A Medical Image Fusion Method Based on NSCT and PCNN with Neighborhood Stimulation

YANCHUN YANG, YANGPING WANG, JIAO LI

Abstract


In order to better guide clinical diagnosis and treatment, this paper proposes medical image fusion method based on NSCT (nonsubsampled contourlet transform) and PCNN (pulse coupled neural network) with neighborhood stimulation. A fusion rule based on PCNN with neighborhood standard deviation stimulation is adopted in low frequency sub-band coefficient. Through the application of the PCNN simplified model, the PCNN model is stimulated by neighborhood standard deviation according to human visual system. The low frequency sub-band coefficient is determined by firing times. When choosing the bandpass directional sub-band coefficients, a fusion rule based on region gradient energy is adopted according to directional characteristics of NSCT. The experiment results show that the proposed method provides abundant details and edge information over conventional methods in achieving the better quality of fusion.

Keywords


nonsubsampled contourlet transform (NSCT); pulse coupled neural network (PCNN); standard deviation; gradient energy; medical image fusionText


DOI
10.12783/dtetr/iceta2017/19922

Refbacks

  • There are currently no refbacks.