摘要
提出了一种改进的基于小波变换的图像融合算法。对待融合图像进行小波变换,从而得到低频和高频分量,然后针对低频分量和高频分量采用不同的融合策略。对于低频分量,提出一种根据相关系数进行加权平均的融合策略;对于高频分量,采用局部方差最大的融合策略。最后,利用熵、平均梯度等评价融合效果。仿真实验表明,该算法有效地提高了熵、平均梯度值,在较好保持原始图像信息的情况下丰富了图像的细节信息。
An improved image fusion algorithm based on wavelet transform was proposed.Firstly,wavelet transform was applied in fusion image to obtain low frequency and high frequency.Secondly,low frequency and high frequency were fused respectively by different methods.As for low frequency,a weighted average fusion method according to correlation coefficient was proposed.As for high frequency,a fusion method based on the biggest local square difference was proposed.Finally,the performance of the image fusion was evaluated using criteria like entropy and average gradient.Simulation experiments show that this algorithm improves entropy and average gradient and enriches the detailed information of images besides keeping original images information.
出处
《重庆理工大学学报（自然科学）》
CAS
2012年第1期61-65,共5页
Journal of Chongqing University of Technology：Natural Science
关键词
图像融合
小波变换
相关系数
熵
平均梯度
image fusion
wavelet transform
correlation coefficient
entropy
average gradient
