Quality Assessment of Image Fusion Based on Image Content and Structural Similarity

Image fusion quality assessment method based on Error-sensitive Model was seen as the issue of the fidelity of the image signal rather than fidelity of image information.Because they didn´t consider the human visual characteristics,they are inconsistent with the subjective evaluation.In this paper,based on SSIM algorithm,considering the fidelity of image information,it will combine the structural and content similarity of fused images and source images,use low-level visual perception information which contain luminance and contrast masking effects as the image content similarity weight,and the mutual information that fused image obtained from source image as the content information weight to calculate with the SSIM,proposes the quality assessment of image fusion based on image content and structural similarity IC_SSIM (Image Content based SSIM, IC_SSIM).The experiment results show that IC_SSIM algorithm is more consistent with human visual characteristics than PSNR,and is better than SSIM in resolution and accuracy.