浏览全部资源
扫码关注微信
纸质出版日期： 2024-08-16 ，
移动端阅览
马晓瑞， 哈林， 谌敦斌， 梅亮， 王洪玉. 2024. 融合特征优化的跨数据集高光谱图像分类. 中国图象图形学报， 29(08):2175-2187
Ma Xiaorui， Ha Lin， Shen Dunbin， Mei Liang， Wang Hongyu. 2024. Cross-dataset hyperspectral image classification based on fusion feature optimization. Journal of Image and Graphics， 29(08):2175-2187
目的
2
高光谱图像分类可实现对地物目标的逐像素识别，是对地观测中的关键技术。由于采集环境变迁与成像设备差异等因素的影响，不同高光谱图像特征分布偏移，影响现有模型的跨数据集分类精度。针对此，提出了一种融合特征优化的无监督跨数据集高光谱图像分类方法。
方法
2
提出了基于奇异值抑制的特征均衡策略实现数据内独立优化，通过限制奇异值正则项兼顾特征的可迁移性与可鉴别性；提出了基于隐式增广的特征匹配策略实现数据间特征协同优化，引导源域特征逼近目标域提高模型的泛化性；设计了基于隐式鉴别器的对抗学习框架实现数据间特征类别级优化，提高了预测多样性，实现跨数据集分类。
结果
2
实验在Pavia数据集组和HyRANK数据集组上进行，与多种最新的跨数据高光谱图像分类方法进行了对比，在Pavia数据集组中，相比于性能第2的模型，总体精度、平均精度和
к
系数分别提高了1.75%、3.55%和2.17%；在HyRANK数据集组中，相比于性能第2的模型，总体精度、平均精度和
к
系数分别提高了6.58%、13.10%和7.96%。同时进行了消融实验，研究了各个模块对高光谱图像分类效果的影响。实验结果表明，每一模块在提高高光谱图像分类效果方面都是有效的。
结论
2
本文提出的融合特征优化的跨数据集高光谱图像分类方法可以在无监督的条件下显著提高跨数据集高光谱图像分类精度，提高分类预测的多样性，得到更好的分类效果。
Objective
2
Hyperspectral image is a 3D data cube of “space-spectrum integration”. Its high-resolution spectral information can realize fine-grained land cover identification， and its wide-coverage spatial information can complete accurate land cover mapping. Therefore， hyperspectral images with spatial information and spectral information are widely used in tasks related to land cover classification. Thus， hyperspectral image classification is an important research topic in remote sensing field and a key supporting technology for many Earth observation tasks， such as smart cities， precision agriculture， and modern national defense. In recent years， many classification methods for hyperspectral images have been proposed to mine spatial and spectral information based on different deep networks， and they have achieved unprecedented high-precision classification results. However， due to various factors such as the change in acquisition environment and the difference in imaging sensors， the feature distribution of different hyperspectral images is shifted， which leads to the difficulty of cross-dataset classification. For this reason， existing classification methods usually retrain the model to deal with new hyperspectral image dataset， which is label intensive and time consuming. In the era of remote sensing big data， developing classification methods for cross-dataset hyperspectral images is important. Therefore， this study investigates classification methods for cross-dataset hyperspectral images to achieve large-scale Earth observation missions.
Method
2
This study proposes an unsupervised classification method for cross-dataset hyperspectral images based on feature optimization. The proposed method consists of three main modules. First， a feature balancing strategy is proposed to optimize the intra-dataset features independently. During the adversarial domain adaptation process， the transferability and discriminability of features are contradictory， and most existing methods sacrifice the feature discriminability of target dataset， which results in blurred class boundaries and affects classification performance. In the proposed method， a regularization term with singular value of the feature vectors from the source and target datasets is minimized to enhance the transferability and discriminability of the learned features. By extracting better features， this method achieves more accurate classification results on the target dataset. Second， a feature matching strategy is proposed to optimize the inter-dataset features collaboratively. No labeled sample is available in the target dataset， and feature discrepancies are obvious between the source dataset and the target dataset. Thus， the model cannot accurately match the two datasets， which leads to inadequate generalization. In the proposed method， an implicit feature augmentation strategy is performed to guide the source features to the target space， which improves the generalization performance of the model. By utilizing the underlying relationships between different datasets， this method adapts better to the target dataset and improves the overall performance of the classification model. Finally， an adversarial learning framework based on implicit discriminator is designed to optimize inter-dataset class-level features. Existing adversarial learning methods often construct an additional discriminator or use a binary classifier as a discriminator. The former focuses only on feature confusion between datasets and ignores class-level information， and the latter considers only class-level differences， which leads to ambiguous predictions. In the proposed method， by reusing the task classifier as an implicit discriminator， inter-dataset alignment and cross-dataset class recognition are achieved. By further optimizing the adversarial learning method， this approach can further enhance the classification performance of the model on the target dataset.
Result
2
All experiments in this study are executed on a desktop computer with Intel Core i7 4.0 GHz CPU， GeForce GTX 1080Ti GPU， and 32 GB memory. PyTorch， which is a widely used deep learning framework， is used in the experiment. The experiments are conducted on Pavia and HyRANK datasets. The evaluation indexes include overall accuracy （OA）， average accuracy （AA）， and
к
coefficient. In addition， the classification results are intuitively represented by classification maps. The experimental results are compared with various recent classification methods for cross-dataset hyperspectral images trained with all labeled source samples and unlabeled target samples. The proposed method is op
timized by a small-batch SGD optimizer with a momentum of 0.9. The learning rates for the Pavia and HyRANK datasets are set to 0.000 1 and 0.001， respectively. The maximum number of iterations is set to 2 000. In the Pavia datasets， OA， AA， and
к
values increase by 1.75%， 3.55%， and 2.17%， respectively， compared with the model with the second performance. In the HyRANK datasets， OA， AA， and
к
values increase by 6.58%， 13.10%， and 7.96%， respectively， compared with the second-ranked model. Experimental results show that， compared with other methods， the classification maps produced by the proposed method are closer to the ground truth， and evaluation indexes of some categories are significantly improved. Moreover， the ablation experiment is conducted to study the effect of each module of the proposed method， which proves that each module is effective in improving the cross-dataset classification effect of hyperspectral images.
Conclusion
2
In this study， a feature-optimized classification method for cross-dataset hyperspectral images is proposed. The proposed method provides a novel solution for unsupervised classification of cross-dataset hyperspectral images. By combining feature equalization， feature matching， and adversarial learning techniques， the method improves the generalization capability and classification performance of the model. Thus， it is an effective approach for cross-dataset image classification tasks. The proposed method is verified on two hyperspectral datasets， and the experimental results show that the proposed method can significantly improve the accuracy of cross-dataset hyperspectral images under unsupervised conditions compared with related methods.
高光谱图像分类跨数据集分类特征优化领域自适应无监督分类域对抗网络
hyperspectral image classificationcross-dataset classificationfeature optimizationdomain adaptationunsupervised classificationdomain adversarial network
Belgiu M and Drăguţ L. 2016. Random forest in remote sensing： a review of applications and future directions. ISPRS Journal of Photogrammetry and Remote Sensing， 114： 24-31 ［DOI： 10.1016/j.isprsjprs.2016.01.011http://dx.doi.org/10.1016/j.isprsjprs.2016.01.011］
Chandler D. 1987. Introduction to Modern Statistical Mechanics. New York： Oxford University Press
Chen X Y， Wang S N， Long M S and Wang J M. 2019. Transferability vs. discriminability： batch spectral penalization for adversarial domain adaptation//Proceedings of the 36th International Conference on Machine Learning. Long Beach， USA： PMLR： 1081-1090
Chen Y， Nasrabadi N M and Tran T D. 2013. Hyperspectral image classification via kernel sparse representation. IEEE Transactions on Geoscience and Remote Sensing， 51（1）： 217-231 ［DOI： 10.1109/TGRS.2012.2201730http://dx.doi.org/10.1109/TGRS.2012.2201730］
Chen Y S， Lin Z H， Zhao X， Wang G and Gu Y F. 2014. Deep learning-based classification of hyperspectral data. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing， 7（6）： 2094-2107 ［DOI： 10.1109/JSTARS.2014.2329330http://dx.doi.org/10.1109/JSTARS.2014.2329330］
Chen Y S， Zhao X and Jia X P. 2015. Spectral-spatial classification of hyperspectral data based on deep belief network. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing， 8（6）： 2381-2392 ［DOI： 10.1109/JSTARS.2015.2388577http://dx.doi.org/10.1109/JSTARS.2015.2388577］
Cheng S Y， Xiong Y J， Yao Y and Li Q L. 2021. Hyperspectral image classification using an inverted feature pyramid network with U-Net. Journal of Image and Graphics， 26（8）： 1994-2008
程嵩阳， 熊玉洁， 姚瑶， 李庆利. 2021. 结合倒置特征金字塔和U-Net的高光谱图像分类. 中国图象图形学报， 26（8）： 1994-2008 ［DOI： 10.11834/jig.210194http://dx.doi.org/10.11834/jig.210194］
Cui S H， Wang S H， Zhuo J B， Li L， Huang Q M and Tian Q. 2021. Fast batch nuclear-norm maximization and minimization for robust domain adaptation ［EB/OL］. ［2023-08-21］. https://arxiv.org/pdf/2107.06154.pdfhttps://arxiv.org/pdf/2107.06154.pdf
Fang Z Q， Yang Y X， Li Z K， Li W， Chen Y S， Ma L and Du Q. 2022. Confident learning-based domain adaptation for hyperspectral image classification. IEEE Transactions on Geoscience and Remote Sensing， 60： 1-16 ［DOI： 10.1109/TGRS.2022.3166817http://dx.doi.org/10.1109/TGRS.2022.3166817］
Fauvel M， Tarabalka Y， Benediktsson J A， Chanussot J and Tilton J C. 2013. Advances in spectral-spatial classification of hyperspectral images. Proceedings of the IEEE， 101（3）： 652-675 ［DOI： 10.1109/JPROC.2012.2197589http://dx.doi.org/10.1109/JPROC.2012.2197589］
Golub G H and Reinsch C. 1970. Singular value decomposition and least squares solutions. Numerische Mathematik， 14（5）： 403-420 ［DOI： 10.1007/BF02163027http://dx.doi.org/10.1007/BF02163027］
He K M， Zhang X Y， Ren S Q and Sun J. 2016. Deep residual learning for image recognition//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas， USA： IEEE： 770-778 ［DOI： 10.1109/CVPR.2016.90http://dx.doi.org/10.1109/CVPR.2016.90］
Hua Y， Zhang T， Xi H W， Wang Y F and Huang X L. 2012. Research on method of hyperspectral remote sensing image classification based on decision tree. Computer Technology and Development， 22（6）： 198-202
华晔， 张涛， 奚后玮， 王玉斐， 黄秀丽. 2012. 基于决策树的高光谱遥感影像分类方法研究. 计算机技术与发展， 22（6）： 198-202 ［DOI： 10.3969/j.issn.1673-629X.2012.06.051http://dx.doi.org/10.3969/j.issn.1673-629X.2012.06.051］
Li D R， Tong Q X， Li R X， Gong J Y and Zhang L P. 2012. Current issues in high-resolution Earth observation technology. Science China Earth Sciences， 55（7）： 1043-1051
李德仁， 童庆禧， 李荣兴， 龚健雅， 张良培. 2012. 高分辨率对地观测的若干前沿科学问题. 中国科学： 地球科学）， 42（6）： 805-813 ［DOI： 10.1007/s11430-012-4445-9http://dx.doi.org/10.1007/s11430-012-4445-9］
Li J， Bioucas-Dias J M and Plaza A. 2010. Semisupervised hyperspectral image segmentation using multinomial logistic regression with active learning. IEEE Transactions on Geoscience and Remote Sensing， 48（11）： 4085-4098 ［DOI： 10.1109/TGRS.2010.2060550http://dx.doi.org/10.1109/TGRS.2010.2060550］
Li Z K， Tang X Y， Li W， Wang C Y， Liu C W and He J R. 2020. A two-stage deep domain adaptation method for hyperspectral image classification. Remote Sensing， 12（7）： #1054 ［DOI： 10.3390/rs12071054http://dx.doi.org/10.3390/rs12071054］
Liang J， Hu D P and Feng J S. 2020. Domain adaptation with auxiliary target domain-oriented classifier//Proceedings of 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Nashville， USA： IEEE： 16627-16637 ［DOI： 10.1109/CVPR46437.2021.01636http://dx.doi.org/10.1109/CVPR46437.2021.01636］
Liu Q C， Xiao L and Yang J X. 2021. Parallel implementation of content-guided deep convolutional network for hyperspectral image classification. Journal of Image and Graphics， 26（8）： 1926-1939
刘启超， 肖亮， 杨劲翔. 2021. 面向高光谱图像分类的内容引导卷积深度网络并行实现. 中国图象图形学报， 26（8）： 1926-1939 ［DOI： 10.11834/jig.200411http://dx.doi.org/10.11834/jig.200411］
Long M S， Cao Z J， Wang J M and Jordan M I. 2018. Conditional adversarial domain adaptation//Proceedings of the 32nd International Conference on Neural Information Processing Systems. Montréal， Canada： Curran Associates Inc.： 1647-1657
Ma X R， Mou X R， Wang J， Liu X K， Geng J and Wang H Y. 2021. Cross-dataset hyperspectral image classification based on adversarial domain adaptation. IEEE Transactions on Geoscience and Remote Sensing， 59（5）： 4179-4190 ［DOI： 10.1109/TGRS.2020.3015357http://dx.doi.org/10.1109/TGRS.2020.3015357］
Peng J T， Sun W W， Wei T H and Fan W Q. 2020. A modified correlation alignment algorithm for the domain adaptation of GF-5 hyperspectral image. Journal of Remote Sensing， 24（4）： 417-426
彭江涛， 孙伟伟， 魏天慧， 范文琦. 2020. 高分五号高光谱影像的关联对齐域适应与分类. 遥感学报， 24（4）： 417-426 ［DOI： 10.11834/jrs.20209212http://dx.doi.org/10.11834/jrs.20209212］
Qin Y， Bruzzone L and Li B. 2019. Tensor alignment based domain adaptation for hyperspectral image classification. IEEE Transactions on Geoscience and Remote Sensing， 57（11）： 9290-9307 ［DOI： 10.1109/TGRS.2019.2926069http://dx.doi.org/10.1109/TGRS.2019.2926069］
Saboori A and Ghassemian H. 2020. Robust transfer joint matching distributions in semi-supervised domain adaptation for hyperspectral images classification. International Journal of Remote Sensing， 41（23）： 9283-9307 ［DOI： 10.1080/01431161.2020.1797221http://dx.doi.org/10.1080/01431161.2020.1797221］
Saboori A and Ghassemian H. 2021. Adversarial discriminative active deep learning for domain adaptation in hyperspectral images classification. International Journal of Remote Sensing， 42（10）： 3981-4003 ［DOI： 10.1080/01431161.2021.1880663http://dx.doi.org/10.1080/01431161.2021.1880663］
Sun B C， Feng J S and Saenko K. 2016. Return of frustratingly easy domain adaptation//Proceedings of the 30th AAAI Conference on Artificial Intelligence. Phoenix， USA： AAAI Press： 2058-2065
Tang X B， Li C C and Peng Y X. 2022. Unsupervised joint adversarial domain adaptation for cross-scene hyperspectral image classification. IEEE Transactions on Geoscience and Remote Sensing， 60： #5536415 ［DOI： 10.1109/TGRS.2022.3202363http://dx.doi.org/10.1109/TGRS.2022.3202363］
Tarabalka Y， Fauvel M， Chanussot J and Benediktsson J A. 2010. SVM- and MRF-based method for accurate classification of hyperspectral images. IEEE Geoscience and Remote Sensing Letters， 7（4）： 736-740 ［DOI： 10.1109/LGRS.2010.2047711http://dx.doi.org/10.1109/LGRS.2010.2047711］
Tuia D， Camps-Valls G， Matasci G and Kanevski M. 2010. Learning relevant image features with multiple-kernel classification. IEEE Transactions on Geoscience and Remote Sensing， 48（10）： 3780-3791 ［DOI： 10.1109/TGRS.2010.2049496http://dx.doi.org/10.1109/TGRS.2010.2049496］
Wang H Y， Cheng Y H， Liu X M and Kong Y. 2023. Bi-classifier adversarial network for cross-scene hyperspectral image classification. IEEE Geoscience and Remote Sensing Letters， 20： #5504005 ［DOI： 10.1109/LGRS.2023.3266407http://dx.doi.org/10.1109/LGRS.2023.3266407］
Wu H H， Wang L G and Shi Y. 2021. Convolution neural network method for small-sample classification of hyperspectral images. Journal of Image and Graphics， 26（8）： 2009-2020
吴鸿昊， 王立国， 石瑶. 2021. 高光谱图像小样本分类的卷积神经网络方法. 中国图象图形学报， 26（8）： 2009-2020 ［DOI： 10.11834/jig.210103http://dx.doi.org/10.11834/jig.210103］
Zhang Y X， Li W， Sun W D， Tao R and Du Q. 2023a. Single-source domain expansion network for cross-scene hyperspectral image classification. IEEE Transactions on Image Processing， 32： 1498-1512 ［DOI： 10.1109/TIP.2023.3243853http://dx.doi.org/10.1109/TIP.2023.3243853］
Zhang Y X， Li W， Zhang M M， Qu Y， Tao R and Qi H R. 2023b. Topological structure and semantic information transfer network for cross-scene hyperspectral image classification. IEEE Transactions on Neural Networks and Learning Systems， 34（6）： 2817-2830 ［DOI： 10.1109/TNNLS.2021.3109872http://dx.doi.org/10.1109/TNNLS.2021.3109872］
Zhao C H， Li T and Feng S. 2021. Hyperspectral image classification based on dense convolution and domain adaptation. Acta Photonica Sinica， 50（3）： #0310001
赵春晖， 李彤， 冯收. 2021. 基于密集卷积和域自适应的高光谱图像分类. 光子学报， 50（3）： #0310001 ［DOI： 10.3788/gzxb20215003.0310001http://dx.doi.org/10.3788/gzxb20215003.0310001］
Zhao W Z， Chen X， Chen J G and Qu Y. 2020. Sample generation with self-attention generative adversarial adaptation network （SaGAAN） for hyperspectral image classification. Remote Sensing， 12（5）： #843 ［DOI： 10.3390/rs12050843http://dx.doi.org/10.3390/rs12050843］
Zhou X and Prasad S. 2018. Deep feature alignment neural networks for domain adaptation of hyperspectral data. IEEE Transactions on Geoscience and Remote Sensing， 56（10）： 5863-5872 ［DOI： 10.1109/tgrs.2018.2827308http://dx.doi.org/10.1109/tgrs.2018.2827308］
相关作者
相关机构
京公网安备11010802024621