基于提升静态小波变换与联合结构组稀疏表示的多聚焦图像融合

摘要

An image fusion algorithm based on Lifting Stationary Wavelet Transform(LSWT) and joint structural group sparse representation was proposed to restrain pseudo-Gibbs phenomenon created by conventional wavelet transform in multifocus image fusion,overcome the defect that the fusion method with conventional sparse representation was likely to lead textures,edges,and other detail features of fused images to the tendency of smoothness,and improve the efficiency and quality of multi-focus image fusion.Firstly,lifting stationary wavelet transform was conducted on the experimental images,different fusion modes were adopted according to the respective physical characteristics of low frequency coefficients and high frequency coefficients after decomposition.When selecting coefficients of low frequency,the scheme of coefficient selection based on joint structural group sparse representation was adopted;When selecting coefficients of high frequency,the scheme of coefficient selection based on Directional Region Sum Modified-Laplacian (DRSML) and matched-degree was adopted.Finally,ultimate fusion image was obtained by inverse transform.According to the experiment results,the improved algorithm can effectively improve such image indicators as mutual information and average gradient,keep textures,edges,and other detail features of images intact,and produce better image fusion effects.%为抑制传统小波变换在多聚焦图像融合中产生的伪吉布斯现象,以及克服传统稀疏表示的融合方法容易造成融合图像的纹理与边缘等细节特征趋于平滑的缺陷,提高多聚焦图像融合的效率与质量,采用一种基于提升静态小波变换(LSWT)与联合结构组稀疏表示的图像融合算法.首先对实验图像进行提升静态小波变换,根据分解后得到的低频系数与高频系数各自不同的物理特征,采用不同的融合方式.选择低频系数时,采用基于联合结构组稀疏表示的系数选择方案;选择高频系数时,采用方向区域拉普拉斯能量和(DRSML)与匹配度相结合的系数选择方案.最后经逆变换重构得到最终融合图像.实验结果表明,改进的算法有效地提高了图像的互信息量、平均梯度等指标,完好地保留图像的纹理与边缘等细节信息,融合图像效果更好.