一种基于提升小波变换的图像融合方法

在针对传统的多尺度分解的融合方法运算速度慢、内存需求量大，不适于实时应用的局限性的基础上，提出了一种基于提升小波变换的图像融合算法。多个源图像分别进行提升小波分解，使用恰当的融合规则合并各尺度对应的分解系数，通过提升小波逆变换得到融合图像。实验结果表明，提出的算法无论在执行时间还是融合图像质量上都优于传统方法，有广泛的应用前景，特别适用于实时系统。

A Image Fusion Method Based on Lifting Wavelet Transform

Recently,multi-resolution techniques especially 2d wavelet transform have been employed for image fusion.Since the traditional wavelet transform is limited with its slow speed and small memory,which impedes its realtime application,a novel image fusion algorithm using lifting wavelet transform is proposed in this paper.Firstly,the source images are decomposed using lifting wavelet transform respectively.Then,the decomposition coefficients are combined by appropriate fusion rules.Finally,the composite image is obtained by using reversed lifting wavelet transform.Experimental results demonstrate that our algorithm is more effective in both computational speed and fused image quality than traditional image fusion methods.It has great significance in practical applications,particularly in realtime system.