基于离散Tchebichef变换的多聚焦图像融合方法

摘要

Image fusion based on image transform technologies is always used in multi-focus image fusion. It transforms images into transform domain and fuses the transformed image according to a specific fusion rule. After that, the fused image is achieved by the inverse image transform. The transform based image fusion methods are robust to noise and the fused results are widely accepted. This paper proposes a multi-focus image fusion method based on discrete Tchebichef orthogonal polynomial transform. Discrete orthogonal polynomial transform is firstly introduced to the field of multi-focus image fusion. The proposed method combines the spatial frequency with the discrete orthogonal polynomial transform coefficients of image, and it directly achieves the value of spatial frequency by the discrete orthogonal polynomial transform coefficients of the image and avoids the process of recalculation that transforms the discrete orthogonal polynomial transform coefficients to space domain. The proposed method can reduce the fusing time in multi-focus image fusion and improves the fusion effect.%基于变换技术的图像融合是多聚焦图像融合中常采用的方法,其过程是将图像转换到变换域按照一定的融合规则进行融合后再反变换回来,具有较强的抗噪能力,融合效果明显.该文提出一种基于离散Tchebichef正交多项式变换的多聚焦图像融合方法,首次将离散正交多项式变换应用到多聚焦图像融合领域.该方法成功地利用了图像的空间频率与其离散Tchebichef正交多项式变换系数之间的关系,通过离散正交多项式变换系数直接得到空间频率值,避免了将离散多项式变换系数变换到空域计算的过程.所提方法节省了融合时间,并提高了融合效果.