基于梯度理论的多聚焦图像融合

Multi-Focus Image Fusion Based on Gradient Theory

多聚焦图像融合是把各个不同聚焦点的图像组合在一起的方法,其结果是融合成信息量互补的图像。梯度信息有表示变化快慢的特性,在图像融合应用中,可以度量图像的活跃程度。因此提出基于改进梯度理论(GDF)的多聚焦图像融合算法。首先利用改进梯度算法求取图像水平、垂直方向的梯度信息,在获得不同方向绝对值之后,再把其相加,从而得到活跃程度度量图。在对比不同度量图的时候,使用滑动窗口技术,最终获得决策图。在决策图的基础上对原始图像进行加权,最终获得的图像就是融合图像。此算法与传统算法相比都有一定的优越性。

Multi-focus image fusion is a method of combining images of different focal points together,and the result is fused into images with comple⁃mentary information.Proposes a new method,improved gradient theory(GDF)to measure the activity level of the image.Firstly,the gradi⁃ent information of the horizontal and vertical directions of the image is obtained by using the improved gradient algorithm.After obtaining the absolute values in different directions,they are added together to obtain the activity metric map.When comparing different metric graphs,uses the sliding window technique to finally obtain the decision graph.The original image is weighted on the basis of the decision graph,and the resulting image is the fused image.This algorithm has certain advantages compared with traditional algorithms.