基于知识图谱语义表示的高光谱遥感图像零样本分类

doi:10.3799/dqkx.2026.210

基于知识图谱语义表示的高光谱遥感图像零样本分类

doi: 10.3799/dqkx.2026.210

马丽^1,2, ,,
汤睿婕²,
李彦胜³

1. 海南大学生物医学工程学院, 海南三亚 572024;
2. 中国地质大学(武汉) 机械与电子信息学院, 湖北武汉 430074;
3. 武汉大学遥感信息工程学院, 湖北武汉 430079

基金项目:

国家自然科学基金项目(No.61771437)

智慧地球重点实验室开放基金(KF2023YB04-05)

详细信息

作者简介:
马丽(1982-),女,教授,博士生导师,主要从事机器学习,深度学习等方面的研究.E-mail:maryparisster@gmail.com.ORCID:0000-0003-3873-5080,E-mail:maryparisster@gmail.com

通讯作者:
马丽，E-mail： maryparisster@gmail.com.

中图分类号: P237;TP751
计量
- 文章访问数: 64
- HTML全文浏览量: 5
- PDF下载量: 12
- 被引次数: 0
出版历程
- 收稿日期: 2026-01-14
- 网络出版日期: 2026-08-25

Zero-Shot Classification of Hyperspectral Remote Sensing Images Based on Knowledge Graph Semantic Representations

Ma Li^{1,2
, ,},
Tang Ruijie²,
Li Yansheng³

1. School of Biomedical Engineering, Hainan University, Sanya 572024, China;
2. School of Mechanical Engineering and Electronic Information, China University of Geosciences, Wuhan 430074, China;
3. School of Remote Sensing and Information Engineering, Wuhan University, Wuhan 430079, China

摘要

摘要: 针对高光谱遥感图像标注类别不完整条件下未知地物类别难以被传统监督学习方法有效识别的问题，本文提出一种基于知识图谱语义表示的高光谱遥感图像零样本分类网络（Knowledge Graph semantic representations based Zero-Shot Network for hyperspectral remote sensing image classification，KG-ZSNet）．该方法通过构建遥感领域知识图谱获取类别层面的丰富语义表示，并在生成式零样本学习框架下引入变分自编码器，实现视觉特征与语义特征的跨模态对齐，进而在共享隐层空间中完成未知类别分类．在多个公开高光谱遥感数据集上的实验结果表明，所提出方法在零样本像素级分类任务中的分类精度优于多种典型零样本学习方法，通过引入结构化知识图谱并结合生成式跨模态对齐机制，在高光谱遥感图像零样本分类任务中表现出良好的有效性与稳定性．
- 高光谱遥感图像 /
- 零样本学习 /
- 知识图谱 /
- 变分自编码器 /
- 未知类分类
Abstract: Under incomplete category annotation conditions, traditional supervised learning methods have difficulty in effectively recognizing unseen land-cover categories in hyperspectral remote sensing images. This paper proposes a Knowledge Graph semantic representations based Zero-Shot Network for hyperspectral remote sensing image classification(KG-ZSNet). We construct a remote sensing knowledge graph to obtain category-level semantic representations and integrate variational autoencoder into a generative zero-shot learning framework to achieve cross-modal alignment between visual and semantic features, thereby enabling unseen category classification in shared latent space. Experiments on multiple public hyperspectral datasets demonstrate that KG-ZSNet achieves higher accuracy than several representative zero-shot learning methods in zero-shot hyperspectral remote sensing image classification. KG-ZSNet, by incorporating structured knowledge-graph semantics and the generative cross-modal alignment mechanism, achieves strong effectiveness and stability in zero-shot hyperspectral remote sensing image classification.
- hyperspectral remote sensing images /
- zero-shot learning /
- knowledge graph /
- variational autoencoder /
- unseen class classification

HTML全文

参考文献(42)

Akata,Z., Perronnin,F., Harchaoui,Z., et al., 2013. Label-Embedding for Attribute-Based Classification. IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Portland, OR, USA: IEEE, 819-826. https://doi.org/10.1109/CVPR.2013.111.
Akata,Z., Reed,S., Walter,D., et al., 2015. Evaluation of Output Embeddings for Fine-Grained Image Classification. IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Boston, MA, USA: IEEE, 2927-2936. https://doi.org/10.1109/CVPR.2015.7298911.
Bollacker,K., Evans,C., Paritosh,P., et al., 2008. Freebase: A Collaboratively Created Graph Database for Structuring Human Knowledge. International Conference on Management of Data (SIGMOD). Vancouver Canada: ACM, 1247-1250. https://doi.org/10.1145/1376616.1376746.
Bordes,A., Usunier,N., Garcia-Durán,A., et al., 2013. Translating Embeddings for Modeling Multi-Relational Data. International Conference on Neural Information Processing Systems (NeurIPS). Red Hook, NY, USA: Curran Associates Inc., 2787-2795.
Chen,J.D., Wang,A., Chen,J.J., et al., 2019. CN-Probase: A Data-Driven Approach for Large-Scale Chinese Taxonomy Construction. IEEE International Conference on Data Engineering (ICDE). Macao, China: IEEE, 1706-1709. https://doi.org/10.1109/ICDE.2019.00178.
Chen,J.Y., Geng,Y.X., Chen,Z., et al., 2023. Zero-Shot and Few-Shot Learning with Knowledge Graphs: A Comprehensive Survey.Proceedings of the IEEE, 111(6): 653-685. https://doi.org/10.1109/JPROC.2023.3279374.
Chen,W.T., Xu,J.H., Wang R., et al., 2025. Hyperspectral Remote Sensing Inversion of Black Soil Organic Matter Content Based on Multi-Scale Feature Enhancement.Earth Science, 50(12): 4909-4918(in Chinese with English abstract).
Chen,Z.X., Wang,Z.H., Bu,H.J., et al., 2026. A Geological Information and Enhanced Graph Convolutional Network Method for Extracting Information on Lithium and Beryllium-Rich Pegmatites from Hyperspectral Imagery.Earth Science, 51(3): 1057-1064(in Chinese with English abstract).
Fensel,D., Şimşek,U., Angele,K., et al., 2020. Introduction: What Is a Knowledge Graph?. Knowledge Graphs. Cham: Springer International Publishing, 1-10. https://doi.org/10.1007/978-3-030-37439-6_1.
Freitas,S., Silva,H., Silva,E., 2022. Hyperspectral Imaging Zero-Shot Learning for Remote Marine Litter Detection and Classification.Remote Sensing, 14(21): Art.no.5516. https://doi.org/10.3390/rs14215516.
Frome,A., Corrado,G.S., Shlens,J., et al., 2013. DeViSE: A Deep Visual-Semantic Embedding Model. Advances in Neural Information Processing Systems (NIPS). Lake Tahoe, Nevada: Curran Associates Inc., 2121-2129.
Goodfellow,I.J., Pouget-Abadie,J., Mirza,M., et al., 2014. Generative Adversarial Networks. arXiv:1406.2661. https://doi.org/10.48550/arXiv.1406.2661.
Jia,C., Yang,Y.F., Xia,Y., et al., 2021. Scaling Up Visual and Vision-Language Representation Learning with Noisy Text Supervision. 38th International Conference on Machine Learning (ICML). PMLR, 4904-4916.
Jiang,B.C., Wan,G., Xu,J., et al., 2018. Geographic Knowledge Graph Building Extracted from Multi-Sourced Heterogeneous Data.Acta Geodaetica et Cartographica Sinica, 47(8): 1051-1061(in Chinese with English abstract).
Kingma,D.P., Welling,M., 2022. Auto-Encoding Variational Bayes. arXiv:1312.6114. https://doi.org/10.48550/arXiv.1312.6114.
Kodirov,E., Xiang,T., Gong,S.G., 2017. Semantic Autoencoder for Zero-Shot Learning. IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Honolulu, HI: IEEE, 4447-4456. https://doi.org/10.1109/CVPR.2017.473.
Kumar,V., Singh,R.S., Rambabu,M., et al., 2024. Deep Learning for Hyperspectral Image Classification: A survey.Computer Science Review, 53: Art.no.100658. https://doi.org/10.1016/j.cosrev.2024.100658.
Lehmann,J., Isele,R., Jakob,M., et al., 2015. Dbpedia-A Large-Scale, Multilingual Knowledge Base Extracted from Wikipedia.Semantic Web, 6(2): 167-195. https://doi.org/10.3233/SW-140134.
Li,Y.S., Kong,D.Y., Zhang,Y.J., et al., 2021. Robust Deep Alignment Network with Remote Sensing Knowledge Graph for Zero-shot and Generalized Zero-Shot Remote Sensing Image Scene Classification.ISPRS Journal of Photogrammetry and Remote Sensing, 179: 145-158. https://doi.org/10.1016/j.isprsjprs.2021.08.001.
Li,Y.S., Wu,M.L., Zhang,Y.J., 2024. Knowledge Graph-Guided Deep Network for High-Resolution Remote Sensing Image Scene Classification.Acta Geodaetica et Cartographica Sinica, 53(4): 677-688(in Chinese with English abstract).
Li,Y.S., Wang,Y., Yu,L., et al., 2025. Learning to Reason over Multi-Granularity Knowledge Graph for Zero-Shot Urban Land-Use Mapping.Remote Sensing of Environment, 330: Art.no.114961. https://doi.org/10.1016/j.rse.2025.114961.
Liu,C., Ma,S.Q., Li,Z., et al., 2023. Multi-Level Cross-Modal Feature Alignment via Contrastive Learning towards Zero-shot Classification of Remote Sensing Image Scenes. arXiv:2306.06066.https://doi.org/10.48550/arXiv.2306.06066.
Mishra,A., Reddy,S.K., Mittal,A., et al., 2018. A Generative Model For Zero Shot Learning Using Conditional Variational Autoencoders. arXiv:1709.00663. https://doi.org/10.48550/arXiv.1709.00663.
Pan,E., Ma,Y., Fan,F., et al., 2021. Hyperspectral Image Classification across Different Datasets: A Generalization to Unseen Categories.Remote Sensing, 13(9): Art.no.1672. https://doi.org/10.3390/rs13091672.
Pourpanah,F., Abdar,M., Luo,Y.X., et al., 2023. A Review of Generalized Zero-Shot Learning Methods.IEEE Transactions on Pattern Analysis and Machine Intelligence, 45(4): 4051-4070. https://doi.org/10.1109/TPAMI.2022.3191696.
Romera-Paredes,B., Torr,P., 2015. An Embarrassingly Simple Approach to Zero-Shot Learning. 32nd International Conference on Machine Learning (ICML). Lille, France: PMLR, 2152-2161. https://doi.org/10.1007/978-3-319-50077-5_2.
Sun,Z.Q., Deng,Z.H., Nie,J.Y., et al., 2019. RotatE: Knowledge Graph Embedding by Relational Rotation in Complex Space. arXiv:1902.10197. http://arxiv.org/abs/1902.10197.
Sung,F., Yang,Y.X., Zhang,L., et al., 2018. Learning to Compare: Relation Network for Few-Shot Learning. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Salt Lake City, UT, USA: IEEE, 1199-1208. https://doi.org/10.1109/CVPR.2018.00131.
Tan,X.M., Xi,B.B., Li,J.J., et al., 2024. Review of Zero-Shot Remote Sensing Image Scene Classification.IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 17: 11274-11289. https://doi.org/10.1109/JSTARS.2024.3410995.
Trouillon,T., Welbl,J., Riedel,S., et al., 2016. Complex Embeddings for Simple Link Prediction. 33rd International Conference on International Conference on Machine Learning (ICML). New York, NY, USA: PMLR, 2071-2080.
Wang,W., Zheng,V.W., Yu,H., et al., 2019. A Survey of Zero-Shot Learning: Settings, Methods, and Applications.ACM Transactions on Intelligent Systems and Technology, 10(2): 13:1-13:37. https://doi.org/10.1145/3293318.
Xie,G.S., Zhang,Z., Shao,L., et al., 2023. Towards Zero-Shot Learning: A Brief Review and an Attention-Based Embedding Network.IEEE Transactions on Circuits and Systems for Video Technology, 33(3). https://doi.org/10.1109/TCSVT.2022.3208071.
Xu,B., Xu,Y., Liang,J.Q., et al., 2017. CN-DBpedia: A Never-Ending Chinese Knowledge Extraction System. International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems (IEA/AIE). Arras, France: Springer, 428-438. https://doi.org/10.1007/978-3-319-60045-1_44.
Yang,B.S., Yih,W.T., He,X.D., et al., 2014. Embedding Entities and Relations for Learning and Inference in Knowledge Bases.International Conference on Learning Representations (ICLR). Banff, AB, Canada: OpenReview, 1-6.
Zhang,L., Xiang,T., Gong,S.G., 2017. Learning a Deep Embedding Model for Zero-Shot Learning. IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Honolulu, HI, USA: IEEE, 3010-3019. https://doi.org/10.1109/CVPR.2017.321.
Zhang,X.Y., Zhang,C.J., Wu,M.G., et al., 2020. Spatio-Temporal Features Based Geographical Knowledge Graph Construction.Scientia Sinica Informationis, 50(7): 1019-1032(in Chinese with English abstract).
Zhu,Y.Z., Elhoseiny,M., Liu,B.C., et al., 2018. A Generative Adversarial Approach for Zero-Shot Learning from Noisy Texts. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Salt Lake City, UT, USA: IEEE, 1004-1013. https://doi.org/10.1109/CVPR.2018.00111.
陈伟涛, 徐佳辉, 王锐, 等, 2025. 基于多尺度特征增强的黑土有机质含量高光谱卫星遥感反演. 地球科学, 50(12): 4909-4918.
陈志行, 王正海,卜浩坚, 等, 2025. 基于地质信息的改进GCN高光谱富锂铍伟晶岩信息提取方法. 地球科学, 51(3): 1057-1064.
蒋秉川, 万刚, 许剑, 等, 2018. 多源异构数据的大规模地理知识图谱构建. 测绘学报, 47(8): 1051-1061.
李彦胜, 吴敏郎, 张永军, 2024. 知识图谱约束深度网络的高分辨率遥感影像场景分类. 测绘学报, 53(4): 677-688.
张雪英, 张春菊, 吴明光, 等, 2020. 顾及时空特征的地理知识图谱构建方法. 中国科学:信息科学, 50(7): 1019-1032.

施引文献

资源附件(0)

访问统计

点击查看大图

计量

文章访问数: 64
HTML全文浏览量: 5
PDF下载量: 12
被引次数: 0

姓名
邮箱
手机号码
标题
留言内容
验证码

留言板

基于知识图谱语义表示的高光谱遥感图像零样本分类

doi: 10.3799/dqkx.2026.210

作者简介:
马丽(1982-),女,教授,博士生导师,主要从事机器学习,深度学习等方面的研究.E-mail:maryparisster@gmail.com.ORCID:0000-0003-3873-5080,E-mail:maryparisster@gmail.com

通讯作者:
马丽，E-mail： maryparisster@gmail.com.

计量

Zero-Shot Classification of Hyperspectral Remote Sensing Images Based on Knowledge Graph Semantic Representations

计量

目录

留言板

基于知识图谱语义表示的高光谱遥感图像零样本分类

doi: 10.3799/dqkx.2026.210

作者简介: 马丽(1982-),女,教授,博士生导师,主要从事机器学习,深度学习等方面的研究.E-mail:maryparisster@gmail.com.ORCID:0000-0003-3873-5080,E-mail:maryparisster@gmail.com

通讯作者: 马丽，E-mail： maryparisster@gmail.com.

计量

出版历程

Zero-Shot Classification of Hyperspectral Remote Sensing Images Based on Knowledge Graph Semantic Representations

计量

出版历程

目录

作者简介:
马丽(1982-),女,教授,博士生导师,主要从事机器学习,深度学习等方面的研究.E-mail:maryparisster@gmail.com.ORCID:0000-0003-3873-5080,E-mail:maryparisster@gmail.com

通讯作者:
马丽，E-mail： maryparisster@gmail.com.