Please wait a minute...
img

官方微信

遥感技术与应用  2016, Vol. 31 Issue (2): 239-246    DOI: doi:10.11873/j.issn.1004-0323.2016.2.0239
遥感与图像处理     
基于半监督鉴别分析的高分辨率SAR影像建筑区提取
李婷1,2,程博1,尤淑撑3
(1.中国科学院遥感与数字地球研究所数据深加工部,北京 100094;
2.中国科学院大学电子电气与通信工程学院,北京 100049;
3.中国土地勘测规划院遥感所,北京 100035)
Extraction of Building Areas from High-resolution SAR Images based on Semi-supervised Discriminant Analysis
Li Ting1,2,Cheng Bo1,You Shucheng3
(1.Division for Value Added Product,Institute of Remote Sensing and Digital Earth
Chinese Academy of Sciences,Beijing 100094,China;
2.School of Electronic,Electrical and Communication Engineering,University of
Chinese Academy of Sciences,Beijing 100049,China;
3.Remote Sensing Department,China Land Surveying & Planning Institute,Beijing 100035,China)
 全文: PDF(11388 KB)  
摘要:

针对建筑物在城市化发展规划、地理国情信息系统更新、数字化城市以及军事侦察等方面的迫切要求,提出将半监督鉴别分析(Semi\|supervised Discriminant Analysis,SDA)算法应用于高分辨率SAR影像的建筑区提取中,实现快速提取建筑区信息以及提高城市地物目标识别能力。以Radarsat\|2影像和TerraSAR\|X影像为实验数据,基于灰度共生矩阵计算影像的各种纹理特征;结合SDA算法进行特征提取,并以新特征作为大津法(Otsu)的输入提取建筑区;最后对分类结果进行后处理。实验结果与线性鉴别分析(Linear Discriminant Analysis,LDA)算法和局部保持投影(Local Preserving Projection,LPP)算法进行比较,结果表明:SDA算法具有较强的泛化能力,在先验类别信息较少时,适用于高分辨率SAR影像的特征提取,可以快速有效地提取建筑区信息。

关键词: 合成孔径雷达建筑区流形学习特征提取半监督鉴别分析    
Abstract:

Considering urgent request of buildings information for urban development and planning,updating of geographical situation information system,digital city and military reconnaissance,Semi\|supervised Discriminant Analysis (SDA) algorithm was employed for extraction of building areas in high resolution SAR image to improve the ability to target recognition and achieve fast extraction of building areas in city district.Radarsat\|2 images and TerraSAR\|X images were used as the experimental data.A variety of texture features of images were calculated based on gray level co\|occurrence matrix,and then the SDA algorithm was employed for feature extraction.The new feature was used as input of Otsu method to extract building areas.Finally,the post\|processing of image classification was performed.A comparison of the recognition result of building areas between SDA algorithm,Linear Discriminant Analysis (LDA) algorithm and Local Preserving Projection (LPP) algorithm was made.It is concluded that SDA algorithm has stronger generalization ability;SDA algorithm is applicable to feature extraction of high resolution SAR image under less prior class information,quickly and effectively extract the information of building areas in city district.

Key words: Synthetic Aperture Radar(SAR)    Building areas    Manifold learning    Feature extraction    Semi-supervised discriminant analysis
收稿日期: 2015-01-16 出版日期: 2016-06-20
:  TP 75  
基金资助:

国家自然科学基金项目“高分辨率SAR图像典型地物目标样本特征提取和识别研究”(61372189),国家科技合作专项项目(2012DFA20930)资助。

通讯作者: 程博(1974-),男,山东潍坊人,博士,高级工程师,主要从事遥感卫星信息处理及应用研究。Email:bocheng@ceode.ac.cn。   
作者简介: 李婷(1989-),女,江西抚州人,硕士研究生,主要从事遥感技术应用与目标识别分析研究。Email:liting@radi.ac.cn。
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章  
李婷
程博
尤淑撑

引用本文:

李婷,程博,尤淑撑. 基于半监督鉴别分析的高分辨率SAR影像建筑区提取[J]. 遥感技术与应用, 2016, 31(2): 239-246.

Li Ting,Cheng Bo,You Shucheng. Extraction of Building Areas from High-resolution SAR Images based on Semi-supervised Discriminant Analysis. Remote Sensing Technology and Application, 2016, 31(2): 239-246.

链接本文:

http://www.rsta.ac.cn/CN/doi:10.11873/j.issn.1004-0323.2016.2.0239        http://www.rsta.ac.cn/CN/Y2016/V31/I2/239

[1] 万玲,尤红建,程跃兵,卢晓军. 合成孔径雷达图像分割研究进展[J]. 遥感技术与应用, 2018, 33(1): 10-24.
[2] 熊伟,徐永力,姚力波,崔亚奇. 基于SVM的高分辨率SAR图像舰船目标检测算法[J]. 遥感技术与应用, 2018, 33(1): 119-127.
[3] 谢亚楠,周明亮,刘志坤. 合成孔径雷达反演降雨量算法的研究进展[J]. 遥感技术与应用, 2017, 32(4): 624-633.
[4] 李海英,温卫斌,王芳,付强,肖媛. 表层穿透雷达在月球与深空探测应用中的研究进展及趋势[J]. 遥感技术与应用, 2017, 32(4): 634-642.
[5] 姜萍,刘修国,陈启浩,邵芳芳. 利用多尺度SVM-CRF模型的极化SAR图像建筑物提取[J]. 遥感技术与应用, 2017, 32(3): 475-482.
[6] 张祥,陈报章,赵慧,汪磊. 基于时序Sentinel-1A数据的农田土壤水分变化检测分析[J]. 遥感技术与应用, 2017, 32(2): 338-345.
[7] 袁春琦,徐佳,程圆娥,陈媛媛,许康. 基于协同训练与集成学习的极化SAR图像半监督分类[J]. 遥感技术与应用, 2017, 32(2): 380-385.
[8] 孙景,唐丹玲,潘刚. Radarsat2全极化数据在监测南海溢油事件中的方法研究[J]. 遥感技术与应用, 2016, 31(6): 1181-1189.
[9] 李兰,陈尔学,李增元,冯琦,赵磊. 合成孔径雷达森林树高和地上生物量估测研究进展[J]. 遥感技术与应用, 2016, 31(4): 625-633.
[10] 孙建勇,车 涛,段克勤,戴礼云. 冰沟流域积雪面积比例与雪水当量关系初探[J]. 遥感技术与应用, 2015, 30(6): 1085-1094.
[11] 蒋少峰,王超,吴樊,张波,汤益先,张红. 基于结构特征分析的COSMO-SkyMed图像商用船舶分类算法[J]. 遥感技术与应用, 2014, 29(4): 607-615.
[12] 钱进,罗鼎. 基于成对约束半监督降维的高光谱遥感影像特征提取[J]. 遥感技术与应用, 2014, 29(4): 681-688.
[13] 李文梅,李增元,陈尔学,孙晗伟,王馨爽. 极化相干层析反演森林垂直结构原理与方法研究[J]. 遥感技术与应用, 2014, 29(2): 232-239.
[14] 吴樊,王超,张波,张红,田小娟. SAR图像船只分类识别研究进展[J]. 遥感技术与应用, 2014, 29(1): 1-8.
[15] 范伟,荀尚培,杨元建,何彬方,张宏群,吴必文,高锷,孙喜波,陈磊. 星载SAR在涡阳县冬小麦测产中的应用[J]. 遥感技术与应用, 2013, 28(6): 1101-1106.