基于Contourlet的SSIM图像融合质量评价方法

利用contourlet能有效表示图像的边缘和轮廓信息的特点,并将其引入到图像融合后的评价中,提出了一种基于Contourlet的结构相似度(SSIM)图像融合质量评价方法。以失真信道模拟融合过程,通过对标准参考图像与融合图像之间的结构相似度来实现对融合方法、融合规则性能优劣的评判。实验表明该方法与主观评价方法有很好的一致性,且与经典的SSIM算法相比具有更好的区分度。

A Method for SSIM Image Fused Quality Assessment Based on Contourlet

As Contourlet transform can give an asymptotic representation of edges and contours in image by virtue of its characteristics of multi-direction and flexibility, it is introduced into the quality assessment of fused image. A new method （SSIM） for fused image quality assessment based on Contourlet and structural similarity of images is proposed in this paper. We introduce distortion channel to simulate the fusion, that is comparing the structural similarity between reference image and the fused image to achieve the evaluation of the performance of fusion method. According to this algorithm, a set of experiments are implemented. The results illustrate that the proposed method has a good consistency with the method of subjective assessment, as well as better differentiation compared to the classical SSIM algorithm.