Fusion of Infrared and Visible Images Based on Pulse Coupled Neural Network and Nonsubsampled Contourlet Transform

摘要

This paper presents a new fusion algorithm which can effectively solve the problem that has unobvious infraredtarget and low contrast in infrared and visible image fusion. This paper’s innovation point is its fusion rule. Other algorithms’fusion rules usually use pulse coupled neural network (PCNN) and region characteristics to select low frequencyor bandpass subband coefficients. The proposed algorithm innovatively applies improved PCNN and region characteristicsto the selection of both low frequency and bandpasssubband coefficients in nonsubsampled contourled transform(NSCT) domain. First, the subband coefficients of original image are obtained by NSCT. Then, the decomposed subbandcoefficients are processed by using PCNN, whose fire mapping images are obtained. The method of region standard deviationisused to choose the fusion coefficients of fire mapping image, which satisfies to get more image information inlow frequency part. For bandpass subband coefficients’ fire mapping image, the method based on region energy isadopted for the fusion coefficients, which makes the bandpass part captures more energy. Finally, the fused image can beobtained by inverse transform of NSCT. Compared with typical wavelet-based, NSCT-based, NSCT-PCNN based fusionalgorithms, experiment shows that the new proposed algorithm improves the fused image’sobjective evaluation index significantly,obtainsa prominent infrared target and better fusion image quality.