面向领域自适应的部分最优传输高光谱图像分类  

Partial optimal transport-based domain adaptation for hyperspectral image classification

作　　者：王碧琳[1] 王生生[1] 张哲 WANG Bilin;WANG Shengsheng;ZHANG Zhe(College of Computer Science and Technology,Jilin University,Changchun 130012,China;Suzhou Institute of Biomedical Engineering and Technology,Chinese Academy of Sciences,Suzhou 215163,China)

机构地区：[1]吉林大学计算机科学与技术学院,吉林长春130012 [2]中国科学院苏州生物医学工程技术研究所,江苏苏州215163

出　　处：《光学精密工程》2023年第17期2555-2563,共9页Optics and Precision Engineering

基　　金：国家重点研发计划资助项目(No.2020YFA0714103);吉林省发展和改革委员会创新能力建设项目(No.2021FGWCXNLJSSZ10,No.2019C053-3)。

摘　　要：针对高光谱遥感图像分类时有标注的源域训练数据与无标注目标域数据分布不一致的问题,提出基于部分最优传输的无监督领域自适应方法,实现对处于不同数据分布的高光谱遥感地物像素级分类。利用深度卷积神经网络将样本映射到潜在高维空间,根据部分最优传输理论建立样本传输方案,最小化域间分布差异,构建适配模型。采用类感知采样技术和质量分数因子自适应调整策略,促进域间类别对齐,建立全局最优传输。在两组公开高光谱遥感图像数据集上进行实验,从总体分类精度OA(%)、类别平均分类精度AA(%)、分类一致性检验Kappa(×100)等3个评价指标对像素分类结果量化比较。实验结果显示,在两组迁移任务上,相较于仅使用源域数据的基线模型,总体分类准确率分别提升2.21%和2.75%,相较于原始最优传输策略提升1.71%和2.01%,表明模型能够有效提升高光谱遥感影像中像素级地物的分类精度。Hyperspectral image classification is a major task in remote sensing data processing.To solve the problem of inconsistent distribution of labeled source and unlabeled target domains,an unsupervised domain adaptive method based on partial optimal transport is proposed to achieve pixel-level classification of hyperspectral ground objects under different data distributions.Specifically,a deep convolution neural network is used to map the sample to the potential high-dimensional space,and the sample transportation scheme is established based on the partial optimal transport theory to minimize the distribution discrepancy between domains.Class-aware sampling and the mass factor adaptive adjustment strategy are used to pro⁃mote the class alignment between domains and establish a global optimal transport.Experiments were con⁃ducted on two open-source hyperspectral image datasets,and the classification accuracies were compared quantitatively from the three evaluation matrices of overall accuracy(OA,%),average accuracy(AA,%),and Kappa(×100).Compared with the source-only method,the improved classification ac⁃curacies with the proposed method for OA and AA were 2.21%and 2.75%,respectively,and compared with the original optimal transport,the improved accuracies were 1.71%and 2.01%,respectively.These results show that the proposed model can effectively improve pixel-level classification accuracy in hyperspectral images.

关 键 词：计算机视觉 神经网络 领域自适应 最优传输 高光谱图像 

分 类 号：P236[天文地球—摄影测量与遥感]