一种自适应协方差的低复杂度图像放大方法

Low complexity image magnification method based on adaptive covariance

对于图像放大技术而言,重要的就是要权衡到图像质量以及计算复杂度。传统的基于线性或三次样条插值的方法会带来图像模糊和锯齿边缘等失真,为了解决这一问题,人们提出一种基于迭代和学习的算法,但是这种方法带来了很高的计算复杂度。综合以上几点本文提出了一种基于自适应协方差的图像放大方法(adaptive covariance-based edge diffusion,ACED)。该方法能很好地权衡图像放大性能和复杂度之间的关系。在这种方法中,提出了一种联合边缘判别准则,并自适应选择扩散模板来估计局部协方差系数,以高效的减少图像放大带来的失真。实验结果表明,所提出的方法在主观质量和客观质量上都有很大的提升,同时也具有较低的计算复杂度。

For image magnification technology,it is important to weigh the image quality and computational complexity. The traditional method based on linear or three spline interpolation can cause the distortion of image blur and jagged edges. In order to solve this problem,we propose an algorithm based on iteration and learning,but this method brings a high computational complexity.Based on the above points,this paper proposes an image magnification method called adaptive covariance-based edge diffusion（ ACED） which is based on adaptive covariance. This method can well balance the relationship between image magnification and complexity. In this method,in order to reduce the distortion caused by the image magnification,this paper proposes a joint edge criterion to estimate the local covariance coefficient by adaptively selecting diffusion template. The experimental results show that the proposed method has a great improvement in both subjective and objective quality,and it also has a low computational complexity.