基于灰度能量差异性的红外与可见光图像融合

Fusion of Infrared and Visible Images Based on Gray Energy Difference

为了获取红外图像中的突出的目标特征,提取可见光图像中重要的细节信息,以及解决传统算法中目标信息不够突出,细节、纹理缺失严重的问题,本文提出了一种基于灰度能量差异性的红外与可见光图像融合方法。首先通过基于灰度能量差异性的显著目标提取算法检测出红外图像中的目标特征;然后采用非下采样轮廓波变换(non-subsampled contourlet transform,NSCT)对红外图像和可见光图像进行高低频的分解;将灰度能量差异图作为融合权重对红外图像和可见光图像的低频部分进行融合,对于高频部分采用加权方差的规则进行融合;最后对融合后的高频系数和低频系数进行NSCT逆变换得到最终的融合图像。本文选取了3组经典的红外与可见光图像进行融合实验,并且通过主观视觉和客观指标两个方面与其他几种方法作比较。实验结果证明了算法在突出目标信息、提高对比度、清晰度和保留纹理细节方面十分有效。

This paper proposes an infrared and visible image fusion method based on gray energy difference for two purposes:one,to obtain the prominent target features in an infrared image for extracting the important details in the visible image,and two,to solve the problem that the target information in traditional algorithms is not sufficiently prominent and that the details and textures are often missing.In this method,first,the target feature in the infrared image is detected by a target extraction algorithm based on gray energy difference.Second,infrared and visible images are decomposed to high and low frequencies using a non-subsampled contourlet transform(NSCT).Third,the gray energy difference map is used as the fusion weight to fuse the low-frequency parts of the infrared image and the visible image.The high-frequency part is fused by the rule of weighted variance.Finally,the NSCT inverse transform is used to fuse the high-frequency and low-frequency coefficients to obtain the final fused image.In this study,three groups of classical infrared and visible images are selected for fusion experiments and compared with other methods through subjective vision and objective indicators.Experimental results show that the algorithm can effectively highlight target information,improving contrast and sharpness and retaining texture details.