基于视差图和复数轮廓波变换的无参考立体图像质量评价

现有的2D图像质量评价方法并不能很好地应用于立体图像质量评价中。为了有效评价不同失真立体图像的质量,提出了一种基于视差图和复数轮廓波变换的无参考图像质量评价方法。首先提取了能够反映3D信息的视差图,然后对左右失真图像和视差图进行复数轮廓波变换,计算能量和能量差特征,最后通过支持向量回归SVR模型训练学习,预测图像质量分数。实验结果表明,此方法优于当前文献报道的立体图像质量评价方法。

Blind stereo image quality assessment based on disparity map and complex contourlet transform

Given that the existing no-reference quality assessment methods for 2D images cannot be well used in stereoscopic images, we propose a no-reference (NR) image quality assessment method based on the disparity map and complex contourlet transform so as to effectively measure the quality of different types of distorted images. Firstly, we extract the disparity map that can reflect the 3D information, decompose the image by the complex contourlet transform and extract the following features: the energy of the sub-band coefficients within scales and the energy differences between scales. Then those features are fed to the support vector regression (SVR) model which can predict the image quality. Experimental results show that the proposed method outperforms current 3D IQA methods.