激光与光电子学进展, 2016, 53 (2): 021001, 网络出版: 2015-12-25
基于多光谱图像超分辨率处理的遥感图像融合 下载: 1245次
Remote Sensing Image Fusion Based on Multispectral Image Super-Resolution
遥感 图像融合 图像超分辨 YUV 变换 静态小波变换 remote sensing image fusion image super-resolution YUV transform stationary wavelet transform
摘要
传统遥感图像融合方法没有充分利用低分辨率多光谱图像本身的空间信息。针对这一问题,提出了一种基于低分辨率多光谱图像超分辨率处理的遥感图像融合方法。通过对低分辨率多光谱图像进行基于稀疏表示的图像超分辨率处理,在保持其光谱信息的基础上增强其空间信息;利用静态小波变换对增强的多光谱图像的亮度分量Y 和全色图像进行融合;由YUV 逆变换得到融合多光谱图像。在真实遥感图像上进行的实验结果表明该算法能有效提高融合图像的空间细节信息,同时保留了光谱信息,对比实验验证了该方法的优越性。
Abstract
The traditional remote sensing image fusion methods is not make good use of spatial information of the low-resolution multispectral images. To solve this problem, a novel remote sensing image fusion method based on low-resolution multispectral image super-resolution processing is proposed. The low-resolution multispectral image is processed by super- resolution based on sparse representation, which enhances spatial information while maintaining the spectral information. The intensity component Y of the enhanced multispectral image and the panchromatic image are fused using the stationary wavelet transform. The fused multispectral image is obtained via inverse YUV transform. The experimental results carried out on real remote sensing images show that the proposed algorithm is able to improve spatial details of the fused image effectively while maintaining the spectral information. The effectiveness of the proposed method is also verified by the comparable experiments.
杨超, 杨斌, 黄国玉. 基于多光谱图像超分辨率处理的遥感图像融合[J]. 激光与光电子学进展, 2016, 53(2): 021001. Yang Chao, Yang Bin, Huang Guoyu. Remote Sensing Image Fusion Based on Multispectral Image Super-Resolution[J]. Laser & Optoelectronics Progress, 2016, 53(2): 021001.