基于形态学和平均梯度的小波图像融合算法

摘要

提出了一种基于形态学和平均梯度的小波图像融合算法。其基本思想是利用形态学对图像预处理，再对各图像进行小波变换，对低频部分根据分块后各子块的平均梯度和局部能量，按照一定规则选择融合系数；高频部分采取最大系数法选择融合系数。最后进行小波逆变换得到融合图像。实验结果表明，该算法能显著改善融合图像的视觉效果。%This paper proposes a novel image fusion algorithm based on morphology and average grads in the area of wavelet. The basic idea is to perform image pre-processing by morphology first,and then to decompose the image into low-frequency component and high-frequency component through wavelet transformation. The average-grads and partial energy within the low-frequency are measured to determine the fused low-frequency component,and select the maximal absolute values in high-frequency coefficients to be the fused high-frequency component. Finally,the fusion image is reconstructed by wavelet inverse transform. Experimental results indicate that the algorithm remarkably improves the visual effect of the fused image.