摘要: 针对退化图像空间分辨率较低的问题,提出了一种基于归一化方差的多分辨率图像融合方法。该方法首先对图像进行二层小波分解,然后根据小波系数的统计特征来定义归一化方差,并以此作为图像融合测度来对退化图像进行多分辨率特征融合,从而得到包含更多信息和有效特征的融合图像。试验结果表明,融合后的图像在保留细节信息和光谱信息上均能获得较好的效果。
关键词:
图像融合,
小波分析,
图像处理
Abstract: A major problem in using degenerative images is the low spatial resolution.In order to overcome that problem,an image fusion method based on the normalized square deviation is proposed.By using that model as the measure of feature selection,the new method carries out degenerative image fusion.The experiments show that the fused images obtain better results for further application in terms of both preserving spectral information and spatial information of the degenerative images.
中图分类号:
陈少辉;张秋文;王乘;周建中;. 基于归一化方差的退化图像融合研究[J]. , 2005, 31(23): 41-.
CHEN Shaohui,ZHANG Qiuwen,WANG Cheng,ZHOU Jianzhong. Degenerative Image Fusion Based on Normalized Square Deviation[J]. , 2005, 31(23): 41-.