基于小波变换的双通道脉冲耦合神经网络图像融合

提出了一种基于小波变换的双通道脉冲耦合神经网络的图像融合方法.先使用小波变换的方法来分解配准后的各个源图像,进而得到各个源图像的低频分量和高频分量,再把得到的低频和高频系数进行融合处理,使用高斯加权平均的低频融合规则来处理低频子带,利用双通道脉冲耦合神经网络的融合规则处理各高频子带,链接系数为图像的清晰度.融合后的小波系数取决于点火图和点火次数的多少,最后的融合图像由小波逆变换得到.实验结果表明,该方法能更有效地提取原始图像的特征信息,在主观视觉效果以及客观性能指标上较传统算法都有所改善.

Image Fusion of Dual Channel Pulse Coupled Neural Network Based on Wavelet Transform

An image fusion method based on wavelet transform for dual channel pulse coupled neural network is proposed. Firstly,the wavelet transform method is used to decompose the registered source images,and then the low-frequency components and high-frequency components of each source image are obtained,and the obtained low-frequency and high-frequency coefficients are all fused,and the Gaussian weighted average low-frequency fusion is used. The rules deal with the low-frequency sub-bands,and the high-frequency sub-bands are processed by the fusion rule of the two-channel pulse-coupled neural network,and the link coefficient is the sharpness of the image. The fused wavelet coefficients depend on the size of the ignition pattern and the number of firings. The final fused image is obtained by inverse wavelet transform. The experimental results show that the proposed method extracts the feature information of the original image more effectively,and the subjective visual effect and objective performance index are improved compared with the traditional algorithm.