A novel image fusion algorithm using PCNN In NSCT domain

This paper proposed a novel image fusion framework based on PCNN in NSCT domain. Compared with the contourlet transform(CT), NSCT has translational invariance and can also overcome the image pseudo Gibbs phenomena around singularites. PCNN is evolved from mammal's visual cortex neuron model, and characterized by its pulse synchronization and acquisition of the neurons. Firstly, the source images were decomposed into low-frequency and high-frequency sub-bands, and in the fusion process, the sum-modified-laplacian(SML) was used to motivate PCNN. In the same time, the coefficients with larger firing times were selected as the fusion coefficients. Finally, the fused image can be obtained by the inverse NSCT. Experimental results show that the method is better than the traditional fusion methods, such as the wavelet transform, the contourlet transform and the PCNN based methods both in subjective appearance and objective criteria.