基于高阶奇异值分解的多聚焦图像融合新方法

A New Multi-focus Image Fusion Method Based on Higher Order Singular Value Decomposition

针对传统多聚焦图像融合方法不能有效度量图像中聚焦区域的问题,提出一种基于高阶奇异值分解的图像融合新方法。高阶奇异值分解可以精确提取图像纹理特征,并对源图像进行分解。充分考虑像素间的区域相关性,提取分解系数的空间频率、平均梯度和区域能量等区域特征,并提出一种新颖的基于区域特征的多活动测度融合策略,该策略可有效度量图像中的聚焦区域。实验结果表明,该融合方法能够更好地保留图像的边缘细节信息。

For the problem of traditional multi-focus image fusion methods can not effectively measure the focus regions in images,a new multi-focus image fusion method is proposed based on higher order singular value decomposition.As higher order singular value decomposition（HOSVD）can exactly extract features in images,HOSVD is used as an effective decomposition tool.Considering the regional correlation between pixels,space frequency,average gradient and regional energy are extracted.And a novel fusion strategy on comprehensive consideration of multiple regional features is applied to decomposition coefficients.Experimental results show that the proposed method performs better than traditional image fusion methods.