基于NSCT及人眼视觉特性的医学图像融合

Based on NSCT and the Human Visual Characteristics of Medical Image Fusion

针对多模态医学图像融合问题,提出一种基于非下采样Contourlet(NSCT)变换与人眼视觉特性(HVC)的图像融合方法。NSCT变换对源图像的分解后,低频部分基于可见性测度的选择,高频部分基于纹理信息的选择,然后采用NSCT逆变换获得最终融合图像。实验结果表明:所提出的融合方法可以提高空间分辨率,同时保持光谱信息,并有改善。无论是在视觉效果和定量分析与传统的方法相比较,有好的优越性。

Fusion problem for multi- modality medical images, presents an image fusion method based on Nonsubsampled Contourlet(NSCT) transform and human visual system(HVC) is. After transformation of the source image NSCT decomposition visibility measure low frequency based on the selection, the high frequency portion of the texture information based on the selection, and then using the inverse transform to obtain a final blend NSCT image. Experimental results show that : the proposed fusion method can improve the spatial resolution, while maintaining spectral information and improved. Both visual and quantitative analysis in comparison with the conventional method, a good advantage.