摘要
针对传统的全色锐化融合中细节信息提取不准确、融合光谱精度不高等问题,结合非下采样轮廓波变换(nonsubsampled Contourlet transform,NSCT)的多尺度多方向分解特点和脉冲耦合神经网络(pulse coupled neural networks,PCNN)脉冲同步发放特性等优点,提出了一种基于NSCT和PCNN的遥感图像全色锐化算法。该算法首先采用NSCT变换提取全色图像细节特征,然后将该特征注入PCNN模型获得的非规则分割区域,最终采用统计加权方式获取高分辨率的多光谱遥感图像锐化融合结果。采用WorldView-2和GF-2高空间分辨率遥感图像数据集实验结果表明,该算法在细节保持和光谱一致性等评价指标上均优于其他遥感图像融合算法,验证了该算法有效性。
Conventional pansharpening fusion methods suffer inaccurate extraction of details and low spectrum fusion accuracy.This study proposed a pansharpening algorithm of remote sensing images based on nonsubsampled Contourlet transform(NSCT)and pulse coupled neural networks(PCNN)by combining the multi-scale and-directional decomposition characteristics of NSCT and the pulse synchronous emission characteristics of PCNN.The process of this pansharpening algorithm is as follows:first,the details of panchromatic images were extracted through NSCT;then,the extracted detail features were injected into the irregular segmentation regions obtained using the PCNN model;finally,the sharpening fusion results of high-resolution multispectral remote-sensing images were obtained through statistical weighting.As corroborated by the experimental results of WorldView-2 and GF-2 data sets,the pansharpening algorithm outperforms other remote sensing image fusion algorithms in detail preservation and spectral consistency,verifying its effectiveness.
作者
徐欣钰
李小军
赵鹤婷
盖钧飞
XU Xinyu;LI Xiaojun;ZHAO Heting;GAI Junfei(Faculty of Geomatics,Lanzhou Jiaotong University,Lanzhou 730070,China;National-Local Joint Engineering Research Center of Technologies and Applications for National Geographic State Monitoring,Lanzhou 730070,China;Gansu Provincial Engineering Laboratory for National Geographic State Monitoring,Lanzhou 730070,China)
出处
《自然资源遥感》
CSCD
北大核心
2023年第3期64-70,共7页
Remote Sensing for Natural Resources
基金
国家自然科学基金项目“基于脉冲耦合神经网络的高光谱遥感图像融合方法研究”(编号:41861055)
中国博士后基金项目(编号:2019M653795)
兰州交通大学(编号:201806)优秀平台共同资助
关键词
脉冲耦合神经网络
非下采样轮廓波变换
遥感图像融合
多光谱遥感图像
pulse coupled neural network
nonsubsampled Contourlet transform
remote sensing image fusion
multispectral remote sensing image
