基于Contourlet和改进牛顿插值的图像超分辨率重建

通过研究现有图像超分辨率重建算法原理,针对其不足,提出了一种基于Contourlet轮廓波和改进牛顿插值算法的图像超分辨率重建方法.该算法首先对原始图像进行Contourlet变换,然后对高频分量做多方向的改进牛顿差值计算,最后进行小波逆变换得到更加清晰的高分辨率图像.实验结果表明,该算法比传统插值算法和小波插值算法具有更好的视觉效果,并且拥有更高的信噪比和结构相似性.

Improved super-resolution algorithm based on Contourlet transformation and interpolation

The image super-resolution reconstruction uses the super-resolution reconstruction algorithm to get high resolution image on the basis of the existing low resolution image .Aiming at the shortcomings of the existing methods, this paper proposes a new algorithm which based on Contourlet transformation and interpolation .This algorithm first deals the original image with Contourlet transformation, and then calculates the high frequency component with the improved Newton interpolation in different directions. Finally, we can get more clear high resolution image through the wavelet inverse transformation. The experimental results show that the proposed algorithm has better visual effect than the traditional interpolation algorithm and wavelet interpolation algorithm and it has a higher PSNR.