交叉双边滤波和视觉权重信息的图像融合

Image fusion via cross bilateral filter and visual weight information

图像融合技术是图像分析领域重点研究内容之一,为了更好地保留原图像中的细节信息,提高融合图像的对比度,提出了基于视觉权重图的多尺度图像融合方法。首先,利用可变参数的交叉双边滤波器对两幅待融合图像进行多尺度分解;然后,在每个分解层分别计算相应的视觉权重图,并针对不同分解层赋予不同的权重值;最后,综合这些结果生成融合图像。由于对原始图像的分解没有采用下采样和上采样操作,因此不会损失图像中的信息,且克服了传统像素级融合方法中融合图像模糊、对噪声敏感等不足。通过4种定量分析实验表明,在多种模式的图像融合应用中,本方法优于其他5种对比方法,融合时间小于0.2 s。融合后图像细节信息、对比度得到增强,同时降低处理时间。

Image fusion technology plays a vital role in the field of image analysis. In order to retain the details of the source images and improve the contrast of fusion image,a novel image fusion method is proposed based on visual weight map and multi-scale decomposition.Firstly,bilateral filter with varied parameters is used for the multi-scale decomposition of two source images. Then,visual weight map in each decomposition level is calculated and different weights are assigned to the different decomposition levels. Finally,fused image is generated by synthetizing these results. The image information is totally retained due to the decomposition without up-sampling and downsampling. Moreover,this method can also overcome some of the well-known problems in pixel level fusion such as blurring effects and high sensitivity to noise. Experimental results by using four metrics show that the proposed algorithm obtains dramatically improved performance compared to the other five state-of-the-art algorithms. In addition,the computation time of our approach is less than 0. 1seconds,which is much better than other methods. The details and contrast of fused image are enhanced,and the computation time is reduced dramatically.