基于RDU-PCNN和NSST的红外和可见图像融合算法

针对可见光和红外图像的融合容易出现块状噪声 、边缘有振铃现象等不足,提出了一种基于区 域双通道脉冲耦合神经网络(RDU-PCNN)和非下采样剪切波变换(NSST)的红外和见光图像融 合算法。首先 对待融合的可见光和红外图像进行NSST分解得到低频系数和高频方向系数,低频系数采用R DU-PCNN的 规则融合,高频方向系数采用离散余弦变换(DCT)的融合规则,对融合后的系数进行逆NSST ,从而得到融合后 的图像。仿真结果表明,与其他5种目前流行或者较为先进的算法相比,本文的算法在视觉 和客观评价指标上优于其他算法。

Fusion algorithm for infrared and visible images based on RDU-PCNN and NSST

To overcome the limitation of noise and edge ringing artifacts in most fusion results,a novel image fusion algorithm based on region dual-channel unit-linking pulse coupled neura l networks (RDU-PCNN) and the nonsubsampled shearlet transform (NSST) is presented.After NSST decomposition, a fusion rule based on RDU-PCNN is used in the low frequency subband and a fusion rule based on discre te cosine transforfm (DCT) is used in the high frequency subbands.The fusion image can be obtained by taking an inverse NSST.Experiment al results show that our proposed algorithm is superior to other fusion algorithms in visual and objectiv e evaluation,when compared with other currently popular or more advanced methods .