提升静态小波与自适应PCNN相结合的图像融合算法  

Image Fusion Using Lifting Stationary Wavelet Transform and Adaptive PCNN

作　　者：郭茂耘[1] 李华锋[1] 柴毅[1] 

机构地区：[1]重庆大学自动化学院,重庆400044

出　　处：《光电工程》2010年第12期67-74,共8页Opto-Electronic Engineering

基　　金：国家自然科学基金项目(60974090);教育部博士点基金项目(102063720090013);中央高校基本科研业务费资助(CDJXS10172205)

摘　　要：提出了一种新的基于提升静态小波变换与自适应PCNN相结合的图像融合算法。该方法定义一种图像单个像素的清晰度作为PCNN的链接强度,使得PCNN能根据像素清晰度的变化来自适应地调整链接强度的大小,接着对图像经提升静态小波分解得到的低频子带系数的改进拉普拉斯能量和及高频子带系数的单个像素的灰度值,分别作为自适应PCNN神经元的外部输入,并根据点火次数来确定图像融合系数。最后由提升静态小波变换的逆变换得到融合图像。实验表明,该方法在视觉效果和客观评价指标上都优于传统的基于小波变换、提升静态小波变换、提升静态小波-PCNN的图像融合算法。A novel image fusion algorithm based on the Stationary Lifting Wavelet Transform （SLWT） and adaptive Pulse Coupled Neural Network （PCNN） is proposed. Compared with the traditional PCNN where the linking strength of each neuron is the same, this adaptive PCNN uses the sharpness of each pixel as its value, so that the linking strength of each pixel can be chosen adaptively. By using a stationary lifting wavelet transform, we can calculate a flexible multiscale and shift-invariant representation of registered images. A Novel Sum-Modified-Laplacian （NSML） in the low frequency subbands and the pixels value of high frequency subbands of SLWT are input into motivate adaptive PCNN, respectively. The coefficients in SWLT domain with large firing times are selected as coefficients of the fused image. Experimental results demonstrate that the proposed fusion approach outperforms the traditional discrete wavelet transform-based SLWT-based and SLWT-PCNN-based image fusion methods in terms of both visual quality and objective evaluation.

关 键 词：图像融合 提升静态小波 脉冲耦合神经网络 拉普拉斯能量和 

分 类 号：TP391[自动化与计算机技术—计算机应用技术] TN911.73[自动化与计算机技术—计算机科学与技术]