基于非下采样剪切波变换域方向信息测度的多聚焦图像融合方法

根据非下采样剪切波变换(non-subsample shearlet transform, NSST)在图像融合领域的优势和多聚焦图像的成像特点,将NSST运用于多聚焦图像融合。在NSST变换域中,利用方向信息测度能够突出图像边缘纹理特征以及降低图像噪声,将其运用在高频子带系数的融合规则中,并对传统的低频融合策略进行改进,将低频系数和高频系数的融合图像进行NSST的逆变换，得到融合图像。计算机仿真实验表明，相比于传统的剪切波变换的融合策略,本文提出的图像融合方法提高了图像的融合质量,增加了图像的信息量,并能降低图像噪声,具有一定的有效性和实用性。

Fusion technique for multi-focus images based on orientation information measurement in non-subsample shearlet transform domain

Non-subsample shearlet transform (NSST) is suggested into the multi-focus image fusion, for its advantage in image fusion and the feature of multi-focus imaging. Orientation information measurement can highlight the edge texture characteristic and reduce the noise, so it is used in the fusion of high-frequency coefficients. Therefore, the fusion image is brought out after the NSST inverse transformation is taken on the fusion of low-frequency and high-frequency coefficients. Simulation is carried out to prove the effectiveness and practicability the proposed technique, which illustrates that the proposed method can improve the quality of fusion images and increase the information of fusion images, meanwhile, the noise of images is also decreased.