基于特征能量加权的红外与可见光图像融合

目前红外与可见光图像直接融合存在红外目标取舍和场景信息提取困难,结合非采样Contourlet的多尺度、多方向性和平移不变性的优点,本文提出了一种基于非采样Contourlet变换(NSCT)的红外与可见光图像融合方法.首先对源图像进行分解,然后低频子带通过构造基于区域的特征像素能量,进行加权融合,高频子带直接选用方差取大法融合.使用该算法进行了融合实验,并给出了融合质量评价.实验结果表明,本文提出的基于NSCT的图像融合算法在保留图像细节信息、增加信息量方面都有显著地提高.

Infrared and Visible Image Fusion Based on Feature Energy

The eurrent infrared and visible image fusion faces the difficult of obtaining infrared target and extracting the scene of visible image.An infrared and visible image fusion method based on Nonsubsampled Contoutlet Transform (NSCT)is proposed by using the advantage of nonsubsampled Contoutlet transform's direction and anisotropy.Firstly,the source images were decomposed.Then,low-frequency sub-band establishes new feature energy,and weights it for the fusion.And high-frequency sub-band selects the max variance of the image fusion roles.By using the algorithm for infrared and visible light images in the fusion experiments,the fusion quality assessment is given.The experimental results show that the proposed image fusion algorithm gets more image detail in the information,and the amount of information has be increased significantly.