基于周至县农耕区Landsat8 OLI影像的融合算法和分类研究

详细信息    查看全文 | 推荐本文 |

英文篇名：Study on fusion algorithm and classification based on Landsat8OLI Image 作者：屈贵兰 ; 张彬 英文作者：QU Gui-lan;ZHANG Bin;College of Urban and Environmental Science,Northwestern University; 关键词：OLI影像 ; 融合算法 ; 支持向量机 英文关键词：OLI images;;fusion algorithm;;classification;;support vector machine 中文刊名：ZGRZ 英文刊名：China Population,Resources and Environment 机构：西北大学城市与环境学院; 出版日期：2016-12-31 出版单位：中国人口·资源与环境 年：2016 期：v.26;No.195 基金：国家自然科学基金项目"基于角色的虚拟地理环境群体感知与协同控制模型研究"(批准号:41271402) 语种：中文; 页：ZGRZ2016S2106 页数：3 CN：S2 ISSN：37-1196/N 分类号：455-457

摘要

本文以陕西省周至县农耕区的OLI影像为研究对象,分析了新型传感器多光谱波段之间的统计特征和相关性,运用分融合、小波融合和PCA—小波变换三种融合算法对OLI影像进行融合实验,研究其算法对影像的适应性,再基于融合效果好的影像采用支持向量机和BP神经网络进行分类实验,并根据地表感兴趣区域进行精度评价。结果表明,依据典型地物的分类效果来看,主成分融合算法更适合OLI影像,同时,支持向量机的分类总体精度和Kappa系数均高于BP神经网络,因此支持向量机的分类算法更适合PCA主成分融合影像。
        Taking OLI image of farming area in the northern of Zhouzhi County,Shaanxi Province as the research object,statistical characteristics and correlation between multispectral bands of OLI image Analysis were studied. PCA fusion,wavelet fusion and PCA-wavelet transform were used to study the adaptability of fusion algorithm of OLI image,and the accuracy evaluation were compared by the typical features. Land use classification of high fusion accuracy image were studied by support vector machine and BP neural network,and the classification accuracy was evaluated by the error matrix. The results showed that,according to the classification accuracy of typical features,principal component fusion algorithm was more suitable for OLI image. The overall accuracy and Kappa coefficient of support vector machine classification were higher than that of BP neural network. Therefore,the classification algorithm of support vector machine was helpful to improve the classification accuracy of fusion image based on PCA.

引文

[1]王乐,牛雪峰,魏斌,等.遥感影像融合质量评价方法研究[J].测绘通报,2015(2):77-79.
    [2]SIMONE G,FARINA A,MORABITO F C,et al.Image fusion techniques for remote sensing applications[J].Information fusion,2002,3(1):3-15.
    [3]KAUR N,MADHU B,HARSIMRAN K.Review on:image fusion using wavelet and curvelet transform[J].International journal of computer science&information technoloy,2014.
    [4]邓磊,李京,陈云浩,等.几种小波融合方法在遥感影像融合中的应用与比较[J].遥感信息,2007(6):23-27.
    [5]李俊杰,何隆华,戴锦芳,等.遥感影像像元级融合方法与试验评价分析[J].地球信息科学,2008,10(1):128-134.
    [6]赖格英,曾祥贵,刘影,等.基于ETM和图像融合的优势植被冠层叶面积指数和消光系数的遥感反演[J].遥感技术与应用,2013,28(4):697-706.
    [7]李存军,刘良云,王纪华,等.基于主成分融合信息失真的城市IKONOS影像阴影自动提取研究[J].武汉大学学报:信息科学版,2008,33(9):947-950.
    [8]申怀飞,胡楠,丁圣彦,等.基于主成分融合的遥感影像分区分类方法探讨[J].河南大学学报(自然科学版),2007,37(3):294-299.
    [9]吴俣,余涛,谢东海,等.高分辨率多光谱遥感图像的自动配准[J].红外与激光工程,2012,41(12):3285-3290.