基于半监督学习的SVM-Wishart极化SAR图像分类方法

滑文强; 王爽; 侯彪

doi:10.12000/JR14138

基于半监督学习的SVM-Wishart极化SAR图像分类方法

DOI: 10.12000/JR14138

1.
(智能感知与图像理解教育部重点实验室西安 710071)
2.
(西安电子科技大学国际智能感知与计算联合研究中心西安 710071)

基金项目:

国家自然科学基金(61173092, 61271302)和陕西省科学技术研究发展计划项目(2013KJXX-64)资助课题

详细信息

作者简介:
滑文强(1987-)，男，陕西人，西安电子科技大学博士研究生，主要研究领域为极化SAR图像处理、机器学习等。E-mail:huawenqiang2013@163.com

王爽(1978-)，女，陕西人，西安电子科技大学教授，博士生导师，智能信息处理研究所副所长，智能感知与图像理解教育部重点实验室成员，国家“111”计划创新引智基地成员，IEEE会员，IET会员，中国电子学会会员，中国计算机学会会员。主要从事SAR/POLSAR处理与分析、稀疏表示、机器学习等方面的研究工作。E-mail:shwang@mail.xidian.edu.cn

侯彪(1974-)，男，陕西人，西安电子科技大学教授，博士生导师，智能感知与图像理解教育部重点实验室副主任，IEEE会员，IET西安分会执行委员会委员，中国电子学会高级会员，陕西信号处理学会理事，教育部创新团队成员。主要研究方向为遥感图像解译、压缩感知、稀疏表示等。E-mail:avcodec@163.com

计量
- 文章访问数: 3918
- HTML全文浏览量: 323
- PDF下载量: 2237
- 被引次数: 0
出版历程
- 收稿日期: 2014-11-20
- 修回日期: 2015-02-28
- 网络出版日期: 2015-02-28

Semi-supervised Learning for Classification of Polarimetric SAR Images Based on SVM-Wishart

1.
(Key Laboratory of Intelligent Perception and Image Understanding of Ministry of Education, Xi’an 710071, China)
2.
(International Research Center for Intelligent Perception and Computation, Xidian University, Xi’an 710071, China)

摘要

摘要: 该文针对极化SAR (Synthetic Aperture Radar)图像分类中的小样本问题，提出了一种新的半监督分类算法。考虑到极化SAR数据反映了地物的散射特性，该方法首先利用目标分解方法提取了多种极化散射特征；其次，在协同训练框架下结合SVM分类器构建了协同半监督模型，该模型可以同时利用有标记和无标记样本对极化SAR图像进行分类，从而在小样本时可以获得更好的分类精度；最后，为进一步改善分类结果，在协同训练分类完成后，该方法又利用Wishart分类器对分类结果进行修正。理论分析与实验表明，该算法在只有少量标记样本的情况下优于传统算法。
- 极化SAR /
- 地物分类 /
- 半监督学习 /
- 协同训练 /
- 支持向量机
Abstract: In this study, we propose a new semi-supervised classification method for Polarimetric SAR (PolSAR) images, aiming at handling the issue that the number of train set is small. First, considering the scattering characters of PolSAR data, this method extracts multiple scattering features using target decomposition approach. Then, a semi-supervised learning model is established based on a co-training framework and Support Vector Machine (SVM). Both labeled and unlabeled data are utilized in this model to obtain high classification accuracy. Third, a recovery scheme based on the Wishart classifier is proposed to improve the classification performance. From the experiments conducted in this study, it is evident that the proposed method performs more effectively compared with other traditional methods when the number of train set is small.
- Polarimetric Synthetic Aperture Radar (SAR) /
- Terrain classification /
- Semi-supervised learning /
- Co-training /
- Support Vector Machine (SVM)

HTML全文

参考文献(17)

[1]	Kersten P R, Lee J S, and Ainsworth T L. Unsupervised classification of polarimetric synthetic aperture radar images using fuzzy clustering and EM clustering[J]. IEEE Transactions on Geoscience and Remote Sensing, 2005, 43(3): 519-527.
[2]	Wang S, Liu K, Pei J J, et al.. Unsupervised classification of fully polarimetric SAR images based on scattering power entropy and copolarized ratio[J]. IEEE Geoscience and Remote Sensing Letters, 2013, 10(3): 622-626.
[3]	Shang F and Hirose A. Use of poincare sphere parameters for fast supervised PolSAR land classification[C]. IEEE Geoscience and Remote Sensing Symposium, Melbourne, Australia, 2013: 3175-3178.
[4]	Shi L, Zhang L F, and Yang J. Supervised graph embedding for Polarimetric SAR image classification[J]. IEEE Geoscience and Remote Sensing Letters, 2013, 10(2): 216-220.
[5]	Hady M and Schwenker F. Co-training by committee: a new semi-supervised learning framework[C]. IEEE International Conference on Data Mining Workshops, 2008: 563-572.
[6]	Hansch R and Hellwich O. Semi-supervised learning for classification of polarimetric SAR-data[C]. IEEE Geoscience and Remote Sensing Symposium, Cape Town, South Africa, 2009: 987-990.
[7]	Lee J S, Grunes M R, and Famil L F. Unsupervised terrain classification preserving polarimetric scattering characteristics[J]. IEEE Transactions on Geoscience and Remote Sensing, 2004, 42(4): 722-731.
[8]	He Y and Cheng J. Classification based on Four-component decomposition and SVM for PolSAR images[C]. IEEE International Conference on Automatic Control and Artificial Intelligence (ACAI 2012), Xiamen, China, 2012: 635-637.
[9]	Blum A and Mitchell T. Combining labeled and unlabeled data with co-training[C]. Proceedings of the 11th Annual Conference on Computational Learning Theory, Wisconsin, USA, 1998: 92-100.
[10]	Cloude S R and Pottier E. A review of target decomposition theorems in radar polarimetry[J]. IEEE Transactions on Geoscience and Remote Sensing, 1996, 34(2): 498-518.
[11]	Cloude S R and Pottier E. An entropy based classification scheme for land application of polarimetric SAR[J]. IEEE Transactions on Geoscience and Remote Sensing, 1997, 35(1): 68-78.
[12]	Krogager E. New decomposition of the radar target scattering matrix[J]. Electronics Letters, 1990, 26(18): 1525-1527.
[13]	Huang J, Shirabad J S, Matwin S, et al.. Improving cotraining with agreement-based sampling[C]. 7th International Conference, RSCTC, Warsaw, Poland, 2010: 197-206.
[14]	边肇祺, 张学工, 等. 模式识别[M]. 北京: 清华大学出版社, 2000: 136-140. Bian Z Q, Zhang X G, et al.. Pattern Recognition[M]. Beijing: Tsinghua University Press, 2000: 136-140.
[15]	Zanchettin C, Bezerra B L, and Azevedo W A. A KNN-SVM hybrid model for cursive handwriting recognition[C]. WCCI IEEE World Congress on Computational Intelligence, Brisbane, Australia, 2012, 6: 10-15.
[16]	Lee J S, Grunes M R, and Kwok R. Classification of multilook polarimetric SAR imagery based on complex Wishart distribution[J]. International Journal of Remote Sensing, 1994, 15(11): 2299-2311.
[17]	Lee J S, Grunes M R, and Grandi G. Polarimetric SAR speckle filtering and its implication for classification[J]. IEEE Transcations on Geoscience and Remote Sensing, 1999, 37(5): 2363-2373.