结合对抗网络与辅助任务的遥感影像无监督域适应方法

doi:10.11947/j.AGCS.2017.20170291

测绘学报 ›› 2017, Vol. 46 ›› Issue (12): 1969-1977.doi: 10.11947/j.AGCS.2017.20170291

结合对抗网络与辅助任务的遥感影像无监督域适应方法

许夙晖¹, 慕晓冬¹, 张雄美¹, 柴栋²

1. 火箭军工程大学信息工程系, 陕西西安 710025;
2. 北京航空工程技术研究中心, 北京 100076

收稿日期:2017-06-05 修回日期:2017-10-24 出版日期:2017-12-20 发布日期:2017-12-28
作者简介:许夙晖(1989-),女,博士生,研究方向为遥感图像处理和模式识别。E-mail:xu_suhui@163.com
基金资助:
国家自然科学基金（61640007）

Unsupervised Remote Sensing Domain Adaptation Method with Adversarial Network and Auxiliary Task

XU Suhui¹, MU Xiaodong¹, ZHANG Xiongmei¹, CHAI Dong²

1. Department of Information Engineering, Rocket Force Engineering University, Xi'an 710025, China;
2. Beijing Aeronautical Technology Research Center, Beijing 100076, China

Received:2017-06-05 Revised:2017-10-24 Online:2017-12-20 Published:2017-12-28
Supported by:
The National Natural Science Foundation of China (No. 61640007)

摘要/Abstract

摘要： 使用机器学习进行遥感影像标注的一个重要前提是有足够的训练样本，而样本的标注是非常耗时的。本文采用了域适应的方法来解决遥感影像场景分类中小样本量的无监督学习问题，提出了结合对抗网络与辅助任务的遥感影像域适应方法。首先建立了基于深度卷积神经网络的遥感影像分类框架；其次，为了学习到域不变特征，在标签分类器的基础上增加域分类器，并使域损失函数在其反射传播时的梯度与标签损失的梯度相反，从而保证域分类器不能区分样本来自于哪个域；最后引入了辅助分类任务，扩充了样本的同时使网络更具泛化能力。试验结果表明，本文方法优于主流的无监督域适应方法，在小样本遥感影像无监督分类中得到了较好的效果。

关键词: 遥感影像, 场景分类, 域适应, 深度卷积神经网络, 对抗网络, 多任务学习

Abstract: An important prerequisite when annotating the remote sensing images by machine learning is that there are enough training samples for training, but labeling the samples is very time-consuming. In this paper, we solve the problem of unsupervised learning with small sample size in remote sensing image scene classification by domain adaptation method. A new domain adaptation framework is proposed which combines adversarial network and auxiliary task. Firstly, a novel remote sensing scene classification framework is established based on deep convolution neural networks. Secondly, a domain classifier is added to the network, in order to learn the domain-invariant features. The gradient direction of the domain loss is opposite to the label loss during the back propagation, which makes the domain predictor failed to distinguish the sample's domain. Lastly, we introduce an auxiliary task for the network, which augments the training samples and improves the generalization ability of the network. The experiments demonstrate better results in unsupervised classification with small sample sizes of remote sensing images compared to the baseline unsupervised domain adaptation approaches.

Key words: remote sensing image, scene classification, domain adaptation, deep convolutional neural network, adversarial network, multi-task learning

中图分类号:

P237

许夙晖, 慕晓冬, 张雄美, 柴栋. 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, CHAI Dong. Unsupervised Remote Sensing Domain Adaptation Method with Adversarial Network and Auxiliary Task[J]. Acta Geodaetica et Cartographica Sinica, 2017, 46(12): 1969-1977.

参考文献

[1] CHENG Gong,HAN Junwei,LU Xiaoqiang.Remote Sensing Image Scene Classification:Benchmark and State of the Art[J].Proceedings of the IEEE,2017,105(10):1865-1883.
[2] HU Fan,XIA Guisong,HU Jingwen,et al.Transferring Deep Convolutional Neural Networks for the Scene Classification of High-resolution Remote Sensing Imagery[J].Remote Sensing,2015,7(11):14680-14707.
[3] ZOU Qin,NI Lihao,ZHANG Tong,et al.Deep Learning Based Feature Selection for Remote Sensing Scene Classification[J].IEEE Geoscience and Remote Sensing Letters,2015,12(11):2321-2325.
[4] 许夙晖,慕晓冬,赵鹏,等.利用多尺度特征与深度网络对遥感影像进行场景分类[J].测绘学报,2016,45(7):834-840.DOI:10.11947/j.AGCS.2016.20150623. XU Suhui,MU Xiaodong,ZHAO Peng,et,al.Scene Classification of Remote Sensing Image Based on Multi-scale Feature and Deep Neural Network[J] Acta Geodaetica et Cartographica Sinica,2016,45(7):834-840.DOI:10.11947/j.AGCS.2016.20150623.
[5] ZHANG Fan,DU Bo,ZHANG Liangpei.Scene Classification via a Gradient Boosting Random Convolutional Network Framework[J].IEEE Transactions on Geoscience and Remote Sensing,2016,54(3):1793-1802.
[6] ROMERO A,GATTA C,CAMPS-VALLS G.Unsupervised Deep Feature Extraction for Remote Sensing Image Classification[J].IEEE Transactions on Geoscience and Remote Sensing,2016,54(3):1349-1362.
[7] NOGUEIRA K,PENATTI O A B,DOS SANTOS J A.Towards Better Exploiting Convolutional Neural Networks for Remote Sensing Scene Classification[J].Pattern Recognition,2017(61):539-556.
[8] ZHOU Xichuan,LI Shengli,TANG Fang,et al.Deep Learning with Grouped Features for Spatial Spectral Classification of Hyperspectral Images[J].IEEE Geoscience and Remote Sensing Letters,2017,14(1):97-101.
[9] GE Weifeng,YU Yizhou.Borrowing Treasures from the Wealthy:Deep Transfer Learning through Selective Joint Fine-tuning[J].arXiv preprint arXiv:1702.08690,2017.
[10] LIN Daoyu,FU Kun,WANG Yang,et al.MARTA GANs:Unsupervised Representation Learning for Remote Sensing Image Classification[J].IEEE Geoscience and Remote Sensing letters, 2017,14(11):2092-2096.
[11] 何小飞,邹峥嵘,陶超,等.联合显著性和多层卷积神经网络的高分影像场景分类[J].测绘学报,2016,45(9):1073-1080.DOI:10.11947/j.AGCS.2016.20150612. HE Xiaofei,ZOU Zhengrong,TAO Chao,et al.Combined Saliency with Multi-convolutional Neural Network for High Resolution Remote Sensing Scene Classification[J].Acta Geodaetica et Cartographica Sinica,2016,45(9):1073-1080.DOI:10.11947/j.AGCS.2016.20150612.
[12] WEISS K,KHOSHGOFTAAR T M,WANG D D.A Survey of Transfer Learning[J].Journal of Big Data,2016,3(1):9-48.
[13] KULIS B,SAENKO K,DARRELL T.What You Saw is not What You Get:Domain Adaptation Using Asymmetric Kernel Transforms[C]//Proceedings of 2011 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).Colorado Springs,CO,USA:IEEE,2011:1785-1792.
[14] HOFFMAN J,RODNER E,DONAHUE J,et al.Efficient Learning of Domain-invariant Image Representations[C]//Proceedings of the International Conference on Learning Representations (ICLR).Scottsdale,Arizona:ICLR,2013.
[15] DUAN Lixin,XU Dong,TSANG I W.Learning with Augmented Features for Heterogeneous Domain Adaptation[C]//Proceedings of the 29th International Conference on Machine Learning.Edinburgh,Scotland,UK:ICML,2012:711-718.
[16] GONG Boqing,SHI Yuan,SHA Fei,et al.Geodesic Flow Kernel for Unsupervised Domain Adaptation[C]//Proceedings of 2012 IEEE Conference on Computer Vision and Pattern Recognition.Providence,RI,USA:IEEE,2012:2066-2073.
[17] GONG Boqing,GRAUMAN K,SHA Fei.Connecting the Dots with Landmarks:Discriminatively Learning Domain-invariant Features for Unsupervised Domain Adaptation[C]//Proceedings of the 30th International Conference on Machine Learning.Atlanta,USA:ICML,2013:222-230.
[18] TZENG E,HOFFMAN J,ZHANG Ning,et al.Deep Domain Confusion:Maximizing for Domain Invariance[J].arXiv preprint arXiv:1412.3474,2014.
[19] TZENG E,HOFFMAN J,DARRELL T,et al.Simultaneous Deep Transfer Across Domains and Tasks[C]//Proceedings of 2015 IEEE International Conference on Computer Vision.Santiago,Chile:IEEE,2015:4068-4076.
[20] GANIN Y,LEMPITSKY V.Unsupervised Domain Adaptation by Backpropagation[C]//Proceedings of the 32nd International Conference on Machine Learning.Lille,France:ICML,2015:1180-1189.
[21] GOODFELLOW I,POUGET-ABADIE J,MIRZA M,et al.Generative Adversarial Nets[C]//Proceedings of the 27th International Conference on Neural Information Processing Systems.Montreal,Canada:MIT Press,2014:2672-2680.
[22] TUIA D,PERSELLO C,BRUZZONE L.Domain Adaptation for the Classification of Remote Sensing Data:An Overview of Recent Advances[J].IEEE Geoscience and Remote Sensing Magazine,2016,4(2):41-57.
[23] LI Xue,ZHANG Liangpei,DU Bo,et al.Iterative Reweighting Heterogeneous Transfer Learning Framework for Supervised Remote Sensing Image Classification[J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2017,10(5):2022-2035.
[24] ZHANG Zhanpeng,LUO Ping,LOY C C,et al.Facial Landmark Detection by Deep Multi-task Learning[C]//Proceedings of the 13th European Conference on Computer Vision.Zurich,Switzerland:Springer,2014:94-108.
[25] SELTZER M L,DROPPO J.Multi-task Learning in Deep Neural Networks for Improved Phoneme Recognition[C]//Proceedings of 2013 IEEE International Conference on Acoustics,Speech and Signal Processing (ICASSP).Vancouver,BC,Canada:IEEE,2013:6965-6969.
[26] KRIZHEVSKY A,SUTSKEVER I,HINTON G E.Imagenet Classification with Deep Convolutional Neural Networks[C]//Advances in Neural Information Processing Systems.Spain:NIPS,2012:1097-1105.
[27] VAN DER MAATEN L,HINTON G.Visualizing Data Using t-SNE[J].Journal of Machine Learning Research,2008,9(11):2579-2605.

结合对抗网络与辅助任务的遥感影像无监督域适应方法

Unsupervised Remote Sensing Domain Adaptation Method with Adversarial Network and Auxiliary Task

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	张祖勋, 姜慧伟, 庞世燕, 胡翔云. 多时相遥感影像的变化检测研究现状与展望[J]. 测绘学报, 2022, 51(7): 1091-1107.
[2]	李荣兴, 李国君, 冯甜甜, 沈强, 乔刚, 叶真, 夏梦莲. 基于光学遥感卫星影像的南极冰流速产品和方法研究综述[J]. 测绘学报, 2022, 51(6): 953-963.
[3]	程结海, 黄中意, 王建如, 何湜. 高空间分辨率遥感影像最优分割结果自动确定方法[J]. 测绘学报, 2022, 51(5): 658-667.
[4]	白坤, 慕晓冬, 陈雪冰, 朱永清, 尤轩昂. 融合半监督学习的无监督遥感影像场景分类[J]. 测绘学报, 2022, 51(5): 691-702.
[5]	廖章回, 姜闯. 高分辨率遥感影像快速去雾[J]. 测绘学报, 2022, 51(3): 446-456.
[6]	何直蒙, 丁海勇, 安炳琪. 高分辨率遥感影像建筑物提取的空洞卷积E-Unet算法[J]. 测绘学报, 2022, 51(3): 457-467.
[7]	洪亮, 冯亚飞, 彭双云, 楚森森. 面向对象的多尺度加权联合稀疏表示的高空间分辨率遥感影像分类[J]. 测绘学报, 2022, 51(2): 224-237.
[8]	葛小三, 陈曦, 赵文智, 李瑞祥. 基于生成对抗网络的建筑物损毁检测[J]. 测绘学报, 2022, 51(2): 238-247.
[9]	余俊鹏, 吴伟东, 孙佳明, 满益云, 沈刚. 附加傅里叶补偿项的卫星遥感影像RFM平差方法[J]. 测绘学报, 2022, 51(1): 127-134.
[10]	张玉鑫, 颜青松, 邓非. 高分辨率遥感影像建筑物提取多路径RSU网络法[J]. 测绘学报, 2022, 51(1): 135-144.
[11]	龚健雅, 许越, 胡翔云, 姜良存, 张觅. 遥感影像智能解译样本库现状与研究[J]. 测绘学报, 2021, 50(8): 1013-1022.
[12]	陶超, 阴紫薇, 朱庆, 李海峰. 遥感影像智能解译:从监督学习到自监督学习[J]. 测绘学报, 2021, 50(8): 1122-1134.
[13]	施慧慧, 徐雁南, 滕文秀, 王妮. 高分辨率遥感影像深度迁移可变形卷积的场景分类法[J]. 测绘学报, 2021, 50(5): 652-663.
[14]	郑凯, 李建胜, 王俊强, 欧阳文, 谷友艺, 张迅. DCLS-GAN:利用生成对抗网络的天绘一号卫星高原地区影像去云方法[J]. 测绘学报, 2021, 50(2): 248-259.
[15]	李雅丽, 向隆刚, 张彩丽, 吴华意, 龚健雅. 车辆轨迹与遥感影像多层次融合的道路交叉口识别[J]. 测绘学报, 2021, 50(11): 1546-1557.