The Nonsubsampled Contourlet Transform in Multisensor Images Fusion
JIA Jian1,2, JIAO Li-cheng1, SUN Qiang1
Author information+
1. Institute of Intelligent Information Processing,Xidian University,Xi’an,Shaanxi 710071,China;2. Department of Mathematics,Northwest University,Xi’an,Shaanxi 710069,China
Based on the properties of NonSubsampled Contourlet Transform,shift invariance,multiscale and multidirectional expansion,a novel multisensor images fusion method was proposed.The proposed method was applied to the multifocus and the multispectral images fusion in the experiments respectively.We gave the subjective judgment and numeric measurement of the fusion image based on visual effect and information indexes.To evaluate the performance of the proposed algorithms,we compared our results with those of the algorithms,such as contourlet transform,wavelet transform and principal component analysis.The simulation results indicate that the proposed method outperforms the others in terms of both visual quality and objective evaluation criteria,and maintains much more detail information as well.At the same time,the proposed algorithm widens the applications of the NSCT.