摘要
提出一种基于小波变换的图像融合新算法。在处理低频部分时,采用了基于局部能量的低频系数融合方案;在处理高频部分时,依照一定的原则,自定义了特征边缘点和非特征边缘点、有效边缘区和有效平滑区,并根据不同的区域特征提出了一种基于小波区域梯度的图像融合新算法。最后,利用信息熵、均方差、空间频率来评价新融合算法的效果。仿真实验结果表明,相比传统的图像融合算法,该融合算法可达到更好的效果。
A new algorithm of image fusion based on wavelet transform is proposed. In the fusion processing, for the low-frequency part, we used a fusion scheme based on local energy according to low-frequency coefficient; to deal with the high-frequency part and in accordance with some principle, we gave the definition with characteristics-edge point and non-characteristics-edge edge area and effective- smooth area. Then according to the different characteristics point, effective- of the area, we proposed a new algorithm based on wavelet regional gradient image fusion. Finally, we used the infor- mation-entropy and mean variance, clarity to evaluate the effect of the new fusion algorithm. The simulation results show that the new fusion algorithm in this paper can get better effect than other traditional image fusion algorithm
出处
《重庆理工大学学报（自然科学）》
CAS
2012年第10期51-55,共5页
Journal of Chongqing University of Technology：Natural Science
关键词
图像融合
小波变换
区域梯度
熵
均方差
image fusion
wavelet transform
local gradient
entropy
clarity
