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

doi:10.13485/j.cnki.11-2089.2014.0163

测绘学报 ›› 2014, Vol. 43 ›› Issue (9): 908-916.doi: 10.13485/j.cnki.11-2089.2014.0163

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

吴田军¹,²,骆剑承³,夏列钢¹,³,杨海平¹,³,沈占锋²,胡晓东³

1. 中国科学院大学
2. 中国科学院遥感与数字地球研究所
3. 中国科学院遥感应用研究所

收稿日期:2014-01-05 修回日期:2014-02-25 出版日期:2014-09-20 发布日期:2014-09-25
通讯作者: 吴田军 E-mail:wutianjun1986@163.com
基金资助:
国家自然科学基金项目;国家自然科学基金项目;国家863计划项目;国家科技支撑计划重点项目课题;国家国际科技合作计划项目;国家科技支撑课题

An Automatic Sample Collection Method for Object-oriented Classification of Remotely Sensed Imageries Based on Transfer Learning

WU Tianjun¹,²,LUO Jiancheng³,XIA Liegang¹,²,YANG Haiping¹,²,SHEN Zhanfeng¹,HU Xiaodong¹

1. Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences
2. University of Chinese Academy of Sciences
3. Institute of Remote Sensing Applications, Chinese Academy of Sciences

Received:2014-01-05 Revised:2014-02-25 Online:2014-09-20 Published:2014-09-25
Contact: WU Tianjun E-mail:wutianjun1986@163.com

摘要/Abstract

摘要：

面向遥感大范围应用的目标，自动化程度仍是遥感影像分类面临的重要问题，样本的人工选择难以适应当前土地覆盖信息自动化提取的实际应用需求。为了构建一套基于先验知识的遥感影像全自动分类流程，本文将空间信息挖掘技术引入到遥感信息提取过程中，提出了一种面向遥感影像对象级分类的样本自动选择方法。该方法通过变化检测将不变地物标示在新的目标影像上，并将过去解译的地物类别知识迁移至新的影像上，建立新的特征与地物关系，从而完成历史专题数据辅助下目标影像的自动化的对象级分类。实验结果表明，在已有历史专题层的图斑知识指导下，该方法能有效地自动选择适用于新影像分类的可靠样本，获得较好的信息提取效果，提高了对象级分类的效率。

关键词: 自动化, 土地覆盖, 对象级分类, 样本选择, 变化检测, 迁移学习

Abstract:

For the large-scale remote sensing applications, the automatic classification of remotely sensed imageries is still a challenge. For example, the artificial sample collection scheme cannot meet the needs of automatic information extraction from the remotely sensed imageries. In order to establish a prior knowledge-based and fully automatic classification method, an automatic sample collection method for object-oriented classification, with the introduction of data mining to the process of information extraction, is proposed. Firstly, the unchanged landmarks are located. Then the prior class knowledge from old interpreted thematic images is transferred to the new target images. And the above knowledge is then used to rebuild the relationship between landmark classes and their spatial-spectral features. The results show that, with the assist of preliminary thematic data, the approach can automatically obtain reliable object samples for object-oriented classification. The accuracy of the classified land-cover types and the efficiency of object-oriented classification are both improved.

Key words: automation, land-cover, object-oriented classification, sample collection, change detection, transfer learning

中图分类号:

TP753

吴田军骆剑承夏列钢杨海平沈占锋胡晓东. 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 YANG Haiping SHEN Zhanfeng HU Xiaodong. 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.

参考文献

[1] LIU Jiyuan. Study on Chinese Resources and Environment Survey and Dynamic Monitoring Using Remote Sensing [M]. Beijing: Press of University of Science and Technology of China, 1996. (刘纪远. 中国资源环境遥感宏观调查与动态研究 [M]. 北京: 中国科学技术出版社, 1996.)
[2] GAN Fuping, WANG Runsheng, WANG Yongjian, et al. The Classification Method Based on Remote Sensing Techniques for Land Use and Cover [J]. Remote Sensing for Land & Resources, 1999, 4(4): 40-45. (甘甫平, 王润生, 王永江, 等. 基于遥感技术的土地利用与土地覆盖的分类方法 [J]. 国土资源遥感, 1999, 4(4): 40-45.)
[3] RICHARDS J. A., JIA X. P. Remote Sensing Digital Image Analysis: An Introduce (4th Edition) [M]. Berlin: Springer-Verlag Inc., 2012.
[4] LU D., WENG Q. A Survey of Image Classification Methods and Techniques for Improving Classification Performance [J]. International Journal of Remote Sensing, 2007, 28(5): 823-870.
[5] TSO B., MATHER P. Classification Methods for Remotely Sensed Data (2nd Edition) [M]. New York: Taylor & Francis Inc., 2009.
[6] LUO Jiancheng, ZHOU Chenghu, YANG Yan. Land-cover and Land-use Classification Based on Remote Sensing Intelligent Geo-interpreting Model [J]. Journal of Natural Resources, 2001, 16(2): 179-183. (骆剑承, 周成虎, 杨艳. 遥感地学智能图解模型支持下的土地覆盖/土地利用分类 [J]. 自然资源学报, 2001, 16(2): 179-183.)
[7] LUO Jiancheng, WANG Qinmin, MA Jianghong, et al. The EM-based Maximum Likelihood Classifier for Remotely Sensed Data [J]. Accta Geodaetica et Cartographica Sinica, 2002, 31(3): 234-239. (骆剑承, 王钦敏, 马江洪, 等. 遥感图像最大似然分类方法EM改进算法 [J]. 测绘学报, 2002, 31(3): 234-239.)
[8] LUO Jiancheng, ZHOU Chenghu, LEUNG Yee, et al. Support Vector Machine for Spatial Feature Extraction and Classification of Remotely Sensed Imagery [J]. Journal of Remote Sensing, 2002, 6(1): 50-55. (骆剑承, 周成虎, 梁怡, 等. 支持向量机及其遥感影像空间特征提取和分类的应用研究 [J]. 遥感学报, 2002, 6(1): 50-55.)
[9] BARROS, RODRIGO C., BASGALUPP M. P., et al. A Survey of Evolutionary Algorithms for Decision-Tree Induction [J]. IEEE Transactions on Systems, Man and Cybernetics, Part C: Applications and Reviews, 2012, 42(3), 291-312.
[10] LEI Zhen. Random Forest and Its Application in Remote Sensing [D]. Shanghai: Shanghai Jiao Tong University, 2012. (雷震. 随机森林及其在遥感影像处理中应用研究 [D]. 上海: 上海交通大学, 2012.)
[11] LUO Jiancheng, ZHOU Chenghu, YANG Yan. ANN Remote Sensing Classification Model and Its Integration Approach with Geo-knowledge [J]. Journal of Remote Sensing, 2001, 5(2): 122-129. (骆剑承, 周成虎, 杨艳. 人工神经网络遥感影像分类模型及其与知识集成方法研究 [J]. 遥感学报, 2001, 5(2): 122-129.)
[12] BRIEM G. J., BENEDIKTSSON J. A., SVEINSSON J. R. Multiple Classifiers Applied to Multisource Remote Sensing Data [J]. IEEE Transactions on Geoscience and Remote Sensing, 2002, 40(10): 2291-2299.
[13] MENNIS J., GUO D. Spatial Data Mining and Geographic Knowledge Discovery: An Introduction [J]. Computers, Environment and Urban Systems, 2009, 33(6): 403-408.
[14] XIA Liegang. Study on Automatic Classification Method for Remotely Sensed Imagery by Incorporating Spatial-Spectral Features [D]. Hangzhou: Zhejiang University of Technology, 2011. (夏列钢. 耦合“图-谱”特征的遥感影像自动分类方法研究 [D]. 杭州: 浙江工业大学, 2011.)
[15] ZHOU Chenghu, LUO Jiancheng. Geo-computation of High-resolution Satellite Remote Sensing Image [M]. Beijing: Science Press, 2009. (周成虎, 骆剑承. 高分辨率卫星遥感影像地学计算 [M]. 北京: 科学出版社, 2009.)
[16] PAN S. J., YANG Q. A Survey on Transfer Learning [J]. IEEE Transactions on Knowledge and Data Engineering, 2010, 22(10): 1345-1359.
[17] DAI W. Y., CHEN Y. Q., XUE G. R., et al. Translated Learning: Transfer Learning across Different Feature Spaces [C]// Processing of 2008 Advances in Neural Information Processing Systems 12(NIPS 2008). Vancouver: 2008, 353-360.
[18] COMANICIU D., MEER P. Mean Shift：A Robust Approach Toward Feature Space Analysis [J]. IEEE Transaction on Pattern Analysis and Machine Intelligence, 2002, 24(5): 603-619.
[19] CHEN Shupeng, YUE Tianxiang, LI Huiguo. Studies on Geo-informatic Tupu and Its Application [J]. Geographical Research, 2000, 19(4): 337-343. (陈述彭, 岳天祥, 励惠国. 地学信息图谱研究及其应用 [J]. 地理研究, 2000, 19(4): 337-343.)
[20] LUO Jiancheng, ZHOU Chenghu, SHEN Zhanfeng, et al. Theoretic and Methodological Review on Sensor Information Tupu Computation [J]. Journal of Geo-information Science, 2009, 11(5): 664-669. (骆剑承, 周成虎, 沈占锋, 等. 遥感信息图谱计算的理论方法研究 [J]. 地球信息科学学报, 2009, 11(5): 664-669.)
[21] GAO Wei. The Study of Information Extraction Technology for Remote Sensing Based On Feature Knowledge [D]. Beijing: China University of Geosciences, 2010. (高伟. 基于特征知识库的遥感信息提取技术研究 [D]. 北京: 中国地质大学, 2010.)
[22] WANG Yan. The Research of Land Use/Cover Change Detection Using High- resolution Remote Sensing Images Based on Image Segments Statistical Analysis [D]. Wuhan: Wuhan University, 2012. (王琰. 基于像斑统计分析的高分辨率遥感影像土地利用/覆盖变化检测方法研究 [D]. 武汉: 武汉大学, 2012.)
[23] BISHOP C. M. Pattern Recognition and Machine Learning [M]. New York: Springer-Verlag Inc., 2006.
[24] ZHOU Z. H., ZHANG M. L., HUANG S. J., et al. Multi-instance Multi-label Learning [J]. Artificial Intelligence, 2012, 176(1), 2291-2320.
[25] NIELSEN A. A. The Regularized Iteratively Reweighted MAD Method for Change Detection in Multi- and Hyperspectral Data [J]. IEEE Transaction on Image Processing, 2007, 16(2), 463-478.
[26] ZHU X. J. Semi-Supervised Learning Literature Survey [R]. Computer Sciences Technical Report TR 1530, University of Wisconsin Madison, 2008.

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

An Automatic Sample Collection Method for Object-oriented Classification of Remotely Sensed Imageries Based on Transfer Learning

PDF

可视化

被引次数

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]	梁哲恒, 黎宵, 邓鹏, 盛森, 姜福泉. 融合多尺度特征注意力的遥感影像变化检测方法[J]. 测绘学报, 2022, 51(5): 668-676.
[3]	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.
[4]	王昶, 张永生, 纪松, 张磊. 建筑物变化的多特征融和及随机多图综合检测法[J]. 测绘学报, 2021, 50(2): 235-247.
[5]	汤寓麟, 金绍华, 边刚, 张永厚, 李凡. 侧扫声呐识别沉船影像的迁移学习卷积神经网络法[J]. 测绘学报, 2021, 50(2): 260-269.
[6]	叶沅鑫, 孙苗苗, 王蒙蒙, 谭鑫. 结合邻域信息和结构特征的遥感影像变化检测[J]. 测绘学报, 2021, 50(10): 1349-1357.
[7]	郑鑫, 潘斌, 张健. 可变形网络与迁移学习相结合的电力塔遥感影像目标检测法[J]. 测绘学报, 2020, 49(8): 1042-1050.
[8]	王舒洋, 慕晓冬, 贺浩, 杨东方, 马晨晖. 航拍图像跨数据域特征迁移道路提取方法[J]. 测绘学报, 2020, 49(5): 611-621.
[9]	赵传, 郭海涛, 卢俊, 余东行, 张保明. 基于深度残差网络的机载LiDAR点云分类[J]. 测绘学报, 2020, 49(2): 202-213.
[10]	赵诣, 蒋弥. 极化SAR参数优化与光学波谱相结合的面向对象土地覆盖分类[J]. 测绘学报, 2019, 48(5): 609-617.
[11]	赵生银, 安如, 朱美如. 联合像元-深度-对象特征的遥感图像城市变化检测[J]. 测绘学报, 2019, 48(11): 1452-1463.
[12]	Yun ZHANG. 智能摄影测量和图像处理在高分辨率光学遥感影像处理中的应用——CRC-AGIP实验室的案例[J]. 测绘学报, 2018, 47(6): 722-729.
[13]	万杰, Alper YILMAZ. 基于深度卷积特征的影像关系表创建方法[J]. 测绘学报, 2018, 47(6): 882-891.
[14]	王建明, 史文中, 邵攀. 自适应距离和模糊拓扑优化的模糊聚类SAR影像变化检测[J]. 测绘学报, 2018, 47(5): 611-619.
[15]	梅莹莹, 张景雄. 土地覆盖变化信息自适应抽样策略及其精度评估[J]. 测绘学报, 2018, 47(5): 644-651.