
浏览全部资源
扫码关注微信
1.宁夏工商职业技术大学 信息技术学院, 宁夏 银川 750021
2.宁夏大学 电子与电气工程学院, 宁夏 银川 750021
3.宁夏大学 前沿交叉学院, 宁夏 中卫 755000
马鑫,男,硕士,讲师,2017年于宁夏大学获得硕士学位,主要从事遥感图像处理方面的研究。E-mail: 2023012@nxgs.edu.cn
收稿:2026-03-11,
修回:2026-04-02,
纸质出版:2026-07-06
移动端阅览
马鑫, 汪西原, 白雪冰. 高低频增强与跨层图卷积聚合的高光谱图像分类[J]. 液晶与显示, 2026,41(7):995-1008.
MA Xin, WANG Xiyuan, BAI Xuebing. Hyperspectral image classification via high-low frequency enhancement and cross-layer graph convolution aggregation[J]. Chinese Journal of Liquid Crystals and Displays, 2026, 41(7): 995-1008.
马鑫, 汪西原, 白雪冰. 高低频增强与跨层图卷积聚合的高光谱图像分类[J]. 液晶与显示, 2026,41(7):995-1008. DOI: 10.37188/CJLCD.2026-0034. CSTR: 32172.14.CJLCD.2026-0034.
MA Xin, WANG Xiyuan, BAI Xuebing. Hyperspectral image classification via high-low frequency enhancement and cross-layer graph convolution aggregation[J]. Chinese Journal of Liquid Crystals and Displays, 2026, 41(7): 995-1008. DOI: 10.37188/CJLCD.2026-0034. CSTR: 32172.14.CJLCD.2026-0034.
针对高光谱图像分类中局部纹理与边缘细节易损失、卷积分支感受野受限以及图分支跨层结构信息利用不足等问题,本文提出一种高低频增强与跨层图卷积聚合的卷积-图卷积联合分类模型。该方法通过高低频残差增强提升输入特征质量,在卷积分支中利用多阶段动态卷积编码提取多尺度空间-光谱特征,在图卷积分支中采用跨层图特征加权聚合增强区域结构建模能力,并通过跨分支注意力融合实现两路特征的协同建模。实验在Indian Pines、Pavia University和Salinas 3个公开数据集上进行,总体分类精度分别达到92.94%、95.11%和97.50%,对应Kappa系数分别为91.94%、93.50%和97.22%。结果表明,所提方法能够兼顾局部细节、空间上下文和区域拓扑结构信息,在不同类型高光谱分类场景下具有较好的综合分类性能。
To address the problems of local texture and edge-detail loss, limited receptive fields in the convolution branch, and insufficient use of cross-layer structural information in the graph branch for hyperspectral image classification, this paper proposes a CNN-GCN joint classification model with high-low frequency enhancement and cross-layer graph convolution aggregation. The model improves input representation through high-low frequency residual enhancement, extracts multi-scale spectral-spatial features using a multi-stage dynamic convolution encoder, enhances regional structural modeling by cross-layer weighted aggregation in the graph branch, and performs collaborative modeling through cross-branch attention fusion. Experiments on three public datasets, Indian Pines, Pavia University, and Salinas, achieve overall accuracies of 92.94%, 95.11%, and 97.50%, with corresponding Kappa coefficients of 91.94%, 93.50%, and 97.22%, respectively. The results show that the proposed method can effectively integrate local details, spatial context, and regional topological structure information, achieving competitive classification performance across different types of hyperspectral classification scenes.
何明一 , 李强 , 郭英 , 等 . 高光谱图像智能分类研究综述与展望 [J]. 中国图象图形学报 , 2025 , 30 ( 6 ): 2207 - 2238 . doi: 10.11834/jig.250045 http://dx.doi.org/10.11834/jig.250045
HE M Y , LI Q , GUO Y , et al . Overview and prospects of intelligent classification of hyperspectral images [J]. Journal of Image and Graphics , 2025 , 30 ( 6 ): 2207 - 2238 . (in Chinese) . doi: 10.11834/jig.250045 http://dx.doi.org/10.11834/jig.250045
PARK J , CHOI Y , KWON S , et al . Microsphere-assisted hyperspectral imaging: super-resolution, non-destructive metrology for semiconductor devices [J]. Light: Science & Applications , 2024 , 13 ( 1 ): 122 . doi: 10.1038/s41377-024-01469-3 http://dx.doi.org/10.1038/s41377-024-01469-3
CEN Q Q , PIAN S , LIU X H , et al . Microtaper leaky-mode spectrometer with picometer resolution [J]. eLight , 2023 , 3 ( 1 ): 9 . doi: 10.1186/s43593-023-00041-7 http://dx.doi.org/10.1186/s43593-023-00041-7
KUHN J , LODIEU N , LÓPEZ R R , et al . Creating ground-based telescopes for imaging interferometry: SELF and ELF [J]. Light: Advanced Manufacturing , 2025 , 6 ( 3 ): 33 . doi: 10.37188/lam.2025.033 http://dx.doi.org/10.37188/lam.2025.033
BAI L F , CHEN Z J , ZHOU L , et al . HSI classification: a comprehensive review, dataset, and future trends [J]. Journal of the Indian Society of Remote Sensing , 2025 , 53 ( 12 ): 3983 - 4004 . doi: 10.1007/s12524-025-02275-z http://dx.doi.org/10.1007/s12524-025-02275-z
ASKER M E . Hyperspectral image classification method based on squeeze-and-excitation networks, depthwise separable convolution and multibranch feature fusion [J]. Earth Science Informatics , 2023 , 16 ( 2 ): 1427 - 1448 . doi: 10.1007/s12145-023-00982-0 http://dx.doi.org/10.1007/s12145-023-00982-0
景海钊 , 陶丽杰 , 张号逵 . 3D-CNN与Transformer混合结构的高光谱图像空谱联合分类 [J]. 光学 精密工程 , 2024 , 32 ( 23 ): 3504 - 3512 . doi: 10.37188/ope.20243223.3504 http://dx.doi.org/10.37188/ope.20243223.3504
JING H Z , TAO L J , ZHANG H K . Specral-spatial classification of hyperspectral imagery with hybrid architecture of 3D-CNN and Transformer [J]. Optics and Precision Engineering , 2024 , 32 ( 23 ): 3504 - 3512 . (in Chinese) . doi: 10.37188/ope.20243223.3504 http://dx.doi.org/10.37188/ope.20243223.3504
刘和 , 宋璎珞 , 胡龙湘 , 等 . 基于空间金字塔注意力机制残差网络的高光谱图像分类 [J]. 液晶与显示 , 2024 , 39 ( 6 ): 833 - 843 . doi: 10.37188/cjlcd.2023-0175 http://dx.doi.org/10.37188/cjlcd.2023-0175
LIU H , SONG Y L , HU L X , et al . Hyperspectral image classification based on spatial pyramid attention mechanism combined with ResNet [J]. Chinese Journal of Liquid Crystals and Displays , 2024 , 39 ( 6 ): 833 - 843 . (in Chinese) . doi: 10.37188/cjlcd.2023-0175 http://dx.doi.org/10.37188/cjlcd.2023-0175
陈海松 , 张康 , 吕浩然 , 等 . 结合高效注意力机制的神经架构搜索高光谱图像分类 [J]. 液晶与显示 , 2025 , 40 ( 4 ): 630 - 641 . doi: 10.37188/cjlcd.2024-0254 http://dx.doi.org/10.37188/cjlcd.2024-0254
CHEN H S , ZHANG K , LYU H R , et al . Neural architecture search combined with efficient attention for hyperspectral image classification [J]. Chinese Journal of Liquid Crystals and Displays , 2025 , 40 ( 4 ): 630 - 641 . (in Chinese) . doi: 10.37188/cjlcd.2024-0254 http://dx.doi.org/10.37188/cjlcd.2024-0254
李英超 , 赵喆浩 , 王祺 , 等 . 面向地物混杂背景的偏振光谱图像融合方法 [J]. 中国光学 , 2024 , 17 ( 5 ): 1098 - 1111 . doi: 10.37188/co.2023-0185 http://dx.doi.org/10.37188/co.2023-0185
LI Y C , ZHAO Z H , WANG Q , et al . Polarization spectral image fusion method for hybrid backgrounds of ground objects [J]. Chinese Optics , 2024 , 17 ( 5 ): 1098 - 1111 . (in Chinese) . doi: 10.37188/co.2023-0185 http://dx.doi.org/10.37188/co.2023-0185
刘敬 , 李银桥 , 刘逸 . 主动学习联合聚类分组网络的高光谱遥感图像分类 [J]. 光学 精密工程 , 2024 , 32 ( 9 ): 1395 - 1407 . doi: 10.37188/ope.20243209.1395 http://dx.doi.org/10.37188/ope.20243209.1395
LIU J , LI Y Q , LIU Y . Active learning-clustering-group convolutions network for hyperspectral images classification [J]. Optics and Precision Engineering , 2024 , 32 ( 9 ): 1395 - 1407 . (in Chinese) . doi: 10.37188/ope.20243209.1395 http://dx.doi.org/10.37188/ope.20243209.1395
邸敬 , 梁婵 , 刘冀钊 , 等 . 基于跨域交互注意力和对比学习引导的红外与可见光图像融合 [J]. 中国光学 , 2025 , 18 ( 2 ): 317 - 332 . doi: 10.37188/co.2024-0147 http://dx.doi.org/10.37188/co.2024-0147
DI J , LIANG C , LIU J Z , et al . Infrared and visible image fusion guided by cross-domain interactive attention and contrastive learning [J]. Chinese Optics , 2025 , 18 ( 2 ): 317 - 332 . (in Chinese) . doi: 10.37188/co.2024-0147 http://dx.doi.org/10.37188/co.2024-0147
吴海滨 , 戴诗语 , 王爱丽 , 等 . CNN-Transformer结合对比学习的高光谱与LiDAR数据协同分类 [J]. 光学 精密工程 , 2024 , 32 ( 7 ): 1087 - 1100 .
WU H B , DAI S Y , WANG A L , et al . Collaborative classification of hyperspectral and LiDAR data based on CNN-transformer [J]. Optics and Precision Engineering , 2024 , 32 ( 7 ): 1087 - 1100 . (in Chinese)
ZHANG Z T , ZHANG K , ZHANG C L , et al . LDS 2 MLP: A novel learnable dilated spectral-spatial MLP for hyperspectral image classificatio n [J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing , 2024 , 17 : 17207 - 17220 . doi: 10.1109/jstars.2024.3441709 http://dx.doi.org/10.1109/jstars.2024.3441709
ZHAO X F , MA J Y , WANG L , et al . A review of hyperspectral image classification based on graph neural networks [J]. Artificial Intelligence Review , 2025 , 58 ( 6 ): 172 . doi: 10.1007/s10462-025-11169-y http://dx.doi.org/10.1007/s10462-025-11169-y
HANACHI R , SELLAMI A , FARAH I R , et al . Multi-view graph representation learning for hyperspectral image classification with spectral-spatial graph neural networks [J]. Neural Computing and Applications , 2024 , 36 ( 7 ): 3737 - 3759 . doi: 10.1007/s00521-023-09275-5 http://dx.doi.org/10.1007/s00521-023-09275-5
LIU J , LI T , ZHAO F , et al . Dual graph convolutional network for hyperspectral images with spatial graph and spectral multigraph [J]. IEEE Geoscience and Remote Sensing Letters , 2024 , 21 : 5506105 . doi: 10.1109/lgrs.2024.3398439 http://dx.doi.org/10.1109/lgrs.2024.3398439
ZHOU H , LUO F L , ZHUANG H P , et al . Attention multihop graph and multiscale convolutional fusion network for hyperspectral image classification [J]. IEEE Transactions on Geoscience and Remote Sensing , 2023 , 61 : 5508614 . doi: 10.1109/tgrs.2023.3265879 http://dx.doi.org/10.1109/tgrs.2023.3265879
GAO H M , SHENG R H , SU Y C , et al . Multiscale segmentation-guided fusion network for hyperspectral image classification [J]. IEEE Transactions on Image Processing , 2025 , 34 : 6152 - 6167 .
YUE J , ZHAO W Z , MAO S J , et al . Spectral-spatial classification of hyperspectral images using deep convolutional neural networks [J]. Remote Sensing Letters , 2015 , 6 ( 6 ): 468 - 477 . doi: 10.1080/2150704x.2015.1047045 http://dx.doi.org/10.1080/2150704x.2015.1047045
Wang W J , Dou S G , Jiang Z M , et al . A fast dense spectral-spatial convolution network framework for hyperspectral images classification [J]. Remote Sensing , 2018 , 10 ( 7 ): 1068 . doi: 10.3390/rs10071068 http://dx.doi.org/10.3390/rs10071068
XI B B , ZHANG Y , LI J J , et al . MCTGCL: mixed CNN-transformer for mars hyperspectral image classification with graph contrastive learning [J]. IEEE Transactions on Geoscience and Remote Sensing , 2025 , 63 : 5503214 . doi: 10.1109/tgrs.2025.3529996 http://dx.doi.org/10.1109/tgrs.2025.3529996
LIU Q C , XIAO L , YANG J X , et al . CNN-enhanced graph convolutional network with pixel- and superpixel-level feature fusion for hyperspectral image classification [J]. IEEE Transactions on Geoscience and Remote Sensing , 2021 , 59 ( 10 ): 8657 - 8671 . doi: 10.1109/tgrs.2020.3037361 http://dx.doi.org/10.1109/tgrs.2020.3037361
0
浏览量
374
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621