我们的网站为什么显示成这样?

可能因为您的浏览器不支持样式,您可以更新您的浏览器到最新版本,以获取对此功能的支持,访问下面的网站,获取关于浏览器的信息:

|本期目录/Table of Contents|

基于Fisher判别的层次分类法的森林遥感影像分类(PDF)

《西南林业大学学报》[ISSN:2095-1914/CN:53-1218/S]

期数:
2017年04期
页码:
175-182
栏目:
出版日期:
2017-06-30

文章信息/Info

Title:
Fisher Discriminant Based Hierarchical Method for Forest Remotely Sensed Data Classification
文章编号:
2095-1914(2017)04-0175-08
作者:
杜靖媛12葛宏立2路伟2孟森2
1. 浙江农林大学信息工程学院,浙江 临安 311300;
2. 浙江省森林生态系统碳循环与固碳减排重点实验室,浙江 临安 311300
Author(s):
Du Jingyuan12 Ge Hongli2 Lu Wei2 Meng Sen2
1. College of Information Engineering, Zhejiang A & F University, Lin′an Zhejiang 311300, China;
2. Zhejiang Provincial Key Laboratory of Carbon Cycling in Forest Ecosystems and Carbon Sequestration, Lin′an Zhejiang 311300, China
关键词:
Fisher判别法层次分类法影像分类最大似然支持向量机随机森林
Keywords:
Fisher discriminant analysis hierarchical classification image classification maximum likelihood SVM random forest
分类号:
S757.2
DOI:
10.11929/j.issn.2095-1914.2017.04.025
文献标识码:
A
摘要:
通过Fisher判别法得到的混淆矩阵计算分类评价指标构建层次分类树模型,提出了一种基于Fisher判别的遥感影像森林地类层次分类法。利用杭州市部分地区的森林资源清查样地数据和Landsat8遥感影像数据进行试验,并与最大似然分类、支持向量机和随机森林分类的结果进行比较。结果表明:基于Fisher判别的层次分类法总分类精度为79.45%,比最大似然分类、支持向量机和随机森林分别高了12.33%、21.00%和10.50%,kappa系数为0.756 8,比最大似然法、支持向量机和随机森林分别高了0.145 5、0.256 4和0.126 4;基于Fisher判别分析方法的层次分类法模型中,层次分类树的节点依次是建设用地—水体—农地—竹林—阔叶林—针叶林—针阔混交林;基于Fisher判别的层次分类法中一个模型只能分出一个类别。
Abstract:
Fisher discriminant analysis has been applied to obtain the confusion matrix, which was used to calculate the classification evaluation index, and hierarchical classification tree model was built in accordance with these index. Thus, Fisher hierarchical classification method based on Fisher discriminant analysis was proposed to forest remote sensed images. A part of sample data from forest resources inventory and Landsat8 remote sensed image data of Hangzhou city were used to test this method, furthermore the results was compared to those of maximum likelihood classification method, support vector machine and random forest classification method. Results showed that the total classification accuracy of Fisher based hierarchical classification is 79.45%, which is higher than that of maximum likelihood classification, support vector machine and random forest by 12.33%, 21.00% and 10.50%, respectively. Kappa coefficient was 0.756  8, which meant it was higher than those from maximum likelihood method, support vector machine and random forest whose values were 0.145 5, 0.256 4 and 0.126 4, respectively. In the hierarchical classification model based on Fisher discriminant analysis, the gradation of hierarchical classification tree are construction land→water→farmland→bamboo grove→broad-leaved forest→coniferous forest→conifer-broadleaf forest. In the hierarchical classification based on Fisher discrimination, a model can only distinguish 1 category.

参考文献/References

[1]李乡儒, 胡占义, 赵永恒. 基于Fisher判别分析的有监督特征提取和星系光谱分类[J]. 光谱学与光谱分析, 2007, 27(9): 1898-1901.
[2]陈科, 张保明, 谢明霞. 改进Fisher判别分类的遥感影像变化检测[J]. 测绘科学, 2010, 35(4): 160-162.
[3]李文, 苗夺谦, 卫志华, 等. 基于阻塞先验知识的文本层次分类模型[J]. 模式识别与人工智能, 2010, 23(4): 456-463.
[4]陆彦婷, 陆建峰, 杨静宇. 层次分类方法综述[J]. 模式识别与人工智能, 2013, 26(12): 1130-1137.
[5]Lu D S, Scott Hetrick, Emilio Moran, et al. Application of time series landsat images to examining land-use/Land-cover dynamic change[J]. Photogrammetric Engineering & Remote Sensing, 2012, 78(7): 747-755.
[6]Chen Yangchi, Crawford M M, Ghosh J. Intergrating support vector machines in a hierarchical output space decomposition framework[J]. Proc of the IEEE International Geoscience and Remote Sensing Symposium, 2002, 5(2): 210-220.
[7]Griffin G, Perona P. Learning and using taxonomies for fast visual categorization[J]. Proc of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Alaska, USA, 2008: 1-8.
[8]熊云波, 李荣陆, 胡运发. 基于混淆矩阵的层次结构构造方法比较[J]. 模式识别与人工智能, 2007, 20(2): 205-210.
[9]Kumar S, Chosh J, Crawford M M. Hierarchical fusion of multiple classifiers for hyperspectral data analysis[J]. Pattern Analysis and Applications, 2002, 5(2): 210-220.
[10]Smith M R, Bandfield J L, Cloutis E A, et al. Hydrated silica on mars: combined analysis with near-infrared and thermalinfrared spectroscopy[J]. Icarus, 2013, 223(2): 633-648.
[11]宋亮, 刘善军, 虞茉莉, 等. 基于可见-近红外和热红外光谱联合分析的煤和矸石分类方法研究[J]. 光谱学与光谱分析, 2017, 37(2): 416-421.
[12]余颂, 陈善雄, 余飞, 等. 膨胀土判别与分类的Fisher判别分析方法[J]. 岩土力学, 2007, 28(3): 499-504.
[13]吴学军. 城市TM遥感图像分类方法研究[D]. 广西: 广西师范大学, 2007.
[14]John R, Jensen. 遥感数字影像处理导论[M]. 陈晓玲, 龚威, 李平湘, 等译. 北京: 机械工业出版社, 2007: 359-361.
[15]李石华, 王金亮, 毕艳, 等. 遥感图像分类方法研究综述[J]. 国土资源遥感, 2005(2): 1-6.
[16]沈明霞, 何瑞银, 丛静华. 基于ETM+遥感影像的森林植被信息提取方法研究[J]. 南京林业大学学报 (自然科学版), 2007, 31(6): 113-116.
[17]刘怀鹏, 安慧君. 利用最大似然法识别呼和浩特市绿化树种[J]. 东北林业大学学报, 2014, 42(7): 157-161.
[18]Breiman L. Random forests[J]. Machine Learning, 2001, 45(1): 5-32.
[19]张晓羽, 李凤日, 甄贞, 等. 基于随机森林模型的陆地卫星-8遥感影像森林植被分类[J]. 东北林业大学学报, 2016, 44(6): 53-57.
[20]刘向东, 陈兆乾. 一种快速支持向量机分类算法的研究[J]. 计算机研究与发展, 2004, 41(8): 1327-1332.

备注/Memo

备注/Memo:
收稿日期:2017-02-16; 修回日期:2017-03-21
基金项目:国家自然科学基金项目(41371411)资助。
第1作者:杜靖媛(1990—),女,硕士生。研究方向:森林资源遥感监测与信息技术。Email: 1138446068@qq.com
通信作者:葛宏立(1960—),男,博士,教授。研究方向:森林数学模型技术、统计与抽样技术、遥感技术在森林资源监测中的应用。Email: jhghlhxl@163.com
更新日期/Last Update: