基于环境卫星数据的沿海滩涂地物类型分类的随机森林方法

doi:10.11873/j.issn.1004-0323.2016.6.1107

遥感技术与应用

2016, Vol. 31

Issue (6): 1107-1113 DOI: 10.11873/j.issn.1004-0323.2016.6.1107

数据与图像处理

基于环境卫星数据的沿海滩涂地物类型分类的随机森林方法

王艳楠¹,王健健¹,龚健新¹,袁帅¹,刘辉¹,罗文^2,3



(1.南京师范大学虚拟地理环境教育部重点实验室,江苏南京 210023;
2.江苏省地理环境演化国家重点实验室培育建设点,江苏南京 210023;
3.江苏省地理信息资源开发与利用协同创新中心,江苏南京 210023)

Detection of Coastal Tidal Flat Using the Rrandom Forests Method and HJ Satellite Images

Wang Yannan¹,Wang Jianjian¹,Gong Jianxin¹,Yuan Shuai¹,Liu Hui¹,Luo Wen^2,3

(1.Key Laboratory of Virtual Geographic Environment (Nanjing Normal University),
Ministry of Education,Nanjing 210023,China; 2.State Key Laboratory Cultivation Base
of Geographical Environment Evolution (Jiangsu Province),Nanjing 210023,China;
3.Jiangsu Center for Collaborative Innovation in Geographical Information Resource
Development and Application,Nanjing 210023,China)

全文: PDF(4700 KB)

摘要：

基于环境卫星遥感影像,利用随机森林数据自适应探测特性,构建了面向沿海滩涂区域的地物类型分类方法,并以大丰为例进行了实证分析,探讨了基于随机森林方法的滩涂分类的参数选择与精度分析。结果显示:基于随机森林的沿海滩涂地物分类方法分类精度较传统监督分类(平行管道法\,最小距离法和最大似然法)精度提高明显,且具有很好的分类稳定性。


关键词： 随机森林; 遥感影像; 沿海滩涂; 地物分类

Abstract:

Taking advantage of the adaptive detection feature of the random forests method,a classification method for coastal tidal flat area has been developed on basis of the HJ satellite images.Two typical areas covering coastal tidal flats,namely the ecological demonstration area of coastal tidal flats and the elk reserve,in the city of Dafeng,Jiangsu Province were selected in this study to showcase the classification effectiveness.The parameter settings and its accuracy were discussed.The results show that the random forests method performs obviously better than other traditional supervised classification methods such as parallelepiped,minimum distance,and maximum likelihood methods,and presents high stability in classification performance.


Key words: Random forests Remote sensing image Coastal tidal flat Classification

收稿日期: 2015-11-24 出版日期: 2016-12-30

TP 75

基金资助:

国家科技支撑项目(2012BAC07B01),国家自然科学基金项目(41571380),江苏省测绘地理信息科研项目资助(JSCHKY201509)。

通讯作者: 罗文(1986-),男,湖北荆州人,博士后,主要从事GIS方面的研究。Email:luow1987@163.com。

作者简介: 王艳楠(1992-),女,北京人,硕士研究生,主要从事环境遥感方面的研究。Email:wangyn_1992@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
	王艳楠
	王健健
	龚健新
	袁帅
	刘辉
	罗文

引用本文:

王艳楠,王健健,龚健新,袁帅,刘辉,罗文. 基于环境卫星数据的沿海滩涂地物类型分类的随机森林方法[J]. 遥感技术与应用, 2016, 31(6): 1107-1113.

Wang Yannan,Wang Jianjian,Gong Jianxin,Yuan Shuai,Liu Hui,Luo Wen. Detection of Coastal Tidal Flat Using the Rrandom Forests Method and HJ Satellite Images. Remote Sensing Technology and Application, 2016, 31(6): 1107-1113.

链接本文:

http://www.rsta.ac.cn/CN/10.11873/j.issn.1004-0323.2016.6.1107 或 http://www.rsta.ac.cn/CN/Y2016/V31/I6/1107

［1］Wu Dingding.The Application of RS Technique in the Extraction of Coastal Beach［J］.Geomatics & Spatial Information Technology,2012,35(11):8384.［吴定定.遥感技术在沿海滩涂提取中的应用［J］.测绘与空间地理信息,2012,35(11):8384.］

［2］Fu Jun,Gu Dongqi.Extracting Method in Tidal Flat based on Decision Fusion of Spectrum and Texture［J］.Marne Environmental Science,2008,30(5):405408.［付军,谷东起.基于光谱与纹理特征融合的滩涂信息提取方法［J］.海洋环境科学,2008,30(5):405408.］

［3］Guan Yujuan,Zhang Liquan.Application of Intertidal Wetlands Classified by Image Fusion Technique［J］.Marine Environmental Science,2008,27(6):647651.［管玉娟,张利权.影像融合技术在滩涂湿地植被分类中的应用［J］.海洋环境科学,2008,27(6):647651.］

［4］Liu Yi,Du Peijun,Zheng Hui,et al.Classification of China Small Satellite Remote Sensing Image based on Random Forests［J］.Science of Surveying and Mapping,2012,37(4):194196.［刘毅,杜培军,郑辉,等.基于随机森林的国产小卫星遥感影像分类研究［J］.测绘科学,2012,37(4):194196.］

［5］Jia Minming,Wang Zongming,Zhang Bai,et al.Land Cover Classification of Compositing HJ1 and MODIS Data based on Objectbased Method［J］.Geomatics and Information Science of Wuhan University,2014,39(3):305309.［贾明明,王宗明,张柏,等.综合环境卫星与MODIS数据的面向对象土地覆盖分类方法［J］.武汉大学学报,2014,39(3):305309.］

［6］Zhang Ce,Zang Shuying,Jin Zhu,et al.Remote Sensing Classification for Zhalong Wetlands based on Support Vector Machine［J］.Wetland Science,2012,10(2):206213.［张策,臧淑英,金竺,等.基于支持向量机的扎龙湿地眼高分类研究［J］.湿地科学,2012,10(2):206213.］

［7］Xie Jing,Wang Zongming,Mao Dehua,et al.Remote Sensing Classification of Wetlands Using Objectoriented Method and Multiseason HJ1 Images——A Case Study in the Sanjiang Plain North of the Wandashan Mountain［J］.Wetland Science,2012,10(4):429438.［谢静,王宗明,毛德华,等.基于面向对象方法和多时相HJ1影像的湿地遥感分类——以完达山以北三江平原为例［J］.湿地科学,2012,10(4):429438.］

［8］Sun Fei,Xu Shiwu,Zhang Yan.Classification Strategy for Remote Sensing Data based on Positive Training Samples［J］.Science of Surveying and Mapping,2016,41(2):133139.［孙飞,徐世武,张岩.利用目标类样本的遥感分类策略［J］.测绘科学,2016,41(2):133139.］

［9］Li Hengkai,Wu Lixin,Li Fashuai.Optimal Fractal Band Selection on HJ1 CCD Image for Land Use Classification［J］.Journal of Remote Sensing,2013,17(6):15721585.［李恒凯,吴立新,李发帅.面向土地利用分类的HJ1 CCD影像最佳分形波段选择［J］.遥感学报,2013,17(6):15721585.］

［10］Pan Chen,Du Peijun,Zhang Hairong.Decision Tree Classification and Its Application in Processing of Remote Sensing Images［J］.Science of Surveying and Mapping,2008,33(1):208211.［潘琛,杜培军,张海荣.决策树分类方法及其在遥感图像处理中的应用［J］.测绘科学,2008,33(1):208211.］

［11］Breiman L.Random Forests［J］.Machine Learning,2001,45(1):532.

［12］Liaw A,Wiener M.Classification and Regression by Random Forest［J］.R News,2002,2(3):1822.

［13］ZhangYing,Lin Ying,Lü Danju.Application of Ensemble Learning in Remote Sensing Image Classification［J］.Computer & Digital Engineering,2013,283(5):697699.［张雁,林英,吕丹桔.集成学习在遥感分类中的应用［J］.计算机与数字工程,2013,283(5):697699.］

［14］Watts J D,Powell S I,Lawrence R L,et al.Improved Classification of Conservation Tillage Adoption Using High Temporal and Synthetic Satellite Imagery［J］.Remote Sensing of Environment,2011,115:6675.

［15］Dong Shishi,Huang Zhexue.A Brief Theoretical Overview of Random Forests［J］.Journal of Integration Technology,2013,2(1):17.［董师师,黄哲学.随机森林理论浅析［J］.集成技术,2013,2(1):17.］

［16］Lei Zhen.Random Forest and Its Application in Remote Sensing［D］.Shanghai:Shanghai Jiaotong University,2012:1631.［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,2012:1631.］

［17］Huang Jinlou,Wang Rusong,Yang Wenrui,et al.Exploration on the Exploitation and Utilization of Coastal Zone Beach Land based on Ecological Service Function in Dafeng City［J］.Journal of Anhui Agricultrue Science,2008,36(21):92159218.［黄锦楼,王如松,阳文瑞,等.基于生态服务功能的大丰市海岸带滩涂土地的开发与利用［J］.安徽农业科学,2008,36(21):92159218.］

［18］Ma Yue,Jiang Qigang,Meng Zhiguo,et al.Classification of Land Use in Farming Area based on Random Forest Algorithm［J］.Transactions of the Chinese Society for Agricultural Machinery,2015,47(1):298303.［马玥,姜琦刚,孟治国,等.基于随机森林算法的农耕区土地利用分类研究［J］.农业机械学报,2015,47(1):298303.］

［19］Wolpert D H,Macready W G.An Efficient Method to Estimate Bagging’s Generalization Error［J］.Machine Learning,1999,35(1):4155.

［20］Wei Yuchun,Tang Guoan,Yang Xin.A Tutorial on Processing of Remote Sensing Digital Images［M］.Beijing:Science Press,2007.［韦玉春,汤国安,杨昕.遥感数字图像处理教程［M］.北京:科学出版社,2007.］

[1]	陈伟民，张凌，宋冬梅，王斌，丁亚雄，许明明，崔建勇. 基于AdaBoost改进随机森林的高光谱图像地物分类方法研究[J]. 遥感技术与应用, 2018, 33(4): 612-620.
[2]	刘振波，邹娴，葛云健，陈健，曹雨濛. 基于高分一号WFV影像的随机森林算法反演水稻LAI[J]. 遥感技术与应用, 2018, 33(3): 458-464.
[3]	周子勇. 运用经验模态分解和压缩感知方法进行遥感影像超分辨率重建[J]. 遥感技术与应用, 2018, 33(1): 96-102.
[4]	周为峰，曹利，李小恕，程田飞. 沿海牡蛎养殖的WorldView-2影像融合方法评价[J]. 遥感技术与应用, 2018, 33(1): 103-109.
[5]	林齐根，邹振华，祝瑛琦，王瑛. 基于光谱、空间和形态特征的面向对象滑坡识别[J]. 遥感技术与应用, 2017, 32(5): 931-937.
[6]	吕利利，颉耀文，董龙龙. 基于不同地形校正模型的影像反射率对比分析[J]. 遥感技术与应用, 2017, 32(4): 751-759.
[7]	吕利利，颉耀文，黄晓君，张秀霞，李汝嫣. 基于CART决策树分类的沙漠化信息提取方法研究[J]. 遥感技术与应用, 2017, 32(3): 499-506.
[8]	朱济帅,尹作霞,谭琨,王雪,李二珠,杜培军. 基于空间邻域信息的高光谱遥感影像半监督协同训练[J]. 遥感技术与应用, 2016, 31(6): 1122-1130.
[9]	吴阿丹,郭建文. 遥感影像Web共享关键技术应用研究[J]. 遥感技术与应用, 2016, 31(6): 1209-1214.
[10]	赵玉,王红,张珍珍. 基于遥感光谱和空间变量随机森林的黄河三角洲刺槐林健康等级分类[J]. 遥感技术与应用, 2016, 31(2): 359-367.
[11]	李盛阳,于海军,韩洁,黑保琴. 基于三维地球的海量遥感影像高效可视化管理系统的设计与实现[J]. 遥感技术与应用, 2016, 31(1): 170-176.
[12]	毛召武,程结海,袁占良. 顾及面积和位置差异的高分遥感影像分割质量评价方法[J]. 遥感技术与应用, 2016, 31(1): 186-193.
[13]	张文博,覃志豪,刘含海. 基于分形理论的土地利用类型变化研究—以广州番禺为例[J]. 遥感技术与应用, 2015, 30(5): 952-958.
[14]	程结海,柏延臣. 一种高空间分辨率遥感影像信噪比测定方法[J]. 遥感技术与应用, 2015, 30(3): 469-475.
[15]	费鲜芸,王婷,魏雪丽. 基于多尺度分割的遥感影像滨海湿地分类[J]. 遥感技术与应用, 2015, 30(2): 298-303.

Viewed

Full text

Abstract

Cited

Shared

Discussed