基于NSCT变换的多传感器图像融合算法

Multi-Sensor Image Fusion Algorithm Based on NSCT

针对多传感器图像融合中偏袒法和平均法易削减图像对比度的缺点,提出一种基于NSCT变换的图像融合算法。首先对来自同一场景配准后的多传感器图像进行NSCT变换;然后采取不同的融合策略分别对低频和高频方向子带系数进行融合:低频子带系数采用区域能量自适应加权的方法,高频方向子带系数采用局部区域能量匹配的加权平均法与选择法相结合的方案;最后通过NSCT逆变换得到融合图像。分别对红外与可见光图像和多聚焦图像融合进行实验,并对融合图像进行主客观评价,实验结果表明:该算法得到的融合图像具有较好的主观视觉效果和客观量化指标,并优于传统的融合方法。

Specific to the drawback that favoritism and average method are weak in maintaining the contrast of fusion image,a novel fusion algorithm based on Nonsubsampled Contourlet Transform（NSCT） is proposed.Firstly,the registered multi-sensor images from the same scene were transformed by Nonsubsampled Contourlet Transform.Then the high and low frequency coefficients are fused separately by using different fusion strategies：the low frequency coefficient is fused by adaptive regional energy,while the high frequency coefficient is fused by using regional energy matching with weighted mean and selection method.Finally,the target image is obtained by performing inverse Nonsubsampled Contourlet Transform.The algorithm has been used to merge infrared and visible images and multi-focus images.The experimental results indicate that the fused image obtained by the proposed method has a better subjective visual effect and objective evaluation criteria and performs better than traditional fusion methods.