知识图谱嵌入的光谱解混算法
Knowledge Graph Embedding Spectral Unmixing
- 2022年 页码:1-17
网络出版日期: 2022-12-16
DOI: 10.11834/jrs.20222253
扫 描 看 全 文
浏览全部资源
扫码关注微信
网络出版日期: 2022-12-16 ,
扫 描 看 全 文
吴瑞,罗文斐,陈江浩.XXXX.知识图谱嵌入的光谱解混算法.遥感学报,XX(XX): 1-17
Wu Rui,Luo Wenfei,Chen Jianghao. XXXX. Knowledge Graph Embedding Spectral Unmixing. National Remote Sensing Bulletin, XX(XX):1-17
在光谱解混过程中,在端元集合中选择有效的端元子集进行解混是至关重要的,但端元子集选择会受到端元光谱变异性的影响,导致选择的结果以及解混精度具有一定的不确定性。本文提出了一种知识图谱嵌入的光谱解混算法KGESU(Knowledge Graph Embedding Spectral Unmixing),在利用光谱特征进行解混的同时,引入一定的先验知识来进一步提高端元选择的可靠性,从而提升解混的精度。主要涉及两个核心问题:一是地学知识图谱的嵌入,二是引入先验的光谱解混。前者借助了TransE模型来实现对地学知识图谱的图嵌入训练;后者则在知识图谱嵌入的基础上进行知识推理,并把推理结果融入到光谱解混的过程,从而实现了一个引入先验知识的光谱解混算法。为了验证算法的有效性,通过高分五号高光谱相机所采集的真实图像进行实验,与其他经典算法作对比,结果表明本文的算法具有更好的解混效果,证明了在光谱解混过程中使用一定的地学先验知识有助于提升解混精度,KGESU算法具有一定的应用前景。
In the process of spectral unmixing
it is very important to select effective endmembers from a set of endmembers. However
the selection of endmembers will be affected by the spectral variability of endmembers
resulting in a certain uncertainty in the results of selection and the accuracy of unmixing. In order to solve this problem
this study combines geoscience prior knowledge with sparse unmixing
and proposes a Knowledge Graph Embedding Spectral Unmixing(KGESU) algorithm. While utilizing spectral features
certain prior knowledge is introduced to further improve the reliability of endmember selection.The implementation steps of the KGESU algorithm involves two issues
i.e. the embedding training of geoscience knowledge graph and spectral unmixing with priori knowledge. The embedding training of geoscience knowledge graph is to transform geoscience knowledge into a structured expression form through knowledge graph. Then TransE model is used for graph embedding. Considering the second issue
we perform knowledge reasoning according to the knowledge graph embedding. Then develop a reasoning-weighting sparse unmixing algorithm to integrate the process of the reasoning and unmixing.Experiments are conducted to validated the effectiveness of the proposed method. The prior knowledge was instantiated with the aid of auxiliary data such as Landsat8 and GDEMV2. The spectral unmixing data were GF-5 satellite data. The GF-2 data with a resolution of 1 meter after graphics fusion were used as the verification data. Compared with the traditional pixel-by-pixel evaluation
this paper expands the evaluation window. By increasing the overlap area between pixels and allocating the residuals
the sensitivity of different resolution images to registration errors is reduced. The root mean square error of each endmember
the mean of the root mean square error of each endmember and the overall root mean square error of the image are used as evaluation indexes to evaluate the unmixing results.The results demonstrate that the KGESU algorithm outperforms the state-of-the-art algorithms.By the guidance of geo-prior knowledge in the unmixing process
the uncertainty caused by factors such as data itself and external noise can be reduced. And the ability to discriminate endmembers can be improved to a certain extent. At the same time
the method proposed in this paper combines the advantages of knowledge reasoning and numerical computation. Furthermore
we use both geoscience knowledge and spectral characteristics to select the endmembers. The unmixing result can be more reliable. In the future
the research has the following issues that need further consideration. (1)In this paper
a knowledge graph is constructed only from the perspective of land use classification
and prior knowledge is introduced. In the follow-up work
secondary and even more precise classification can be considered to highlight the advantages of hyperspectral data.(2) In the future work
we will consider more complex relationships between ground objects
introduce more abundant geoscience knowledge
and further build a more perfect geoscience knowledge graph. (3) Knowledge reasoning based on graph embedding is a relatively good method to integrate reasoning results into spectral unmixing at present. With the continuous development of technology
we further try to introduce knowledge through other knowledge reasoning mechanisms.
知识图谱知识图谱嵌入端元子集选择光谱解混
knowledge graphknowledge graph embeddingendmember selectionspectral unmixing
Bhatt J S and Joshi M V. 2020. Deep Learning in Hyperspectral Unmixing:A Review. IEEE International Symposium on Geoscience and Remote Sensing IGARSS, pp, 2189-2192 [DOI: 10.1109/IGARSS39084.2020.9324546http://dx.doi.org/10.1109/IGARSS39084.2020.9324546]
Bioucas-Dias J M, Plaza A, Dobigeon N, Parente M, Du Q, Gader P and Chanussot J. 2012. Hyperspectral Unmixing Overview: Geometrical, Statistical, and Sparse Regression-Based Approaches. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 5(2): 354-79 [DOI: 10.1109/jstars.2012.2194696http://dx.doi.org/10.1109/jstars.2012.2194696]
Bordes A, Usunier N, Garcia-Duran A, Weston J and Yakhnenko O. 2013. Translating embeddings for modeling multi-relational Data. Neural Information Processing Systems (NIPS), Dec 2013, South Lake Tahoe, United States. pp.1-9. hal-00920777.
De'Ath G and Fabricius K. 2000. Classification and regression trees: Apowerful yet simple technique for ecological data analysis. Ecology,81(11):3178-92.https://www.researchgate.net/publication/312976423_Classification_and_regression_trees_A_powerful_yet_simple_technique_for_ecological_data_analysis(2021-10-01)https://www.researchgate.net/publication/312976423_Classification_and_regression_trees_A_powerful_yet_simple_technique_for_ecological_data_analysis(2021-10-01)
Dempster A P, Laird N M and RubinD B. 1997. Maximum Likelihood from Incomplete Data via the EM Algorithm. Journal of the Royal Statistical Society,39(1): 1-38. http://links.jstor.org/sici?sici=0035-9246%281977%2939%3A1%3C1%3AMLFIDV%3E2.0.CO%3B2-Z (2021-09-28)http://links.jstor.org/sici?sici=0035-9246%281977%2939%3A1%3C1%3AMLFIDV%3E2.0.CO%3B2-Z(2021-09-28)
Dobigeon N, Moussaoui S, Coulon M, Tourneret J Y and Hero A O. 2009. Joint Bayesian Endmember Extraction and Linear Unmixing for Hyperspectral Imagery.IEEE Transactions on Signal Processing, 57(11): 4355-68 [DOI: 10.1109/tsp.2009.2025797http://dx.doi.org/10.1109/tsp.2009.2025797]
Du C, Ren H Z, Qin Q M, Meng J J and Zhao S H. 2015. A Practical Split-Window Algorithm for Estimating Land Surface Temperature from Landsat 8 Data. Remote Sensing, 7 (1): 647-65 [DOI: 10.3390/rs70100647http://dx.doi.org/10.3390/rs70100647]
Ha K, Ding Q L, Men M X and Xu H. 2015. Land Use Distribution and Its Relationship with Topographic Factors in Mountainous and Hilly Areas——A Case Study of Huailai County, Hebei Province. Geographical Research, 34(5): 909-21
哈凯, 丁庆龙, 门明新.2015. 许皞山地丘陵区土地利用分布及其与地形因子关系——以河北省怀来县为例.地理研究, 34 (5): 909-21 [DOI: 10.11821/dlyj201505010http://dx.doi.org/10.11821/dlyj201505010]
Han Z, Hong D F, Gao L R, Zhang B and Chanussot J.2021.Deep Half-Siamese Networks for Hyperspectral Unmixing.IEEE Geoscience and Remote Sensing Letters. 18(11): 1996-2000 [DOI:10.1109/LGRS.2020.3011941http://dx.doi.org/10.1109/LGRS.2020.3011941]
Heylen R, Burazerovic D and Scheunders P.2011.Fully Constrained Least Squares Spectral Unmixing by Simplex Projection.IEEE Transactions on Geoscience and Remote Sensing, 49(11): 4112-4122 [DOI:10.1109/TGRS.2011.2155070http://dx.doi.org/10.1109/TGRS.2011.2155070]
Hou SW, Sun WF, Guo BL, Li XD, Zhang JH, Xu CZet al.2022.RFSDAF:A New Spatiotemporal Fusion Method Robust to Registration Errors, IEEE Transactions on Geoscien--ce and Remote Sensing, 60: 5616018 [DOI:10.1109/TGRS.20http://dx.doi.org/10.1109/TGRS.20
213138078] (Iordache M D, Bioucas-Dias J M and Plaza A. 2011. Sparse Unmixing of Hyperspectral Data. IEEE Transactions on Geoscience and Remote Sensing, 49(6): 2014-39 [DOI: 10.1109/tgrs.2010.2098413http://dx.doi.org/10.1109/tgrs.2010.2098413])
Jia L F, Duan J N and Qiao Z M. 2007. Correlation Analysis between Spatial Distribution of Land Use and Terrain Factors. Economic 200Geography, 27 (02): 310-2
贾宁凤, 段建南, 乔志敏.2007.土地利用空间分布与地形因子相关性分析方法.经济地理, 27(02): 310-2 [DOI: 10.15957/j.cnki.jjdl.2007http://dx.doi.org/10.15957/j.cnki.jjdl.2007.02029]
Jia Y, Cui M B, Wei X P, Gong D Z and Hu X T. 2016. Applicability Evaluation of Reference Crop Evapotranspiration Algorithm in Yangtze River Basin Based on Inverse Distance Weight Method. Transactions of the Chinese Society of Agricultural Engineering, 32(6): 130-8
贾悦, 崔宁博, 魏新平, 龚道枝, 胡笑涛.2016.基于反距离权重法的长江流域参考作物蒸散量算法适用性评价.农业工程学报, 32(6): 130-8 [DOI: 10.11975/j.issn.1002-6819.2016.06.018http://dx.doi.org/10.11975/j.issn.1002-6819.2016.06.018]
Li J, Agathos A, Zaharie D, Bioucas-Dias J M, Plaza A and Xia L. 2015. Minimum Volume Simplex Analysis: A Fast Algorithm for Linear Hyperspectral Unmixing. IEEE Transactions on Geoscience and Remote Sensing, 53(9): 5067-82 [DOI: 10.1109/tgrs.2015.2417162http://dx.doi.org/10.1109/tgrs.2015.2417162]
Li W L and Wu C S. 2015. Incorporating land use land cover probability information into endmember class selections for temporal mixture analysis. ISPRS Journal of Photogrammetry and Remote Sensing, 101: 163-73 [DOI: 10.1016/j.isprsjprs.201412.007http://dx.doi.org/10.1016/j.isprsjprs.201412.007] (LiY S, KongD Y, ZhangY J, Chen R X and ChenJ D,2021.Representation Learning of Remote Sensing Knowled-) (-ge Graph for Zero-Shot Remote Sensing Image Scene Classification. IEEE International Geoscience and Remote Sensing Symposium IGARSS, cp, 1351-4 [DOI:10.1109/IGARhttp://dx.doi.org/10.1109/IGAR) (SS47720.2021.9553667]) (LiY S, OuyangS and ZhangY J.2022.Combining deep learning and ontology reasoning for remote sensing image semantic segmentation. Knowledge-Based Systems, 243(2022)108469 [DOI:10.1016/j.knosys.2022.108469http://dx.doi.org/10.1016/j.knosys.2022.108469])
Liu X T, Zhou L, Shi H, Wang S Q and Chi Y G. 2018. Phenological characteristics of temperate coniferous and broad-leaved mixed forests based on multiple remote sensing vegetation indices, chlorophyII fluorescence and CO2 flux data. Acta Ecologica Sinica, 38 (10): 3482-3494
刘啸添, 周蕾, 石浩, 王绍强, 迟永刚. 2018. 基于多种遥感植被指数、叶绿素荧光与CO2通量数据的温带针阔混交林物候特征对比分析. 生态学报, 38(10): 3482-3494 [DOI:10.5846/stxb20http://dx.doi.org/10.5846/stxb201708211508]
Lu F, Li S, Cao J J, E X Z and Zhou Y. 2019. Point cloud boundary point detection algorithm based on inverse distance weight and density. Computer Engineering and Design, 40 (2): 364-9
陆帆, 李松, 曹菁菁, 鄂晓征,周勇.2019.基于反距离权重和密度的点云边界点检测算法.计算机工程与设计, 40(2): 364-9 [DOI: 10.16208/j.issn1000-7024.2019.02.012http://dx.doi.org/10.16208/j.issn1000-7024.2019.02.012]
Ma W K, Bioucas-Dias J M, Chan T H, Gillis N, Gader P, Plaza A J, Ambikapathi A and Chi C Y. 2014. A Signal Processing Perspective on Hyperspectral Unmixing. IEEE Signal Processing Magazine, 31 (1): 67-81 [DOI: 10.1109/MSP.2013.2279731http://dx.doi.org/10.1109/MSP.2013.2279731]
Nascimento J M P and Dias J M B. 2005. Vertex component analysis: a fast algorithm to unmix hyperspectral data. IEEE Transactions on Geoscience and Remote Sensing, 43(4): 898-910 [DOI: 10.1109/TGRS.2005.844293http://dx.doi.org/10.1109/TGRS.2005.844293]
Roberts D A, Gardner M, Church R, Ustin S, Scheer G and Green R O. 1998. Mapping chaparral in the Santa Monica Mountains using multiple endmember spectral mixture models. Remote Sensing of Environment, 65: 267-79 [DOI: 10.1016/S0034-4257(98)00037-6http://dx.doi.org/10.1016/S0034-4257(98)00037-6]
Shi S K, Zhang L J, Altmann Y and Chen J.2022.Deep Generative Model for Spatial-Spectral Unmixing With Multiple Endmember Priors.IEEE Transactions on Geoscience and Remote Sensing, 60: 1:14 [DOI:10.1109/TGRS.2022.3168712http://dx.doi.org/10.1109/TGRS.2022.3168712]
Singhal. 2012. Introducing the Knowledge Graph: Things, Not Strings.(2012-05-12):https://www.blog.google/products/search/introducing-knowledge-graph-things-not/ (2021-10-29
Sun Z Q, Deng Z H, Nie J Y and Tang J. 2019. Rotate:knowledge graph embedding by relational rotation complex space. ICLR 2019https://openreview.net/forum?id=HkgEQnRqYQ(2021-12-1)
Tong Q X, Zhang B and Zhang L F.. 2016. Current progress of hyperspectral remote sensing in China. National Remote Sensing Bulletin, 20(5): 689-707
童庆禧, 张兵, 张立福.2016.中国高光谱遥感的前沿进展.遥感学报, 20 (5): 689-707 [DOI: 10.11834/jrs.20166264http://dx.doi.org/10.11834/jrs.20166264]
Tong Q X, Zhang B and Zheng L F. 2006a. Hyperspectral Remote Sensing and It's Multidisciplinary Applications. Beijing: Publishing House of Electronics Industry
童庆禧, 张兵, 郑兰芬.2006a.高光谱遥感的多学科应用.北京:电子工业出版社
Tong Q X, Zhang B and Zheng L F. 2006b. Hyperspectral Remote Sensing: the Principle, Technology and Application. Beijing: Higher Education Press
童庆禧, 张兵, 郑兰芬.2006b.高光谱遥感:原理、技术与应用. 北京: 高等教育出版社
Wang J and Chang C I. 2006. Applications of Independent Component Analysis in Endmember Extraction and Abundance Quantification for Hyperspectral Imagery. IEEE Transactions on Geoscience and Remote Sensing, 44(9): 2601-16 [DOI: 10.1109/tgrs.2006.874135http://dx.doi.org/10.1109/tgrs.2006.874135]
Wang M Y,Luo Y,Zhang Z Y,Xie Q Y,Wu X D and Ma X L. 2022. Recent advances in remote sensing of vegetation phenology: Retrieval algorithm and validation strategy. National Remote Sensing Bulletin,26(3):431-455
王敏钰,罗毅,张正阳,谢巧云,吴小丹,马轩龙.2022.植被物候参数遥感提取与验证方法研究进展.遥感学报, 26(3): 431-455 [DOI: 10.11834/jrs.20211601http://dx.doi.org/10.11834/jrs.20211601]
Wang N, Du B and Zhang L P. 2013. An Endmember Dissimilarity Constrained Non-Negative Matrix Factorization Method for Hyperspectral Unmixing. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 6(2): 554-69 [DOI: 10.1109/jstars.2013.2242255http://dx.doi.org/10.1109/jstars.2013.2242255]
Wang Q, Mao Z D, Wang B and Guo L. 2017. Knowledge Graph Embedding: A Survey of Approaches and Applications. IEEE Transactions on Knowledge and Data Engineering, 29 (12): 2724-43 [DOI: 10.1109/tkde.2017.2754499http://dx.doi.org/10.1109/tkde.2017.2754499]
Wang Z H, Yang X M and Zhou C H. 2021. Geographic knowledge graph for remote sensing big data. Journal of Geo-information Science, 23(1): 16-28
王志华,杨晓梅,周成虎.2021.面向遥感大数据的地学知识图谱构想.地球信息科学, 23(1): 16-28 [DOI: 10.12082/dqxxkx.2021.200632http://dx.doi.org/10.12082/dqxxkx.2021.200632]
Winter M E.1999. N-FINDR: an algorithm for fast autonomous spectral end-member determination in hyperspectral data. //Proceedings of SPIE,3753, Imaging Spectrometry V, 266. Denver, CO, USA: SPIE:266–75 [DOI: 10.1117/12.366289http://dx.doi.org/10.1117/12.366289]
Xiang M T, Wei W and Wu W B. 2018. Review of vegetation phenology estimation by using remote sensing. China Agricultural Information,30(1): 55-66
项铭涛, 卫炜, 吴文斌. 2018. 植被物候参数遥感提取研究进展评述. 中国农业信息, 30(1): 55-66 [DOI:10.12105/j.issn.1672-0423.20180106http://dx.doi.org/10.12105/j.issn.1672-0423.20180106]
Xiong F C, Zhou J, Ye M C, Lu J F and Qian Y T.2021.NMF-SAE:
An Interpretable sparse autoencoder for hyperspectral unmixing.IEEE International Conference on Acoustics,Speech and Signal Processing(ICASSP 2021), pp, 1865-1869[DOI:10.1109/ICASSP39728.2021.9414084http://dx.doi.org/10.1109/ICASSP39728.2021.9414084]
Xu N, Xiao X Y, Geng X R, You H J and Cao Y H. 2016. Spectral-spatial constrained sparse unmixing of hyperspectral imagery using a hybrid spectral library. Remote Sensing Letters, 7(7): 641-50 [DOI: 10.1080/2150704x.2016.1177240http://dx.doi.org/10.1080/2150704x.2016.1177240]
Xue J, Anderson MC, Gao F, Hain C, Sun L, Yang Yet al.2020.Sharpening ECOSTRESS and VIIRS land surface temperature using harmonized Landsat-Sentinel surface reflectances. Remote Sensing of Environment, 215: 112055 [DOI:10.1016/j.rse.2020.112055http://dx.doi.org/10.1016/j.rse.2020.112055]
Yuan B. 2018. NMF hyperspectral unmixing algorithm combined with spatial and spectral correlation analysis. National Remote Sensing Bulletin, 22(2): 265-76
袁博.2018.空间与谱间相关性分析的NMF高光谱解混. 遥感学报, 22 (2): 265-76 [DOI: 10.11834/jrs.20186445http://dx.doi.org/10.11834/jrs.20186445]
Zhang J X, Gu H Y, Yang Y, Zhang H and Li H T. 2021. Research progress and trend of high-resolution remote sensing imagery intelligent interpretation. National Remote Sensing Bulletin, 25 (11): 2198-210
张继贤, 顾海燕,杨懿,张鹤,李海涛. 2021.高分辨率遥感影像智能解译研究进展与趋势. 遥感学报, 25(11): 2198-210 [DOI: 10.11834/jrs.20210382http://dx.doi.org/10.11834/jrs.20210382]
Zhang L P and Li J Y.2016. Development and prospect of sparse representation-based hyperspectral image processing and analysis. National Remote Sensing Bulletin, 20(5): 1091-101
张良培, 李家艺.2016.高光谱图像稀疏信息处理综述与展望.遥感学报, 20(5): 1091-101 [DOI: 10.11834/jrs.20166050http://dx.doi.org/10.11834/jrs.20166050]
Zhang S Q, Li J, Li H C, Deng C Z and Plaza A. 2018. Spectral-Spatial Weighted Sparse Regression for Hyperspectral Image Unmixing. IEEE Transactions on Geoscience and Remote Sensing, 56(6): 3265-3276 [DOI:10.1109/TGRS.2018.2797200http://dx.doi.org/10.1109/TGRS.2018.2797200]
Zhou C H, Wang C H, Hou Z Q, Zheng Z , Shen S Z, Cheng Q Met al. 2021. Geoscience knowledge graph in the big data era.Scientia Sinice Terrae, 51(7): 1070-9
周成虎, 王华, 王成善, 侯增谦, 郑志明, 沈树忠, 成秋明等.2021.大数据时代的地学知识图谱研究.中国科学:地球科学, 51(7): 1070-9 [DOI: 10.1360/SSTe-2020-0337http://dx.doi.org/10.1360/SSTe-2020-0337]
Zhou J X, Qiu Y, Chen J and Chen X H. 2021.A geometric misregistration resistant data fusion approach for adding red-edge (RE) and short-wave infrared (SWIR) bands to high spatial resolution imagery,Science of Remote Sensing, 4: 100033 [DOI:10.1016/j.srs.2021.100033http://dx.doi.org/10.1016/j.srs.2021.100033]
Zhu Q Q, Lei Y, Sun X L, Guan Q F, Zhong Y F, Zhang L Pet al.2022.Knowledge-guided land pattern depiction for urban land use mapping: A case study of Chinese cities. Remote Sensing of Environment, 272,112916 [DOI:10.1016/j.rse.2022.112916http://dx.doi.org/10.1016/j.rse.2022.112916]
Zhu Q Q,Wang L L,Chen J L,Zeng W,Zhong Y F,Guan Q Fet al.2022.S3TRM: Spectral-Spatial Unmixing of Hyperspectral Imagery Based on Sparse Topic Relaxation-Clustering Model. IEEE Transactions on Geoscience and Remote Sensing, 60: 1-13601-13 [DOI: 10.1109/tgrs.2021.3117250http://dx.doi.org/10.1109/tgrs.2021.3117250]
相关作者
相关机构