1.湖北珞珈实验室,湖北 武汉,430079
2.武汉大学资源与环境科学学院,湖北 武汉,430079
3.中国地质大学(武汉)地球物理与空间信息学院,湖北 武汉,430074
4.武汉大学计算机学院,湖北 武汉,430072
董燕妮,博士,教授,研究方向为遥感图像智能解译。dongyanni@whu.edu.cn
杜博,博士,教授。dubo@whu.edu.cn
收稿:2025-05-14,
纸质出版:2026-06-05
移动端阅览
董燕妮,秦学翔,张玉香,等.多级空间特征的平衡测度学习遥感图像域适应分类[J].武汉大学学报(信息科学版),2026,51(6):1182-1193.
DONG Yanni,QIN Xuexiang,ZHANG Yuxiang,et al.Balanced Metric Learning in Multilevel Spatial Features for Domain Adaptation in Remote Sensing Image Classification[J].Geomatics and Information Science of Wuhan University,2026,51(6):1182-1193.
董燕妮,秦学翔,张玉香,等.多级空间特征的平衡测度学习遥感图像域适应分类[J].武汉大学学报(信息科学版),2026,51(6):1182-1193. DOI:10.13203/j.whugis20240212
DONG Yanni,QIN Xuexiang,ZHANG Yuxiang,et al.Balanced Metric Learning in Multilevel Spatial Features for Domain Adaptation in Remote Sensing Image Classification[J].Geomatics and Information Science of Wuhan University,2026,51(6):1182-1193. DOI:10.13203/j.whugis20240212
遥感图像域适应分类通常以光谱特征或简单的空间特征作为数据特征,通过对齐域间的特征分布来标注缺乏标签样本的域。域适应方法忽略了遥感图像更深层的空间特征,导致对空间局部信息的发掘不够;同时,域适应方法很少考虑类别非均衡的问题,导致小类样本的重要性被忽略。针对上述问题,提出一种多级空间特征的平衡测度学习遥感图像域适应分类方法。首先,通过迭代均值滤波器提取遥感图像更深层的空间特征;其次,构建测度学习来对齐边缘分布,最小化类内距离和类间散度;最后,根据类的先验概率构建自适应权重,最大化平衡类间距离,以缓解类别不平衡的问题。所提方法在Worldview-2数据集和Pavia数据集上设计了4个任务来评估有效性,使用遥感图像分类任务中常用的总体精度、Kappa系数、可视化结果和时间成本作为评价指标,通过实验分析不同参数对分类结果的影响以及各个组成部分对分类结果的影响。相较于次优的算法,所提方法的总体准确率提高了1.64%~9.18%,Kappa系数提高了0.023 1~0.171 7。实验结果表明,所提方法不仅通过多级空间特征加强了局部特征的提取和有效利用,还有效缓解了类不平衡问题,使分类器取得了更好的预测效果。
Objective
2
Domain adaptation classification of remote sensing images usually takes spectral features or simple spatial features as data features
and labels domains lacking labeled samples by aligning feature distributions among domains. The domain adaptation method ignores the deeper spatial features of remote sensing images
resulting in insufficient exploitation of spatially localized information. At the same time
the domain adaptation method seldom takes into account the problem of class imbalance
which leads to the importance of the small class of samples being ignored.
Methods
2
To address the above problems
balanced metric learning in multilevel spatial features for domain adaptation in remote sensing image classification is proposed. First
deeper spatial features of remote sensing images are extracted by an iterative mean filter. Then
metric learning is constructed to align the marginal distributions and minimize the intraclass distance and interclass scatter. Finally
adaptive weights are constructed based on the prior probability of the classes to maximize the balanced interclass distances to alleviate the problem of class imbalance.
Results
2
The proposed method designs four tasks on the Worldview-2 dataset and Pavia dataset to evaluate the effectiveness. The overall accuracy
Kappa coefficient
visualization results
and time cost
which are commonly used in remote sensing image classification tasks
are used as the evaluation indicators. A series of experiments are carried out to analyze the effects of different parameters on the classification results. The effects of the individual components on the classification results are analyzed. Compared with the suboptimal algorithm
the proposed method improves the overall accuracy by 1.64% to 9.18%
and the Kappa coefficient by 0.023 1 to 0.171 7.
Conclusions
2
The experimental results show that the proposed method not only enhances the extraction and effective utilization of local features through multilevel spatial features
but also alleviates the problem of class imbalance effectively so that the classifier achieves a better prediction result.
徐恩恩 , 郭颖 , 陈尔学 , 等 . 基于无人机LiDAR和高空间分辨率卫星遥感数据的区域森林郁闭度估测模型 [J]. 武汉大学学报(信息科学版) , 2022 , 47 ( 8 ): 1298 - 1308 .
XU En'en , GUO Ying , CHEN Erxue , et al . An Estimation Model for Regional Forest Canopy Closure Combined with UAV LiDAR and High Spatial Resolution Satellite Remote Sensing Data [J]. Geomatics and Information Science of Wuhan University , 2022 , 47 ( 8 ): 1298 - 1308 .
高奎亮 , 刘冰 , 余旭初 , 等 . 面向高光谱影像分类的网络结构自动搜索方法 [J]. 武汉大学学报(信息科学版) , 2024 , 49 ( 2 ): 225 - 235 .
GAO Kuiliang , LIU Bing , YU Xuchu , et al . Automatic Network Structure Search Method for Hyperspectral Image Classification [J]. Geomatics and Information Science of Wuhan University , 2024 , 49 ( 2 ): 225 - 235 .
YU J , ZENG P , YU Y Y , et al . A Combined Convolutional Neural Network for Urban Land-Use Classification with GIS Data [J]. Remote Sensing , 2022 , 14 ( 5 ): 1128 .
LU Y W , WONG W K , ZENG B Q , et al . Guided Discrimination and Correlation Subspace Learning for Domain Adaptation [J]. IEEE Transactions on Image Processing , 2023 , 32 : 2017 - 2032 .
梁敏 , 汪西莉 . 结合超分辨率和域适应的遥感图像语义分割方法 [J]. 计算机学报 , 2022 , 45 ( 12 ): 2619 - 2636 .
LIANG Min , WANG Xili . Semantic Segmentation Model for Remote Sensing Images Combining Super Resolution and Domain Adaption [J]. Chinese Journal of Computers , 2022 , 45 ( 12 ): 2619 - 2636 .
ZHOU Y S , CHEN P Y , LIU N , et al . Graph-Embedding Balanced Transfer Subspace Learning for Hyperspectral Cross-Scene Classification [J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing , 2022 , 15 : 2944 - 2955 .
PAN S J , TSANG I W , KWOK J T , et al . Domain Adaptation via Transfer Component Analysis [J]. IEEE Transactions on Neural Networks , 2011 , 22 ( 2 ): 199 - 210 .
LONG M S , WANG J M , DING G G , et al . Transfer Feature Learning with Joint Distribution Adaptation [C]// IEEE International Conference on Computer Vision , Sydney, Australia , 2014 .
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 .
DONG Y N , LIANG T Y , ZHANG Y X , et al . Spectral–Spatial Weighted Kernel Manifold Embedded Distribution Alignment for Remote Sensing Image Classification [J]. IEEE Transactions on Cybernetics , 2021 , 51 ( 6 ): 3185 - 3197 .
吴田军 , 骆剑承 , 夏列钢 , 等 . 迁移学习支持下的遥感影像对象级分类样本自动选择方法 [J]. 测绘学报 , 2014 , 43 ( 9 ): 908 - 916 .
WU Tianjun , LUO Jiancheng , XIA Liegang , et al . An Automatic Sample Collection Method for Object-Oriented Classification of Remotely Sensed Imageries Based on Transfer Learning [J]. Acta Geodaetica et Cartographica Sinica , 2014 , 43 ( 9 ): 908 - 916 .
许夙晖 , 慕晓冬 , 张雄美 , 等 . 结合对抗网络与辅助任务的遥感影像无监督域适应方法 [J]. 测绘学报 , 2017 , 46 ( 12 ): 1969 - 1977 .
XU Suhui , MU Xiaodong , ZHANG Xiongmei , et al . Unsupervised Remote Sensing Domain Adaptation Method with Adversarial Network and Auxiliary Task [J]. Acta Geodaetica et Cartographica Sinica , 2017 , 46 ( 12 ): 1969 - 1977 .
罗芳 , 艾廷华 , 贾小斌 . 空间自相关支撑下的地类分布模式一致性评价 [J]. 武汉大学学报(信息科学版) , 2022 , 47 ( 7 ): 1017 - 1024 .
LUO Fang , AI Tinghua , JIA Xiaobin . Consistency Evaluation of Land Use Distribution Pattern Supported by Spatial Autocorrelation [J]. Geomatics and Information Science of Wuhan University , 2022 , 47 ( 7 ): 1017 - 1024 .
TOBLER W R . A Computer Movie Simulating Urban Growth in the Detroit Region [J]. Economic Geography , 1970 , 46 ( 1 ): 234 - 240 .
GONG B Q , SHI Y , SHA F , et al . Geodesic Flow Kernel for Unsupervised Domain Adaptation [C]// IEEE Conference on Computer Vision and Pattern Recognition , Providence, USA , 2012 .
DONG Y N , LIANG T Y , YANG C , et al . Joint Distance Transfer Metric Learning for Remote-Sensing Image Classification [J]. IEEE Geoscience and Remote Sensing Letters , 2022 , 19 : 6506205 .
HUANG J C , ZHOU Z H . Transfer Metric Learning for Unsupervised Domain Adaptation [J]. IET Image Processing , 2019 , 13 ( 5 ): 804 - 810 .
WANG J D , CHEN Y Q , HAO S J , et al . Balanced Distribution Adaptation for Transfer Learning [C]// IEEE International Conference on Data Mining (ICDM) , New Orleans, USA , 2017 .
BELKIN M , NIYOGI P , SINDHWANI V . Manifold Regularization: A Geometric Framework for Learning from Labeled and Unlabeled Examples [J]. Journal of Machine Learning Research , 2006 , 7 : 2399 - 2434 .
孙一帆 , 余旭初 , 谭熊 , 等 . 面向小样本高光谱影像分类的轻量化关系网络 [J]. 武汉大学学报(信息科学版) , 2022 , 47 ( 8 ): 1336 - 1348 .
SUN Yifan , YU Xuchu , TAN Xiong , et al . Lightweight Relational Network for Small Sample Hyperspectral Image Classification [J]. Geomatics and Information Science of Wuhan University , 2022 , 47 ( 8 ): 1336 - 1348 .
李晓旭 , 刘忠源 , 武继杰 , 等 . 小样本图像分类的注意力全关系网络 [J]. 计算机学报 , 2023 , 46 ( 2 ): 371 - 384 .
LI Xiaoxu , LIU Zhongyuan , WU Jijie , et al . Total Relation Network with Attention for Few-Shot Image Classification [J]. Chinese Journal of Computers , 2023 , 46 ( 2 ): 371 - 384 .
毛丽君 , 李明诗 . GEE环境下联合Sentinel主被动遥感数据的国家公园土地覆盖分类 [J]. 武汉大学学报(信息科学版) , 2023 , 48 ( 5 ): 756 - 764 .
MAO Lijun , LI Mingshi . Integrating Sentinel Active and Passive Remote Sensing Data to Land Cover Classification in a National Park from GEE Platform [J]. Geomatics and Information Science of Wuhan University , 2023 , 48 ( 5 ): 756 - 764 .
王严 , 刘万军 , 谭亚丽 , 等 . 卷积神经网络滑窗识别结合语义推理的城市功能区分类 [J]. 武汉大学学报(信息科学版) , 2023 , 48 ( 6 ): 950 - 959 .
WANG Yan , LIU Wanjun , TAN Yali , et al . Classification of Urban Functional Areas by Convolution Neural Network Recognition Combined with Sliding Window and Semantic Reasoning [J]. Geomatics and Information Science of Wuhan University , 2023 , 48 ( 6 ): 950 - 959 .
李毅夫 , 孙斌 , 高志海 , 等 . 5 m光学02星多光谱影像农田防护林信息提取 [J]. 遥感学报 , 2024 , 28 ( 3 ): 624 - 634 .
LI Yifu , SUN Bin , GAO Zhihai , et al . Farmland Shelterbelt Information Extraction Based on Multispectral Image of the ZY1-02E Satellite [J]. National Remote Sensing Bulletin , 2024 , 28 ( 3 ): 624 - 634 .
刘倩 , 吴泽彬 , 徐洋 , 等 . 基于多方向自适应感知网络的高光谱遥感图像分类 [J]. 遥感学报 , 2024 , 28 ( 1 ): 168 - 186 .
LIU Qian , WU Zebin , XU Yang , et al . Hyperspectral Remote Sensing Image Classification Based on Multidirectional Adaptive Aware Network [J]. National Remote Sensing Bulletin , 2024 , 28 ( 1 ): 168 - 186 .
汤圣君 , 张韵婕 , 李晓明 , 等 . 超体素随机森林与LSTM神经网络联合优化的室内点云高精度分类方法 [J]. 武汉大学学报(信息科学版) , 2023 , 48 ( 4 ): 525 - 533 .
TANG Shengjun , ZHANG Yunjie , LI Xiaoming , et al . A High-Precision Indoor Point Cloud Classification Method Jointly Optimized by Super Voxel Random Forest and LSTM Neural Network [J]. Geomatics and Information Science of Wuhan University , 2023 , 48 ( 4 ): 525 - 533 .
SUN B C , SAENKO K . Deep CORAL: Correlation Alignment for Deep Domain Adaptation [C]// Computer Vision–ECCV 2016 Workshops , Amsterdam, The Netherlands , 2016 .
WANG J D , FENG W J , CHEN Y Q , et al . Visual Domain Adaptation with Manifold Embedded Distribution Alignment [C]// The 26th ACM International Conference on Multimedia , Seoul Republic of Korea , 2018 .
0
浏览量
1
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