基于随机森林的地理要素面向对象自动解译方法

顾海燕; 闫利; 李海涛; 贾莹

doi:10.13203/j.whugis20140102

基于随机森林的地理要素面向对象自动解译方法

doi: 10.13203/j.whugis20140102

1.
武汉大学测绘学院, 湖北武汉, 430079;
2.
中国测绘科学研究院, 北京, 100830;
3.
高德软件有限公司, 北京, 100080

基金项目: 国家科技支撑计划(2012BAH28B03);地理空间信息工程国家测绘地理信息局重点实验室开放基金(201101)。

详细信息

作者简介:
顾海燕,博士生,主要从事遥感影像智能解译方法与应用研究。guhy@casm.ac.cn

中图分类号: P208;P237

An Object-based Automatic Interpretation Method for Geographic Features Based on Random Forest Machine Learning

1.
School of Geodesy and Geomatics, Wuhan University, Wuhan 430079, China;
2.
Key Laboratory of Geo-informatics of State Bureau of Surveying and Mapping, Chinese Academy of Surveying and Mapping, Beijing 100830, China;
3.
AutoNavi Software Co. Ltd., Beijing 100080, China

Funds: The National Science & Technology Pillar Program, No. 2012BAH28B03; Key Laboratory of Geoinformatics of National Administration of Surveying, Mapping and Geoinformation, No.201101.

摘要: 面向地理对象影像分析(GEOBIA)技术取得了显著的进展,代表了遥感影像解译的发展范式,其主要目标是发展智能化分析方法。随机森林机器学习方法是一种相对新的、数据驱动的非参数分类方法,具有自动特征优选、自动模型构建等优势,为智能化分析提供了有效手段。充分利用GEOBIA及随机森林机器学习的优势,提出了基于随机森林的地理要素面向对象自动解译方法,阐述了随机森林面向对象分类方法的技术流程,为设计和实现该方法提供了详细指导,有助于指导用户优选特征和构建分类模型。通过与支持向量机分类的对比实验证明,该方法可以自动进行特征优选及分类模型的构建,利用较少的特征得到较高的分类精度,在不损失性能的前提下减少了计算量和内存使用,能够为大范围、大区域地理要素自动解译提供先验知识及自动化的手段。
- 面向地理对象影像分析 /
- 随机森林 /
- 分类模型 /
- 特征选择
Abstract: Geographic object-based image analysis (GEOBIA) techniques have recently seen considerable development in comparison to traditional pixel-based image analysis, representing a paradigm shift in remote sensing interpretation. The main aim is to incorporate and develop geographic-based intelligence. The random forest (RF) machine learning method is a relatively new, non-parametric, data-driven classification method that can supply intelligent means for feature selection and classification modelling. This paper presents a novel RF GEOBIA method for land-cover classification that makes full use of the advantages of GEOBIA and RF. A detailed RF GEOBIA workflow is proposed to guide the design and implementation of the method, and to guide experts during elaboration of feature selection and classification modelling. Theoretical and experimental results are compared with the support vector machine (SVM) approach, demonstrating that it is a robust and intelligent method for land-cover classification with wrapper feature selection and classification modelling. The RF GEOBIA method reduces the number of features required, computing time, and memory requirements, with no associated reduction in performance. It also provides a priori knowledge for further classification and supports large scale applications where "big data" is involved.
- geographic object-based image analysis (GEOBIA) /
- random forest /
- classification model /
- feature selection

[1]	Wiki. GEO-Object-Based Image Analysis[OL]. http://wiki.ucalgary.ca/page/GEOBIA,2015
[2]	Hay G J, Castilla G. Object-based Image Analysis:Strengths, Weaknesses, Opportunities and Threats (SWOT)[C].The 1st International Conference on Object-based Image Analysis (OBIA), Salzburg, Austria, 2006
[3]	Robertson L D, King D J. Comparison of Pixel-and Object-based Classification in Land-cover Change Mapping[J]. International Journal of Remote Sensing,2011,32(6):1505-1529
[4]	Li Haitao, Gu Haiyan, Han Yanshun, et al. Object-oriented Classification of High-resolution Remote Sensing Imagery Based on an Improved Colour Structure Code and a Support Vector Machine[J]. International Journal of Remote Sensing,2010,31(6):1453-1470
[5]	Blaschke T. Object Based Image Analysis for Remote Sensing[J]. ISPRS Journal of Photogrammetry and Remote Sensing,2010, 65:2-16
[6]	Definiens Imaging GmbH. Developer 8 Reference Book; Definiens Imaging GmbH[OL]. http://www.doc88.com/p-0963997861736.html
[7]	Baraldi A, Boschetti L. Operational Automatic Remote Sensing Image Understanding Systems:Beyond Geographic Object-based and Object-oriented Image Analysis (GEOBIA/GEOOIA). Part 1:Introduction[J]. Remote Sensing,2012,4:2694-2735
[8]	Baraldi A, Boschetti L. Operational Automatic Remote Sensing Image Understanding Systems:Beyond Geographic Object-based and Object-oriented Image Analysis (GEOBIA/GEOOIA). Part 2:Novel System Architecture, Information/Knowledge Representation, Algorithm Design and Implementation[J]. Remote Sensing,2012,4:2768-2817
[9]	Stumpf A, Kerle N. Object-oriented Mapping of Landslides Using Random Forests[J]. Remote Sensing of Environment,2011,115(10):2564-2577
[10]	Breiman L, Cutler A. Random Forests[OL].http://www.stat.berkeley.edu/users/breiman/RandomForests/cc_home.htm,2015
[11]	Breiman L. Random Forests[J]. Machine Learning, 2001,45(1):5-32
[12]	Rodriguez-Galiano V F, Chica-Olmo M, Abarca-Hernandez F,et al. Random Forest Classification of Mediterranean Land Cover Using Multi-seasonal Imagery and Multi-seasonal Texture[J]. Remote Sensing of Environment,2012,121:93-107
[13]	Guo L, Chehata N, Mallet C,et al. Relevance of Airborne LiDAR and Multispectral Image Data for Urban Scene Classification Using Random Forests[J].ISPRS Journal of Photogrammetry and Remote Sensing,2011,66(1):56-66
[14]	Stumpf A, Kerle N. Combining Random Forests and Object-oriented Analysis for Landslide Mapping from very High Resolution Imagery[J].Procedia Environmental Sciences, 2011,3:123-129
[15]	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):194-196(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):194-196)
[16]	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)
[17]	Li Haitao, Gu Haiyan, Han Yanshun, et al. An Efficient Multi-scale SRMMHR (Statistical Region Merging and Minimum Heterogeneity Rule) Segmentation Method for High-resolution Remote Sensing Imagery[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2009,2(2):67-73
[18]	Verikas A, Gelzinis A, Bacauskiene M. Mining Data with Random Forests:A Survey and Results of New Tests[J].Pattern Recognition, 2011,44(2):330-349
[19]	OpenCV Documentation.OpenCV 2.4.6 Online Documentation[OL].http://docs.opencv.org/modules/ml/doc/random_trees.html,2015
[20]	Huang C, Davis L S, Townshend J R G. An Assessment of Support Vector Machines for Land Cover Classification[J]. International Journal of Remote Sensing,2002,23(4):725-749
[21]	Pal M. Random Forest Classifier for Remote Sensing Classification[J]. International Journal of Remote Sensing, 2005,26(1):217-222

[1]	李秋萍, 刘逸诗, 巩诗瑶, 周素红, 卓莉, 陶海燕, 栾学晨. 基于居民出行活动特征的个体经济水平推断方法 . 武汉大学学报 ● 信息科学版, 2019, 44(10): 1575-1580. doi: 10.13203/j.whugis20170426
[2]	韩波, 李衡, 王志波, 崔晓晖. 气溶胶光学厚度估测中的LASSO特征选择方法 . 武汉大学学报 ● 信息科学版, 2018, 43(4): 536-541. doi: 10.13203/j.whugis20160014
[3]	刘坚, 李树林, 陈涛. 基于优化随机森林模型的滑坡易发性评价 . 武汉大学学报 ● 信息科学版, 2018, 43(7): 1085-1091. doi: 10.13203/j.whugis20160515
[4]	程诗尧, 梅天灿, 刘国英. 顾及结构特征的多层次马尔科夫随机场模型在影像分类中的应用 . 武汉大学学报 ● 信息科学版, 2015, 40(9): 1180-1187. doi: 10.13203/j .whu g is20130692
[5]	王云艳, 何楚, 涂峰, 陈东, 廖明生. 特征选择双层svm的融合算法用于极化sar图像分类 . 武汉大学学报 ● 信息科学版, 2015, 40(9): 1157-1162. doi: 10.13203/j .whu g is20140351
[6]	孙杰, 赖祖龙. 利用随机森林的城区机载LiDAR数据特征选择与分类 . 武汉大学学报 ● 信息科学版, 2014, 39(11): 1310-1313.
[7]	李雪, 舒宁, 李井冈, 刘小利. 基于特征贡献选择的遥感影像变化检测方法研究 . 武汉大学学报 ● 信息科学版, 2013, 38(10): 1158-1162.
[8]	刘梦玲, 陈嘉宇, 孙洪. Joint Boost特征选择的SAR信息可视化技术研究 . 武汉大学学报 ● 信息科学版, 2012, 37(10): 1240-1243.
[9]	何楚, 刘明, 许连玉, 刘龙珠. 利用特征选择自适应决策树的层次SAR图像分类 . 武汉大学学报 ● 信息科学版, 2012, 37(1): 46-49.
[10]	余晓敏, 湛飞并, 廖明生, 胡金星. 利用改进SEaTH算法的面向对象分类特征选择方法 . 武汉大学学报 ● 信息科学版, 2012, 37(8): 921-924.
[11]	刘海, 陈晓玲, 宋珍, 殷守敬. MODIS影像雪深遥感反演特征参数选择与模型研究 . 武汉大学学报 ● 信息科学版, 2011, 36(1): 113-116.
[12]	李祖传, 马建文, 张睿, 李利伟. 利用融合纹理与形态特征进行地震倒塌房屋信息自动提取 . 武汉大学学报 ● 信息科学版, 2010, 35(4): 446-450.
[13]	杨柳, 刘艳芳. 将微粒群和支持向量机用于耕地驱动因子选择的研究 . 武汉大学学报 ● 信息科学版, 2010, 35(2): 248-251.
[14]	薄树奎, 韩新超, 丁琳. 面向对象影像分类中分割参数的选择 . 武汉大学学报 ● 信息科学版, 2009, 34(5): 514-517.
[15]	张睿, 马建文. 一种SVM-RFE高光谱数据特征选择算法 . 武汉大学学报 ● 信息科学版, 2009, 34(7): 834-837.
[16]	叶志伟, 郑肇葆, 万幼川, 虞欣. 基于蚁群优化的特征选择新方法 . 武汉大学学报 ● 信息科学版, 2007, 32(12): 1127-1130.
[17]	金淑英, 李德仁, 龚健雅. 基于偏最小二乘回归的纹理特征线性组合 . 武汉大学学报 ● 信息科学版, 2006, 31(5): 399-402.
[18]	许才军. 随机模型误差对函数模型选择的影响 . 武汉大学学报 ● 信息科学版, 1992, 17(3): 36-42.
[19]	吴京航, 桂志鹏, 申力, 吴华意, 刘洪波, 李锐, 梅宇翱, 彭德华. 顾及格网属性分级与空间关联的人口空间化方法 . 武汉大学学报 ● 信息科学版, 0, 0(0): 0-0. doi: 10.13203/j.whugis20200379
[20]	毛丽君, 李明诗. GEE环境下联合Sentinel主被动遥感数据的国家公园土地覆盖分类 . 武汉大学学报 ● 信息科学版, 0, 0(0): 0-0. doi: 10.13203/j.whugis20200633

点击查看大图

计量

文章访问数: 1067
HTML全文浏览量: 35
PDF下载量: 699
被引次数: 0

姓名
邮箱
手机号码
标题
留言内容
验证码

留言板

基于随机森林的地理要素面向对象自动解译方法

doi: 10.13203/j.whugis20140102

作者简介:
顾海燕,博士生,主要从事遥感影像智能解译方法与应用研究。guhy@casm.ac.cn

An Object-based Automatic Interpretation Method for Geographic Features Based on Random Forest Machine Learning

计量

基于随机森林的地理要素面向对象自动解译方法

doi: 10.13203/j.whugis20140102

1. 武汉大学测绘学院, 湖北武汉, 430079;

2. 中国测绘科学研究院, 北京, 100830;

3. 高德软件有限公司, 北京, 100080

作者简介:
顾海燕,博士生,主要从事遥感影像智能解译方法与应用研究。guhy@casm.ac.cn

English Abstract

An Object-based Automatic Interpretation Method for Geographic Features Based on Random Forest Machine Learning

1. School of Geodesy and Geomatics, Wuhan University, Wuhan 430079, China;

2. Key Laboratory of Geo-informatics of State Bureau of Surveying and Mapping, Chinese Academy of Surveying and Mapping, Beijing 100830, China;

3. AutoNavi Software Co. Ltd., Beijing 100080, China

全文HTML

目录

留言板

基于随机森林的地理要素面向对象自动解译方法

doi: 10.13203/j.whugis20140102

作者简介: 顾海燕,博士生,主要从事遥感影像智能解译方法与应用研究。guhy@casm.ac.cn

An Object-based Automatic Interpretation Method for Geographic Features Based on Random Forest Machine Learning

计量

出版历程

基于随机森林的地理要素面向对象自动解译方法

doi: 10.13203/j.whugis20140102

1. 武汉大学测绘学院, 湖北武汉, 430079; 2. 中国测绘科学研究院, 北京, 100830; 3. 高德软件有限公司, 北京, 100080

作者简介: 顾海燕,博士生,主要从事遥感影像智能解译方法与应用研究。guhy@casm.ac.cn

English Abstract

An Object-based Automatic Interpretation Method for Geographic Features Based on Random Forest Machine Learning

1. School of Geodesy and Geomatics, Wuhan University, Wuhan 430079, China; 2. Key Laboratory of Geo-informatics of State Bureau of Surveying and Mapping, Chinese Academy of Surveying and Mapping, Beijing 100830, China; 3. AutoNavi Software Co. Ltd., Beijing 100080, China

全文HTML

目录

作者简介:
顾海燕,博士生,主要从事遥感影像智能解译方法与应用研究。guhy@casm.ac.cn

1. 武汉大学测绘学院, 湖北武汉, 430079;

2. 中国测绘科学研究院, 北京, 100830;

3. 高德软件有限公司, 北京, 100080

作者简介:
顾海燕,博士生,主要从事遥感影像智能解译方法与应用研究。guhy@casm.ac.cn

1. School of Geodesy and Geomatics, Wuhan University, Wuhan 430079, China;

2. Key Laboratory of Geo-informatics of State Bureau of Surveying and Mapping, Chinese Academy of Surveying and Mapping, Beijing 100830, China;

3. AutoNavi Software Co. Ltd., Beijing 100080, China