Please wait a minute...
img

官方微信

遥感技术与应用  2016, Vol. 31 Issue (4): 731-738    DOI: 10.11873/j.issn.1004-0323.2016.4.0731
数据与图像处理     
基于形态学属性剖面的高光谱影像集成分类
鲍蕊1,2,3,夏俊士2,3,薛朝辉2,3,杜培军2,3,车美琴2,3
(1.天津市地质调查研究院,天津 300191;2.卫星测绘技术与应用国家测绘地理信息局重点实验室,
南京大学,江苏 南京 210023;3.江苏省地理信息技术重点实验室,南京大学,江苏 南京 210023)
Ensemble Classification for Hyperspectral Imagery based on Morphological Attribute Profiles
Bao Rui1,2,3,Xia Junshi2,3,Xue Zhaohui2,3,Du Peijun2,3,Che Meiqin2,3
1.Tianjin Institute of Geological Survey,Tianjin 300191,China;
2.Key Laboratory for Satellite Mapping Technology and Applications of National Administration of
Surveying,Mapping and Geoinformation of China,Nanjing University,Nanjing 210023,China;
3.Jiangsu Provincial Key Laboratory of Geographic Information Science and Technology,
Nanjing University,Nanjing 210023,China
 全文: PDF(10198 KB)  
摘要:

传统高光谱遥感影像逐像素分类方法未考虑像元之间的空间关联性且泛化性能较低。形态学属性剖面是表征影像空间结构的有效方法,同时集成学习可显著提升分类算法的泛化能力。为了在高光谱影像分类中充分利用影像的空间信息并提高分类的稳定性,提出一种基于形态学属性剖面高光谱遥感影像集成学习分类方法。首先,用主成分分析和最小噪声变换进行特征提取,并借助形态学属性剖面获取影像的多重空间特征;然后用极限学习和支持向量机的方法进行分类;最后将多个分类结果以多数投票的方式集成。区别于已有集成学习方法,综合考虑了不同特征提取和不同分类方法的联合集成,并将形态学属性剖面引入其中以充分利用影像的空间信息。采用AVIRIS和ROSIS两组高光谱数据检验该方法的分类性能,实验结果表明该方法可获得高精度和高稳定性的分类结果,总体精度分别达到83.41%和95.14%。

关键词: 形态学属性剖面集成学习支持向量机极限学习高光谱影像分类    
Abstract:

The traditional pixel\|wised classification methods is lack of hyperspectral image (HIS) ignore spatial information and generalization ability,resulting in a big limit of classification performance.Morphological attribute profiles is an effective method to express spatial information of remote sensing image.In the meanwhile,ensemble learning machines possess high ability of power generalization so that to improve classification stability.An ensemble method based on morphological attribute profiles is proposed for hyperspectral image in order to make full advantage of spatial information to improve stability.Firstly,principal component analysis and minimum noise fraction are utilized for feature extraction,and then morphological attribute profiles operations are carried out on the first a few features which have most information of the image;Secondly,supporting vector machines and extreme learning machines are used as base classifiers for their good classification performance.Finally,the results of each base classifier are combined by 〖JP3〗the way of majority voting.Compared with other ensemble method,different feature extraction algorithm and different base classifiers are both combined to form the joint integrated model.In addition,spatial information deriving from morphological attribute profiles is introduced to improve classification accuracy. Experiment on AVIRIS data set and ROSIS data set respectively illustrate that the method can obtain better classification performance in terms of precision and stability,and the overall accuracy of them reach to 83.41% and 95.14%,respectively.
 

Key words: Ensemble learning    Support vector machines (SVMs)    Extreme learning machines (ELMs)    Hyperspectral imagery classification
收稿日期: 2015-04-28 出版日期: 2016-10-14
:  TP 75  
基金资助:

江苏省杰出青年基金项目(BK2012018),国家重大科学仪器设备开发专项(012YQ050250)资助。

通讯作者: 杜培军(1975-),男,山西五台人,博士,教授,主要从事遥感图像处理与模式识别、资源环境遥感与信息系统、高光谱遥感方面的研究。Emaildupjrs@126.com。   
作者简介: 鲍蕊(1990-),女,山东济宁人,硕士研究生,主要从事高光谱影像光谱—空间分类研究。E\|mail:baoruijiayou@163.com。
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  
鲍蕊
夏俊士
薛朝辉
杜培军
车美琴

引用本文:

鲍蕊,夏俊士,薛朝辉,杜培军,车美琴. 基于形态学属性剖面的高光谱影像集成分类[J]. 遥感技术与应用, 2016, 31(4): 731-738.

Bao Rui,Xia Junshi,Xue Zhaohui,Du Peijun,Che Meiqin. Ensemble Classification for Hyperspectral Imagery based on Morphological Attribute Profiles. Remote Sensing Technology and Application, 2016, 31(4): 731-738.

链接本文:

http://www.rsta.ac.cn/CN/10.11873/j.issn.1004-0323.2016.4.0731        http://www.rsta.ac.cn/CN/Y2016/V31/I4/731

[1]Pu Ruiliang,Gong Peng.Hyperspectral Remote Sensing and Its Application[M].Beijing:Higher Education Press,2000.[蒲瑞良,宫鹏.高光谱遥感及其应用[M].北京:高等教育出版社,2000.]

[2]Tong Qingxi,Zhang Bing,Zheng Lanfen.Hyperspectral Remote Sensing:The Principle,Technology and Application[M].Beijing:Higher Education Press,2006.[童庆禧,张兵,郑兰芬.高光谱遥感——原理、技术与应用[M].北京:高等教育出版社,2006.]

[3]Bioucas-Dias J M,Plaza A,Camps-Valls G,et al.Hyperspectral Remote Sensing Data Analysis and Future Challenges[J].IEEE Geoscience and Remote Sensing Magazine,2013,1(2):6-36.

[4]Plaza A,Benediktsson J A,Boardman J W,et al.Recent Advances in Techniques for Hyperspectral Image Processing[J].Remote Sensing of Environment,2009,113(9):S110-S122.4.

[5]Zhou Ziyong.Progress in Hyperspectral Remote Sensing for Petroleum Prospecting[J].Remote Sensing Technology and Application,2014,29(2):352-361.[周子勇.高光谱遥感油气勘探进展[J].遥感技术与应用,2014,29(2):352-361.]

[6]Mura M D,Benediktsson J A,Waske B,et al.Morphological Attribute Profiles for the Analysis of very High Resolution Images[J].IEEE Transactions on Geoscience and Remote Sensing,2010,48(10):3747-3762.

[7]Dalla Mura M,Villa A,Benediktsson J A,et al.Classification ofHyperspectral Images by Using Extended Morphological Attribute Profiles and Independent Component Analysis[J].IEEE Geoscience and Remote Sensing Letters,2011,8(3):542-546.

[8]Song B,Li J,Dalla Mura M,et al.Remotely Sensed Image Classification Using Sparse Representations of Morphological Attribute Profiles[J].IEEE Transactions on Geoscience and Remote Sensing,2014,52(8):5122-5136.

[9]Ghamisi P,Benediktsson J A,Cavallaro G,et al.Automatic Framework for Spectral-Spatial Classification based on Supervised Feature Extraction and Morphological Attribute Profiles[J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2014,7(6):2147-2160.

[10]Marpu P R,Pedergnana M,Mura M D,et al.Automatic Generation of Standard Deviation Attribute Profiles for Spectral-Spatial Classification of Remote Sensing Data[J].IEEE Geoscience and Remote Sensing Letters,2013,10(2):293-297.

[11]Zhang Xiongmei,Yi Zhaoxiang,Tian Song,et al.Change Detection of SAR Images Using Morphologic Attribute Profile and Support Vector Machine[J].Editorial Office of Optics and Precision Engineeri,2014,22(10):2832-2839.[张雄美,易昭湘,田淞,等.结合形态学属性断面与支持向量机的合成孔径雷达图像变化检测[J].光学精密工程,2014,22(10):2832-2839.][12]Samat A,Du P,Liu S,et al.E 2 LMs:Ensemble Extreme Learning Machines for Hyperspectral Image Classification[J].Selected Topics in Applied Earth Observations and Remote Sensing IEEE Journal of,2014,7(4):1060-1069.

[13]Bazi Y,Alajlan N,Melgani F,et al.Differential Evolution Extreme Learning Machine for the Classification of Hyperspectral Images[J].IEEE Geoscience and Remote Sensing Letters,2014,11(6):1066-1070.

[14]Han M,Liu B.Ensemble ofExtreme Learning Machine for Remote Sensing Image Classification[J].Neurocomputing,2015,149:65-70.

[15]Chen C,Li W,Su H,et al.Spectral-Spatial Classification of Hyperspectral Image based on Kernel Extreme Learning Machine[J].Remote Sensing,2014,6(6):5795-5814.

[16]Wu Jun,Wang Shitong,Zhao Xin.Positive andNegative Fuzzy Rule and Image System,Extreme Learning Machine Classification[J].Journal of Image and Graphics,2011,16(8):1408-1417.[吴军,王士同,赵鑫.正负模糊规则系统,极限学习机与图像分类[J].中国图象图形学报,2011,16(8):1408-1417.]

[17]Ayerdi B,Marqués I,Graa M.Spatially Regularized Semisupervised Ensembles of Extreme Learning Machines for Hyperspectral Image Segmentation[J].Neurocomputing,2015,149:373-386.

[18]Du P,Xia J,Zhang W,et al.Multiple Classifier System for Remote Sensing Image Classification:A Review[J].Sensors,2012,12(4):4764-4792.

[19]Waske B,Sebastian V D L,Benediktsson J A,et al.Sensitivity of Support Vector Machines to Random Feature Selection in Classification of Hyperspectral Data[J].IEEE Transactions on Geoscience and Remote Sensing,2010,48(7):2880-2889.

[20]Gurram P,Kwon H.Ensemble Learning based on Multiple Kernel Learning for Hyperspectral Chemical Plume Detection[J].Proceedings of SPIE - The International Society for Optical Engineering,2010,7695(5):223-263.

[21]Wang Jie,Ja Yuheng.A Classification Method of Extreme Learning Machine based on AdaBoost[J].Journal of Zhengzhou University:Natural Science Edition,2014,46(2):55-58.[王杰,贾育衡.一种基于 AdaBoost 的极限学习机分类方法[J].郑州大学学报:理学版,2014,46(2):55-58].

[22]Huang G B,Zhu Q Y,Siew C K.Extreme Learning Machine:Theory and Applications[J].Neurocomputing,2006,70(1-3):489-501.

[23]Huang G B,Zhu Q Y,Siew C K.Extreme Learning Machine:A New Learning Scheme of Feedforward Neural Networks[C]//Neural Networks,2004.Proceedings 2004.IEEE International Joint Conference on,2004,2:985-990.

[24]Vapnik V N.Statistical Learning Theory[M].Beijing:Tsinghua University Press,2000.[Vapnik V N.统计学习理论的本质[M].北京:淸华大学出版社,2000.]

[25]Ardjani F,Sadouni K.Optimization of SVMMulticlass by Particle Swarm(PSO-SVM)[J].International Journal of Modern Education and Computer Science (IJMECS),2010,2(2):1-4.

[26]Dietterichl T G.Ensemble Learning[M].Mit Press:Handbook of Brain Theory and Neural Networks,2002:125-142.

[27]Dietterich T G.Ensemble Methods in Machine Learning[C]// International Workshop on Multiple Classifier Systems.Springer-Verlag,2015:1-15.

[1] 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.
[2] 陈洋波,张涛,窦鹏,董礼明,陈华. 基于SVM的东莞市土地利用/覆被自动分类误差来源与后处理[J]. 遥感技术与应用, 2017, 32(5): 893-903.
[3] 史飞飞,高小红,杨灵玉,何林华,贾伟. 基于HJ-1A高光谱遥感数据的湟水流域典型农作物分类研究[J]. 遥感技术与应用, 2017, 32(2): 206-217.
[4] 胡根生,吴问天,黄文江,梁栋,黄林生. 粒子群优化的最小二乘支持向量机在小麦白粉病监测中的应用[J]. 遥感技术与应用, 2017, 32(2): 299-304.
[5] 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.
[6] 林志垒,晏路明. 高光谱影像的BDT-SVM地物分类算法与应用[J]. 遥感技术与应用, 2016, 31(1): 177-185.
[7] 尹小君,张清,赵庆展,汪传建,宁川. 基于SVM的加工番茄细菌性斑点病氮素含量反演[J]. 遥感技术与应用, 2015, 30(3): 461-468.
[8] 周艳飞,张绘芳,李霞,杨帆,丁程锋. 基于高分辨遥感数据的胡杨与柽柳树冠提取[J]. 遥感技术与应用, 2015, 30(3): 510-517.
[9] 胡俊,谭琨,吴立新. 一种基于多项式逻辑回归高光谱影像分类方法的改进[J]. 遥感技术与应用, 2015, 30(1): 135-139.
[10] 邹利东,郭航,朱秀芳,王艳艳,郑学昌,王建阔. 设施农业空间分布信息自动提取方法研究[J]. 遥感技术与应用, 2014, 29(4): 669-674.
[11] 周颂洋,谭琨,吴立新. 基于邻域距离ISOMAP算法的高光谱遥感降维算法[J]. 遥感技术与应用, 2014, 29(4): 695-700.
[12] 刘朝相,宫兆宁,赵文吉. 基于SVM模型的妫水河叶绿素a浓度的遥感反演[J]. 遥感技术与应用, 2014, 29(3): 419-427.
[13] 任俊英,苏彩霞,曹永锋. 基于中间层特征的全极化SAR监督地物分类[J]. 遥感技术与应用, 2014, 29(2): 330-337.
[14] 张汉奎,黄 波,俞 乐. SVM-RFE高光谱数据波段选择中核函数的研究[J]. 遥感技术与应用, 2013, 28(5): 747-752.
[15] 林娜,杨武年. 基于核最小噪声分离变换的高光谱遥感影像特征提取研究[J]. 遥感技术与应用, 2013, 28(2): 245-251.