基于双分解的双通道PCNN红外与可见光图像融合

为解决红外与可见光图像融合过程中存在的对比与清晰度较低和小目标易丢失等问题，提出了基于在双分解模型下的双通道PCNN(dPCNN)图像融合算法。首先对两幅源图像进行预增强处理，通过鲁棒的主成分分析(RPCA)将处理后图像分解为稀疏层与低秩层，接着，再利用非下采用剪切波变换(NSST)对的稀疏层进行多尺度分解得到低频子带与高频子带，然后对低秩层和低频子带采用局部加权能量与拉普拉斯能量两者取大的规则进行融合，对高频子带则利用dPCNN的点火图进行融合，最后将得到的融合成分进行逆变换或合成来得到最终融合图像。实验表明，该算法的融合图像目标信息对比突出、小目标信息明显，对源图像信息保留较好，客观评价指标也明显也优于其他算法，其中互信息有了大幅度的提升，有效地提升了红外与可见光图像的融合效果。

Dual channel PCNN infrared and visible image fusion based on dual decomposition

In order to solve the problems of low contrast and definition and easy loss of small objects in the fusion process of infrared and visible light images,a dual channel PCNN(dPCNN)image fusion algorithm based on the dual decomposition model is proposed in this paper.Firstly,the two source images are pre enhanced,and the processed images are decomposed into sparse layers and low rank layers by robust principal component analysis(RPCA).Next,the low frequency sub band and high frequency sub band are obtained by multi scale decomposition of the sparse layer using non lower shear wave transform(NSST).Then,the low rank layer and the low frequency sub band are fused using the rule that the local weighted energy and the Laplace energy are larger,and the high frequency sub band is fused using the ignition map of dPCNN.Finally,the obtained fusion components are inversely transformed or synthesized to obtain the final fusion image.Experiments show that the fusion image target information of the algorithm is outstandingly contrasted,the small target information is obvious,the source image information is well preserved,and the objective evaluation index is also significantly better than other algorithms,in which the mutual information is greatly improved,and the fusion effect of infrared and visible light images is effectively improved.