浏览全部资源
扫码关注微信
1.哈尔滨理工大学 测控技术与通信工程学院 黑龙江省激光光谱技术及应用重点实验室, 黑龙江 哈尔滨 150080
2.国网黑龙江省电力有限公司 综合信息中心,黑龙江 哈尔滨 150010
3.中部大学 计算机科学学院,日本 爱知 487-8501
[ "王爱丽(1979-),女,天津人,博士,副教授,硕士生导师,2008年于哈尔滨工业大学获得博士学位,主要从事机器视觉、深度学习图像分类的研究。Email:aili925@hrbust.edu.cn" ]
[ "吴海滨(1977-),男,上海人,博士,教授,博士生导师,2002年于哈尔滨工业大学获得硕士学位,2008年于哈尔滨理工大学获得博士学位,主要从事机器视觉、医学虚拟现实、深度学习图像分类的研究。E-mail:woo@hrbust.edu.cn" ]
收稿日期:2022-09-27,
修回日期:2022-11-08,
纸质出版日期:2023-07-10
移动端阅览
王爱丽,丁姗姗,刘和等.空谱域适应与XGBoost结合的跨场景高光谱图像分类[J].光学精密工程,2023,31(13):1950-1961.
WANG Aili,DING Shanshan,LIU He,et al.Cross-scene hyperspectral image classification combined spatial-spectral domain adaptation with XGBoost[J].Optics and Precision Engineering,2023,31(13):1950-1961.
王爱丽,丁姗姗,刘和等.空谱域适应与XGBoost结合的跨场景高光谱图像分类[J].光学精密工程,2023,31(13):1950-1961. DOI: 10.37188/OPE.20233113.1950.
WANG Aili,DING Shanshan,LIU He,et al.Cross-scene hyperspectral image classification combined spatial-spectral domain adaptation with XGBoost[J].Optics and Precision Engineering,2023,31(13):1950-1961. DOI: 10.37188/OPE.20233113.1950.
针对跨场景高光谱遥感图像分类中源域和目标域的频谱偏移问题,提出一种结合空谱域适应与极度梯度提升树(eXtreme Gradient Boosting, XGBoost)的跨场景高光谱图像分类模型。将深度超参数卷积模型(Depthwise Over-parameterized Convolution Model,DOCM)和大核注意力(Large Kernel Attention,LKA)结合,构成空谱注意力模型,提取源域空谱特征。利用相同的空谱注意力模型对目标域进行特征提取,并与鉴别器完成对抗域适应,减少源域与目标域之间的频谱偏移;通过目标域中少量有标签数据对目标域特征提取器进行有监督域适应,使目标域特征提取器进一步学习目标域的真实分布,并对源域和目标域的特征进行映射,形成相似的空间分布,完成聚类域适应。最后,使用集成分类器XGBoost进行高光谱图像分类,进一步提高模型的训练速度与置信度。在Pavia和Indiana高光谱数据集上的实验结果表明,本文算法的总体分类精度分别达到了91.62%和 65.98%。相比较于其他跨场景高光谱图像分类模型,本文所提模型具有更高的地物分类精度。
For solving the problem of spectral shift between the source domain and target domain in cross-scene hyperspectral remote sensing image classification, this study proposes a cross-scene hyperspectral image classification model combining spatial-spectral domain adaptation and Xtreme Gradient Boosting (XGBoost). First, the Depth Over Parametric Convolution Model (DOCM) and Large Kernel Attention (LKA) was combined to form a spatial-spectral attention model and extract the spatial-spectral features of the source domain. Next, the same spatialspectral attention model was used to extract features from the target domain, and the discriminator was used to adapt to the confrontation domain to reduce the spectral shift between the source and target domains. Second, the feature extractor of the target domain was adapted to the supervised domain through a small amount of labeled data in the target domain such that the feature extractor of the target domain can learn the true distribution of the target domain and map the features of the source and target domains to form a similar spatial distribution and complete the clustering domain adaptation. Finally, the ensemble classifier XGBoost was used to classify hyperspectral images to further improve the training speed and confidence of the model. Experimental results for the Pavia and Indiana hyperspectral datasets indicate that the overall classification accuracy of this algorithm reaches 91.62% and 65.98%, respectively. Compared with other cross-scene hyperspectral image classification models, the proposed model has a higher classification accuracy.
黄鸿 , 张臻 , 李政英 . 面向高光谱影像分类的深度流形重构置信网络 [J]. 光学 精密工程 , 2021 , 29 ( 8 ): 1985 - 1998 . doi: 10.37188/OPE.20212908.1985 http://dx.doi.org/10.37188/OPE.20212908.1985
HUANG H , ZHANG ZH , LI ZH Y . Deep manifold reconstruction belief network for hyperspectral remote sensing image classification [J]. Opt. Precision Eng. , 2021 , 29 ( 8 ): 1985 - 1998 . (in Chinese) . doi: 10.37188/OPE.20212908.1985 http://dx.doi.org/10.37188/OPE.20212908.1985
樊星皓 , 刘春雨 , 金光 , 等 . 轻小型高分辨率星载高光谱成像光谱仪 [J]. 光学 精密工程 , 2021 , 29 ( 3 ): 463 - 473 . doi: 10.37188/ope.20212903.0463 http://dx.doi.org/10.37188/ope.20212903.0463
FAN X H , LIU CH Y , JIN G , et al . Small and high-resolution spaceborne hyperspectral imaging spectrometer [J]. Opt. Precision Eng. , 2021 , 29 ( 3 ): 463 - 473 . (in Chinese) . doi: 10.37188/ope.20212903.0463 http://dx.doi.org/10.37188/ope.20212903.0463
QIN H M , ZHOU W Q , YAO Y , et al . Estimating aboveground carbon stock at the scale of individual trees in subtropical forests using UAV LiDAR and hyperspectral data [J]. Remote Sensing , 2021 , 13 ( 24 ): 4969 . doi: 10.3390/rs13244969 http://dx.doi.org/10.3390/rs13244969
CRUZ-RAMOS C , GARCIA-SALGADO B P , REYES-REYES R , et al . Gabor features extraction and land-cover classification of urban hyperspectral images for remote sensing applications [J]. Remote Sensing , 2021 , 13 ( 15 ): 2914 . doi: 10.3390/rs13152914 http://dx.doi.org/10.3390/rs13152914
LIU Q C , XIAO L , YANG J X , et al . Content-guided convolutional neural network for hyperspectral image classification [J]. IEEE Transactions on Geoscience and Remote Sensing , 2020 , 58 ( 9 ): 6124 - 6137 . doi: 10.1109/tgrs.2020.2974134 http://dx.doi.org/10.1109/tgrs.2020.2974134
BINTA ZAKARIA Z , ISLAM M R . Hybrid 3DNet: hyperspectral image classification with spectral-spatial dimension reduction using 3D CNN [J]. International Journal of Computer Applications , 2022 , 184 ( 23 ): 6 - 11 . doi: 10.5120/ijca2022922270 http://dx.doi.org/10.5120/ijca2022922270
JIA S , JIANG S G , ZHANG S Y , et al . Graph-in-graph convolutional network for hyperspectral image classification [J]. IEEE Transactions on Neural Networks and Learning Systems , 2022 , PP. DOI: 10.1109/TNNLS.2022.3182715 http://dx.doi.org/10.1109/TNNLS.2022.3182715 .
刘桂雄 , 黄坚 . 基于标签预留Softmax算法的机器视觉检测鉴别语义分割迁移学习技术 [J]. 光学 精密工程 , 2022 , 30 ( 1 ): 117 - 125 . doi: 10.37188/OPE.20223001.0117 http://dx.doi.org/10.37188/OPE.20223001.0117
LIU G X , HUANG J . Transfer learning techniques for semantic segmentation of machine vision inspection and identification based on label-reserved Softmax algorithms [J]. Opt. Precision Eng. , 2022 , 30 ( 1 ): 117 - 125 . (in Chinese) . doi: 10.37188/OPE.20223001.0117 http://dx.doi.org/10.37188/OPE.20223001.0117
HUANG Y , PENG J T , NING Y J , et al . Graph embedding and distribution alignment for domain adaptation in hyperspectral image classification [J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing , 2021 , 14 : 7654 - 7666 . doi: 10.1109/jstars.2021.3099805 http://dx.doi.org/10.1109/jstars.2021.3099805
CHEN J F , CHEN G , FANG B , et al . Class-aware domain adaptation for coastal land cover mapping using optical remote sensing imagery [J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing , 2021 , 14 : 11800 - 11813 . doi: 10.1109/jstars.2021.3128527 http://dx.doi.org/10.1109/jstars.2021.3128527
DENG B , JIA S , SHI D M . Deep metric learning-based feature embedding for hyperspectral image classification [J]. IEEE Transactions on Geoscience and Remote Sensing , 2020 , 58 ( 2 ): 1422 - 1435 . doi: 10.1109/tgrs.2019.2946318 http://dx.doi.org/10.1109/tgrs.2019.2946318
赵春晖 , 李彤 , 冯收 . 基于密集卷积和域自适应的高光谱图像分类 [J]. 光子学报 , 2021 , 50 ( 3 ): 0310001 . doi: 10.3788/gzxb20215003.0310001 http://dx.doi.org/10.3788/gzxb20215003.0310001
ZHAO CH H , LI T , FENG SH . Hyperspectral image classification based on dense convolution and domain adaptation [J]. Acta Photonica Sinica , 2021 , 50 ( 3 ): 0310001 . (in Chinese) . doi: 10.3788/gzxb20215003.0310001 http://dx.doi.org/10.3788/gzxb20215003.0310001
周钢 , 郭福亮 . 集成学习方法研究 [J]. 计算技术与自动化 , 2018 , 37 ( 4 ): 148 - 153 . doi: 10.16339/j.cnki.jsjsyzdh.201804026 http://dx.doi.org/10.16339/j.cnki.jsjsyzdh.201804026
ZHOU G , GUO F L . Research on ensemble learning [J]. Computing Technology and Automation , 2018 , 37 ( 4 ): 148 - 153 . (in Chinese) . doi: 10.16339/j.cnki.jsjsyzdh.201804026 http://dx.doi.org/10.16339/j.cnki.jsjsyzdh.201804026
陈伟民 , 张凌 , 宋冬梅 , 等 . 基于AdaBoost改进随机森林的高光谱图像地物分类方法研究 [J]. 遥感技术与应用 , 2018 , 33 ( 4 ): 612 - 620 . doi: 10.11873/j.issn.1004-0323.2018.4.0612 http://dx.doi.org/10.11873/j.issn.1004-0323.2018.4.0612
CHEN W M , ZHANG L , SONG D M , et al . Research on hyperspectral imagery land cover classification method based on AdaBoost improved random forest [J]. Remote Sensing Technology and Application , 2018 , 33 ( 4 ): 612 - 620 . (in Chinese) . doi: 10.11873/j.issn.1004-0323.2018.4.0612 http://dx.doi.org/10.11873/j.issn.1004-0323.2018.4.0612
李润祥 , 高小红 , 汤敏 . 基于双树复小波分解的Boosting集成学习土地覆被分类研究 [J]. 遥感技术与应用 , 2022 , 37 ( 2 ): 354 - 356, 358 .
LI R X , GAO X H , TANG M . Study on boosting ensemble learning land cover classification based on dual-tree complex wavelet transform [J]. Remote Sensing Technology and Application , 2022 , 37 ( 2 ): 354 - 356, 358 . (in Chinese)
CHEN T , GUESTRIN C . XGBoost : a scalable tree boosting system [EB/OL]. 2016 : arXiv : 1603 . 02754 . https://arxiv.org/abs/1603.02754 https://arxiv.org/abs/1603.02754 . doi: 10.1145/2939672.2939785 http://dx.doi.org/10.1145/2939672.2939785
DU W , CHEN H , LIAO P , et al . Visual attention network for low dose CT [EB/OL]. 2018 : arXiv : 1810 . 13059 . https://arxiv.org/abs/1810.13059 https://arxiv.org/abs/1810.13059 . doi: 10.1109/lsp.2019.2922851 http://dx.doi.org/10.1109/lsp.2019.2922851
CAO J M , LI Y Y , SUN M C , et al . DO-conv: depthwise over-parameterized convolutional layer [J]. IEEE Transactions on Image Processing: a Publication of the IEEE Signal Processing Society , 2022 , 31 : 3726 - 3736 . doi: 10.1109/tip.2022.3175432 http://dx.doi.org/10.1109/tip.2022.3175432
PLAZA A , BENEDIKTSSON J A , BOARDMAN J W , et al . Recent advances in techniques for hyperspectral image processing [J]. Remote Sensing of Environment , 2009 , 113 : S110 - S122 . doi: 10.1016/j.rse.2007.07.028 http://dx.doi.org/10.1016/j.rse.2007.07.028
LI P , HU H , CHENG T , et al . High-resolution multispectral image classification over urban areas by image segmentation and extended morphological profile [C]. 2006 IEEE International Symposium on Geoscience and Remote Sensing. July 31 - August 4 , 2006 , Denver, CO, USA. IEEE , 2007 : 3252 - 3254 . doi: 10.1109/igarss.2006.835 http://dx.doi.org/10.1109/igarss.2006.835
CHEN Y S , WANG Y , GU Y F , et al . Deep learning ensemble for hyperspectral image classification [J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing , 2019 , 12 ( 6 ): 1882 - 1897 . doi: 10.1109/jstars.2019.2915259 http://dx.doi.org/10.1109/jstars.2019.2915259
WANG A L , LIU C Y , XUE D , et al . Hyperspectral image classification based on cross-scene adaptive learning [J]. Symmetry , 2021 , 13 ( 10 ): 1878 - 1894 . doi: 10.3390/sym13101878 http://dx.doi.org/10.3390/sym13101878
0
浏览量
432
下载量
2
CSCD
关联资源
相关文章
相关作者
相关机构