Applying NSCT (nonsubsampled contourlet transform) theory to achieving effective image fusion

Aim: In our opinion, traditional image fusion algorithms based on Contourlet transform are not quite effective. So we propose applying Ref. 5's NSCT theory to achieving effective image fusion. Section 2 of the full paper explains the NSCT based algorithm. It explains essentially three things: (1) the low frequency information fusion rule uses the weighted average mean to obtain the fused low-frequency information as eq. (1); (2) the high-frequency fusion rule uses the spatial frequency of image as its measurement criterion and obtains the fused high-frequency information; (3) the final fusion image is obtained by performing the NSCT reconstruction of the low-frequency information and the high-frequency information. Section 3 analyzes the results of experiments, which are given in Fig. 2 and Table 1. The analysis shows that the NSCT based algorithm is effective for retaining the original image information and extracting the characteristics of the image to be fused.