一种改进的基于方向区域能量的多聚焦图像融合算法

摘要

通过分析图像经小波变换后各子带的方向区域特性和比较各种图像融合算法的效果,提出了一种改进的基于方向区域能量选大图像融合算法.对于低频子带系数,采用基于归一化区域梯度加权的融合规则确定低频融合系数；对于高频子带系数,根据其所在高频子带的方向特性,采用基于方向区域能量选大的融合规则,进而确定高频融合系数.实验结果表明:改进方法能够获得具有更丰富的信息量和更高清晰度的融合图像.%An improved image fusion algorithm based on maximum region-energy with its directional correlation is presented by analyzing the directional characteristics of the wavelet coefficients and comparing the effects of applying various image fusion algorithms. For the coefficients of low-frequency sub-band,the low frequency fusion coefficient is determined by the fusion rule based on normalized regional gradient weighting. For the coefficients of high-frequency sub-band, the fusion coefficient is determined by the maximum directional region-energy in different high-frequency regions based on the directional characteristics. Experiments show that the fusion image of proposed algorithm has more information content and higher clarity.