有理分形插值函数-中国图象图形学报.DOCVIP

中图法分类号：TP301.6 文献标识码：A 文章编号：1006-8961(2015 )  多尺度有理分形的图像插值 姚勋祥1，2， 张云峰1，2，*， 宁阳1，2， 刘一方1,2 1. 山东财经大学计算机科学与技术学院，山东济南，250014； 2. 山东省数字媒体技术重点实验室，山东济南，250014 *E-mail: yfzhang@ 摘 要：目的 图像插值是图像处理中的重要问题，为了提高纹理图像的放大质量，结合以往的有理函数的插值算法，本文提出了一种新的基于有理分形函数的图像插值算法。方法 对于输入图像，首先，运用中值滤波和直方图均衡化对输入图像预处理；其次，通过毯子覆盖法求出图像的多尺度分形特征值，进行纹理区域和平滑区域的划分；最后，在纹理区域采用有理分形插值函数，在平滑区域采用有理插值函数。结果 对于一般图像，本文算法与nonlocal autoregressive model (NARM)，new edge-directed interpolation (NEDI)相当，在纹理区域较多的图像中，本文算法在峰值信噪比(PSNR)和结构相似性Multi-scale Feature Image Interpolation Based on Rational Fractal Function Yao Xunxiang 1,2, Zhang Yunfeng1,2,*, Ning Yang1,2, Liu Yifang1,2 1. School of Computer Science Technology, Shandong University of Finance and Economics, Jinan 250014,China; 2. Shandong Provincial Key Laboratory of Digital Media Technology, Jinan 250014, China; *E-mail: yfzhang@ Abstract: Objective: Image interpolation plays a vital role in image processing. A new image interpolation algorithm based on rational fractal function is proposed for the purpose of improving the quality of texture image magnification. This method also combined with previous rational function interpolation algorithm. Method: For the input image, first of all, median filter and histogram equalization is used to input image preprocessing; Secondly, the image is classified the texture area and smooth area through the blanket method and multi-scale fractal characteristic value of the image; Thirdly, in the region of the texture using rational fractal interpolation function, adopting rational interpolation function in smooth area. Finally, optimization technique is utilized to further modify the interpolation model, which is proved to be effective. Result: A rational fractal interpolation algorithm is proposed in this article. As for common images，the quality of interpolation corresponds to nonlocal autoregressive model (NARM)，new edge-directed interpolation (NEDI). But for texture images, the proposed method is more competitive not on