图像质量评价已经从图像处理中分离出来成为一个独立的研究领域。图像信息的最终接受者是人，所以主观方法是可靠的。但需要多次重复实验，费时，难以应用到实际中。客观图像质量评价方法成为了研究的热点，传统的客观评价方法计算简单，但由于它没有考虑人类的视觉感受以及图像本身的特点，经常与人的主观评价脱节。Zhou.Wang等人结合结构失真理论提出的结构相似度 (SSIM)算法虽然很优越，但也存在不足。 本文深入分析了SSIM算法，通过实验验证SSIM的优越性，同时指出了SSIM在评价同一幅图像不同失真类型时存在一定的误差，该问题主要存在于严重模糊的图像中。模糊图像的纹理特性受到破坏，而SSIM方法中的结构比较(s)分量并不能准确的反映这种变化，图像的梯度可以很好的反映出边缘及纹理的变换情况。所以本文在引入梯度信息的SSIM算法的基础上，考虑Contourlet拥有良好的类似于人类视觉系统(HVS)的特性，具有局部化、多尺度、多方向等优良特性，提出了基于Contourlet的梯度结构相似度(CGSSIM)图像质量评价算法。 CGSSIM算法将梯度结构相似度(GSSIM)算法用于Contourlet分解频带的各方向子带，得到各带通方向子带的梯度结构相似度，然后对各带通方向子带的梯度结构相似度求加权和得到整幅图像的基于Contourlet的梯度结构相似度(CGSSIM)。权值通过实验根据人眼对不同带通方向子带敏感度确定。通过实验证明了CGSSIM算法对同一幅图像的高斯模糊和白噪声失真图像交叉评价时可以得到较好的结果。同时验证了CGSSIM算法比PSNR、SSIM等算法更符合人眼的视觉特性，与主观感知更加接近，评价准确率和一致性更好。特别是对高斯模糊图像的评价，其准确性有了一定的提高。

Image quality assessment has been separated from the image processing to become an independent field of study. The ultimate recipient of the image information is a person, so the subjective method is reliable. But it requires repeated experiments, time-consuming, difficult to be applied to practical applications. Objective image quality assessment method has become a hot research, traditional objective evaluation method is simple, but because it doesn’t consider the human visual perception and image characteristics, often is out of touch with people's subjective evaluation. Z.Wang combined the structural distortion theory and proposed structural similarity (SSIM, structural similarity) algorithm.The algorithm is superior, but there is also drawback. SSIM algorithm is analysed in this paper. The superiority of SSIM has been verified through experimental result.The SSIM also pointed out that there are some errors in the evaluation with different types of distortion image for the same image. The main problem exists in severely blurred image. The texture characteristics of blurred image are destroyed, the structure(s) component of SSIM methodand does not accurately reflect the change. The gradient of the image can be well reflected in the edge and the texture transformation. Follow the existing gradirent structure similarity algorithm; consider Contourlet has good characteristics similar to that of the human visual system (HVS). Contourlet has excellent characteristics of the local, multi-scale, multi-direction. This paper proposes an image quality assessment algorithm based on Contourlet and gradient structural similarity (CGSSIM). In this method, the gradient structure similarity (GSSIM) algorithm is used to each band-pass directional subband of Contourlet decomposition. Gradient structure similarities of all directional subbands are weighted, and get the Contourlet transform domain of the whole image gradient structure similarity (CGSSIM). Weights experimentally determine the sensitivity of the human eye with different band-pass directional subband. By the experiment, the CGSSIM algorithm is proved that can get better results with an image of the Gaussian blur and the image of white noise distortion cross evaluation. The experimental results also show that the CGSSIM algorithm is more suitable for the human visual system characteristics than the PSNR and SSIM algorithm, and it is closer with the subjective perception, the accuracy of evaluation and consistency better. Especially Gaussian blur image evaluation, its accuracy has been improved to some extent.