[J]. IEEE Transactions on Geoscience and Remote Sensing, 2022, 60: 5515214 doi: 10.1109/tgrs.2021.3114552
22	Yuan Q, Wei Y, Meng X, et al A multiscale and multidepth convolutional neural network for remote sensing imagery pan-sharpening[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2018, 11 (3): 978- 989 doi: 10.1109/JSTARS.2018.2794888
23	Zhang Y, Yang X, Li H, et al DCPNet: a dual-task collaborative promotion network for pansharpening[J]. IEEE Transactions on Geoscience and Remote Sensing, 2024, 62: 5404016 doi: 10.1109/tgrs.2024.3377635
24	Liu Z, Lin Y, Cao Y, et al. Swin transformer: hierarchical vision transformer using shifted windows [C]//Proceedings of the IEEE/CVF International Conference on Computer Vision. Montreal: IEEE, 2021: 9992–10002.
25	Vrabel J Multispectral imagery advanced band sharpening study[J]. Photogrammetric Engineering and Remote Sensing, 2000, 66 (1): 73- 80
26	Simões M, Bioucas-dias J, Almeida L B, et al A convex formulation for hyperspectral image superresolution via subspace-based regularization[J]. IEEE Transactions on Geoscience and Remote Sensing, 2015, 53 (6): 3373- 3388 doi: 10.1109/TGRS.2014.2375320
27	Rui X, Cao X, Pang L, et al Unsupervised hyperspectral pansharpening via low-rank diffusion model[J]. Information Fusion, 2024, 107: 102325 doi: 10.1016/j.inffus.2024.102325
28	Hou S, Xiao S, Qu J, et al Multiscale common-private feature adversarial decoupling network for hyperspectral pansharpening[J]. Knowledge-Based Systems, 2025, 310: 113031 doi: 10.1016/j.knosys.2025.113031
29	Liu C, Zhang Z, Wang M RAMSF: a novel generic framework for optical remote sensing multimodal spatial-spectral fusion[J]. IEEE Transactions on Geoscience and Remote Sensing, 2025, 63: 5617722 doi: 10.1109/tgrs.2025.3552937
30	Wald L, Ranchin T, Mangolini M Fusion of satellite images of different spatial resolutions: assessing the quality of resulting images[J]. Photogrammetric Engineering and Remote Sensing, 1997, 63: 691- 699
31	杨睿哲, 李小军, 张剑鑫, 等多尺度级联的遥感图像全色锐化融合网络[J]. 激光与光电子学进展, 2025, 62 (22): 306- 314 Yang Ruizhe, Li Xiaojun, Zhang Jianxin, et al Multi-scale cascaded remote sensing image pansharpening fusion network[J]. Laser and Optoelectronics Progress, 2025, 62 (22): 306- 314 doi: 10.3788/LOP250991
32	Vivone G, Dalla Mura M, Garzelli A, et al A new benchmark based on recent advances in multispectral pansharpening: revisiting pansharpening with classical and emerging pansharpening methods[J]. IEEE Geoscience and Remote Sensing Magazine, 2021, 9 (1): 53- 81 doi: 10.1109/MGRS.2020.3019315
33	Sara D, Mandava A K, Kumar A, et al Hyperspectral and multispectral image fusion techniques for high resolution applications: a review[J]. Earth Science Informatics, 2021, 14 (4): 1685- 1705 doi: 10.1007/s12145-021-00621-6
34	Ren K, Sun W, Meng X, et al Fusing China GF-5 hyperspectral data with GF-1, GF-2 and sentinel-2A multispectral data: which methods should be used?[J]. Remote Sensing, 2020, 12 (5): 882 doi: 10.3390/rs12050882
35	Yokoya N, Grohnfeldt C, Chanussot J Hyperspectral and multispectral data fusion: a comparative review of the recent literature[J]. IEEE Geoscience and Remote Sensing Magazine, 2017, 5 (2): 29- 56 doi: 10.1109/MGRS.2016.2637824

[1]	连远锋,范树玉,张本哲,王森. 融合几何约束特征提取与多奖励协同优化的仿生步态控制[J]. 浙江大学学报(工学版), 2026, 60(9): 1862-1871.
[2]	火久元,姜灏,常琛. 基于改进EAST的高速公路交通标识文本检测[J]. 浙江大学学报(工学版), 2026, 60(9): 1881-1889.
[3]	李珺,丁彬彬,史维娟,杨琳. 基于多尺度特征聚合的航拍图像检测算法[J]. 浙江大学学报(工学版), 2026, 60(9): 1890-1900.
[4]	武晓春,李梓宁. 基于YOLOv11n的改进铁路工人安全穿戴检测模型[J]. 浙江大学学报(工学版), 2026, 60(9): 1901-1911.
[5]	邬开俊,郑云琦,魏鼎,袁海翔. 基于YOLOv8s的轻量化航拍图像小目标检测算法[J]. 浙江大学学报(工学版), 2026, 60(9): 1912-1923.
[6]	于鑫淼,夏楠,江佳鸿,郝子莹,把云胜. 基于多尺度特征相似性匹配的低照度目标检测[J]. 浙江大学学报(工学版), 2026, 60(7): 1464-1474.
[7]	张乃洲,赵云超,曹薇,张啸剑. 基于多视图跨模态特征融合的图像描述生成[J]. 浙江大学学报(工学版), 2026, 60(6): 1205-1212.
[8]	李国燕,于威,梅玉鹏,张明辉,王新强. 全局局部特征融合的遥感图像建筑物提取[J]. 浙江大学学报(工学版), 2026, 60(5): 1100-1108.
[9]	于天河,王文龙,刘镛,杨壮壮,侯善冲. 改进的有雾图像中被遮挡车辆及行人识别算法[J]. 浙江大学学报(工学版), 2026, 60(4): 738-750.
[10]	李国燕,李鹏辉,刘榕,梅玉鹏,张明辉. 融合多尺度分辨率和带状特征的遥感道路提取[J]. 浙江大学学报(工学版), 2026, 60(3): 585-593.
[11]	雒伟群,陆敬蔚,吴佳缔,梁钰迎,申传鹏,朱睿. 藏区高原典型环境地形目标的轻量化检测模型[J]. 浙江大学学报(工学版), 2026, 60(3): 594-603.
[12]	闫光辉,黄霄,常文文. 基于脑电多尺度特征和图神经网络的紧急制动行为识别[J]. 浙江大学学报(工学版), 2026, 60(2): 404-414.
[13]	刘春娟,包易航,吴小所,闫浩文,朱铁栓,赵浩然. 基于细节增强和跨尺度注意力融合的PCB缺陷检测算法[J]. 浙江大学学报(工学版), 2026, 60(11): 2431-2441.
[14]	刘宇轩,苏恩祈,井佩光,孙宏博,苏育挺. 基于多任务对比约束与可控缺陷生成的异常检测算法[J]. 浙江大学学报(工学版), 2026, 60(11): 2442-2452.
[15]	黄腾飞,火久元,曹芳,武泽宇,孔垂乐. 面向小目标与模糊目标的轨道缺陷检测框架[J]. 浙江大学学报(工学版), 2026, 60(11): 2453-2461.

Viewed

Full text

Abstract

Cited

Shared

Discussed