[1]依力亚斯江·努尔麦麦提,丁建丽,塔西甫拉提·特依拜,等.基于支持向量机分类的遥感土壤盐渍化信息监测[J].水土保持研究,2007,14(04):209-214,222.
　Ilyas·Nurmuhammat,DING Jian-li,Tashpolat·Tiyip,et al.Soil Salinization Monitoring Based on Support Vector Machine[J].Research of Soil and Water Conservation,2007,14(04):209-214,222.
点击复制

基于支持向量机分类的遥感土壤盐渍化信息监测

《水土保持研究》[ISSN:1005-3409/CN:61-1272/P] 卷: 14 期数: 2007年04期 页码: 209-214,222 栏目: 出版日期: 1900-01-01
Title:
Soil Salinization Monitoring Based on Support Vector Machine
作者:
依力亚斯江·努尔麦麦提1,2, 丁建丽1,2,3, 塔西甫拉提·特依拜1,2, 吐尔逊·艾山1,2
1. 新疆大学资源与环境科学学院, 乌鲁木齐, 830046;
2. 新疆大学绿洲生态教育部重点实验室, 乌鲁木齐, 830046;
3. 新疆大学理论经济学博士后流动站, 乌鲁木齐, 830046
Author(s):
Ilyas·Nurmuhammat1,2, DING Jian-li1,2,3, Tashpolat·Tiyip1,2, Tursun·Hasan1,2
1. College of Resources and Environment Science,Xinjiang University, Urumqi, 830046, China;
2. Key Laboratory of Oasis Ecology, Xinjiang University, Urumqi, 830046, China;
3. Post Doctoral Laboratory of Theoretical Economics, Xinjiang University, Urumqi, 830046, China
关键词:
支持向量机(SVM); 遥感图像; 盐渍化
Keywords:
support vector machine; remote sensing image; salinization
分类号:
TP79;S155.293
摘要:
遥感技术在大尺度土壤盐渍化检测方面有着宏观性、实时性、动态性等优势和广阔的应用前景,但是传统的遥感图像分类方法精度不高、分类效率较低和不确定性.提出了基于支持向量机(SupportVectorMachine,SVM)的分类方法,介绍了SVM算法的基本原理,通过支持向量机分类法与传统分类方法(最大似然法和最小距离法)在盐渍化信息提取结果上进行对比,表明基于SVM的遥感图像分类方法能够较好的检测土壤的盐渍化信息,分类总精度达到95.66%,比最大似然法和最小距离法分类精度(分别为91.54%和85.42%)更高,因此更适合于遥感图像分类和盐渍化信息检测.
Abstract:
The remote sensing technique have many superiorities such as offering wide regional coverage,providing up-to-date information,offering large archive of dynamic data for monitoring of large scale soil salinization,while the traditional extraction methods of the remote sensing image information has shortcomings including the low classification accuracy and classification efficiency,high indeterminacy.The classification method based on support vector machine(SVM) was put forward and the fundamental theory was introduced,moreover,the classification results of SVM classification with maximum likelihood classification and minimum distance classification were compared qualitatively and quantitatively in terms of classification accuracy.The results indicated that RS image classification based on SVM monitored the soil salinization more accurately than those of maximum likelihood classification and minimum distance classification,classification overall accuracy reached 95.66%,while the classification overall accuracy of maximum likelihood classification and minimum distance classification reached 91.54% and 85.42% respectively.The classification method based on SVM may be adapted to RS image classification and monitoring of soil salinization.

参考文献/References:

[1] 朱庭芸.灌区土壤盐渍化防治[M].北京:农业出版社,1992.32-38.
[2] Taylor G R,Mah A H.Characterization of saline soils using airborne radar imagery[J].Remote Sensing of Environment,1996,57(3):127-142.
[3] Verma K S,Saxena R K,et al.Remote sensing technique for mapping salt affected soils[J].International Journal of Remote Sensing,1994,15(9):1901-1914.
[4] Metternicht G I,Zinck J A.Remote sensing of soil salinity potentials and constraints[J].Remote Sensing of Environment,2003,85:1-20.
[5] 关元秀,刘高焕.区域土壤盐渍化遥感监测研究综述[J].遥感技术与应用,2001,16(1):40-44.
[6] Vapnik V N.Statistical learning theory [M].New York:Wiley Publishers Science,1998.
[7] Vapnik V N.The Nature of Statistical Learning Theory[M].New York:Springer,2000.
[8] Cortes C,Vapnik.V N.Support Vector Networks.Machine Learning,1995,20:1-25.
[9] Huang Hanpang,Liu Yihung.Fuzzy support vector machines for pattern recognition and data mining[J].International Journal of Fuzzy Systems,2002,4(3):826-835.
[10] 章毓晋.中国图像工程及当前的几个研究热点[J].计算机辅助设计与图形学学报,2002,14(6):489-500.
[11] Joachims T.Text categorization with support vector machines:learning with many relevant features [M].Berlin:Springer,In European Conference on Machine Learning(EC2ML),1998.137-142.
[12] 王宏漫,欧宗瑛.采用PCA/ICA特征和SVM分类的人脸识别[J].计算机辅助设计与图形学学报,2003,15(4):416-420.
[13] 朱晓霞,孙同景.基于二叉树和SVM的指纹分类[J].山东大学学报(工学版),2006,36(1):121-124.
[14] 何昕,刘重庆.基于支撑向量机的文本无关的说话人识别系统[J].计算机工程,2000,26(6):61-63.
[15] Qun Zhao,Jose C Principe.Support vector machines for SAR automatic target recognition[J].IEEE Transactions on Aero space and Electronic Systems,2001,37(2):643-654.
[16] Klaus-Robert Muller.An introduction to kernel-based learning algorithms[J].IEEE Transactions on Neural Networks,2001,12(2):181-201.
[17] Rolif,Fumira G.Support vector machines for remote sensing image classification [EB/OL].http://citeseer.nj.nec.com,2004-09-12.
[18] Weston J,Watkins C.Multi-class support vector machines[A].In:Verleysen M,ed.Proceedings of ESANN99[M].Brussels:D Facto Press,1999.
[19] 钱云,郝毓灵.新疆绿洲[M].乌鲁木齐:新疆人民出版社,2000.358-359.
[20] Kerrthi S S,Lin C I.Asymptotic behaviors of support vector machines with gaussian kernel[J].Neural Computation,2003,15(7):1667-1689.

相似文献/References:

[1]傅文杰,洪金益.基于支持向量机的滑坡灾害信息遥感图像提取研究[J].水土保持研究,2006,13(04):120.
　FU Wen-jie,HONG Jin-yi.Discussion on Application of Support Vector Machine Technique in Extraction of Information on Landslide Hazard from Remote Sensing Images[J].Research of Soil and Water Conservation,2006,13(04):120.

备注/Memo

收稿日期:2007-4-13。
基金项目:国家自然科学基金项目(编号:40261006);自治区高校科研计划项目(编号:XJEDU2004I06,XJEDU2005I07);教育厅创新研究群体基金项目(编号:XJEDU2004G04)
作者简介:依力亚斯江·努尔麦麦提(1982- ),男,硕士,主要从事干旱区资源环境与遥感应用研究.
通讯作者:塔西甫拉提·特依拜(1958- ),男,教授.E-mail:tash@xju.edu.cn

更新日期/Last Update: 1900-01-01