Adaptive Fusion Algorithm of Multisensor Image in Nonsubsampled Contourlet Domain

Abstract

Considering of the characteristics of low visible light images and infrared images and combining with the human visual system, a novel image fusion technique is presented in this paper. The fusion technique is based on immune clonal selection (ICS) in natural immune selection and nonsubsampled contourlet transform (NSCT). NSCT can give the asymptotic optimal representation of the edges and contours in image by virtue of the characteristics of good multiresolution, shift-invariance and multi-directionality. And then ICS is introduced into NSCT domain to optimize the fusing weights adaptively. Numerical tests show that this algorithm provides improvements both in visual effects and quantitative analysis. And the fusion images hold edge and texture information well and have ideal contrast and definition.