检索规则说明:AND代表“并且”;OR代表“或者”;NOT代表“不包含”;(注意必须大写,运算符两边需空一格)
检 索 范 例 :范例一: (K=图书馆学 OR K=情报学) AND A=范并思 范例二:J=计算机应用与软件 AND (U=C++ OR U=Basic) NOT M=Visual
出 处:《遥感信息》2010年第6期14-18,共5页Remote Sensing Information
摘 要:ARSIS概念下的遥感影像融合中多尺度分析方法层出不穷,经典的小波分析方法因为方向有限且是各向同性的,因此不能"稀疏"表达二维图像中的线奇异以及边缘方向,在融合影像中引入一定程度模糊。曲波(Curvelet)作为一种新的多尺度分析方法比小波分析更加适合分析二维图像中的曲线或直线状边缘特征,而且具有更高的逼近精度和更好的稀疏表达能力。本文利用第二代Curvelet作为多尺度分析工具,并提出一种新的分量融合模型,粗尺度分量线性加权融合,不同层次细尺度分量采用不同融合模型,1°和2°层对全色影像A分量进行调整使其与多光谱影像B分量具有相同的概率密度函数;3°层基于对全色影像和多光谱影像分量的相关性判断来调整融合分量。利用该方法对SPOT-5影像进行了融合实验,并对比小波融合方法和传统模型下的Curvelet融合方法。实验结果表明Curvelet方法无论在光谱保真度还是空间细节增强方面都优于小波方法,且本文提出的融合模型提高了传统模型下Curvelet的融合效果。Multiscale analysis methods based on the ARSIS concept have been emerging,in which the classical wavelet analysis appears not competent enough to provide the approximation precision and sparsity description of 2-D image edge due to its isotropic nature and limited direction,leading to some blur images.Curvelet,as a new multiscale analysis algorithm,is more appropriate for analysis of the image edge fusion method such as curve and line characteristics than wavelet.The second-generation curvelet with a brand new fusion model is proposed in this paper.The new fusion model involves both the linear weight sum model for the coarse coefficient and exclusive models for different scale fine coefficients,the adjustment of pan image A coefficients at 1°and 2°aiming at the same probability density function with the counterpart of B as well as that at 3°based on the determination of coefficient correlation between A and B.This method is applied to SPOT-5 image fusion and compared with the results produced by wavelet analysis and another curvelet analysis with the traditional fusion model.The fusion results demonstrate the advantage of curvelet over wavelet analysis both in spectral preservation and spatial detail enhancement.Meanwhile,the fusion model proposed by this paper can improve the performance of curvelet analysis with the traditional fusion model.
关 键 词:ARSIS CURVELET变换 Wavelet变换 融合模型 SPOT
分 类 号:TP79[自动化与计算机技术—检测技术与自动化装置]
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在链接到云南高校图书馆文献保障联盟下载...
云南高校图书馆联盟文献共享服务平台 版权所有©
您的IP:216.73.216.205