Medical image fusion based on Maclaurin expansion and PCNN

To address the problem that traditional multi-scale transform-based methods use single features while neglecting complementary features such as curves and edges. A Pulse Coupled Neural Network medical image fusion method based on a combination of Maclaurin expansion and Gaussian homomorphic filtering enhancement is proposed. The source image is firstly decomposed into deviance components and multi-level energy components (hereafter decomposed to three levels) by Maclaurin expansion, and then the three-level energy components are enhanced by Gaussian homomorphic filtering to obtain the enhanced three-level energy component sub-map The experimental results show that the method is more universal than other methods in terms of image sharpness, detail information retention and image fusion quality.