Fusion Algorithm of Infrared and Visible Images Based on NSCT and PCNN

This paper proposes a novel method for image fusion of infrared images and visible images based on PCNN(Pulse-coupled Neural Networks) and NSCT(Non-subsampled Contourlet Transform). First, a multi-scale decomposition on each registered image is performed by NSCT, which get the low-coefficients and high-coefficients. Second, the high-coefficients are employed to be the inputs of the PCNN. BFC is computed and used to combine the high coefficients. Low-coefficients are fused by the average. Last, the fused coefficients are used to reconstruct the fused image by an inverse NSCT. Experiments including infrared images and visible images are designed to testify the performance of the proposed method. The results shows that the proposed method not only keeps the more useful information than other methods, but also improves the quality of source images.