一种边缘定向平滑图像插值算法AnEdge-directed-电子与信息学报.PDF

第 34 卷第 7 期 电 子 与 信 息 学 报 Vol.34No.7 2012 年 7 月 Journal of Electronics Information Technology Jul. 2012 一种边缘定向平滑图像插值算法 赵旦峰 王 博* 杨大伟 (哈尔滨工程大学信息与通信工程学院 哈尔滨 150001) 摘 要：针对传统图像放大算法边缘处理效果较差，自适应图像插值方法存在高计算复杂度的问题，该文提出一种 有效增强图像边缘轮廓的插值放大算法。该算法结合边缘定向平滑滤波器和双线性插值的特点，使得图像在平坦和 非平坦区域均能取得理想效果。仿真测试结果表明，与基于统计特征的自适应插值算法相比，该文提出算法能显著 提高插值速度，平均运行时间降低 8.33 s；与双三次插值算法相比，图像峰值信噪比平均增加0.30 dB。 关键词：图像处理；边缘定向；平滑滤波器；双线性；插值 中图分类号：TN911.73 文献标识码： A 文章编号：1009-5896(2012)07-1543-04 DOI: 10.3724/SP.J.1146.2011.01275 An Edge-directed Smoothness Algorithm for Image Interpolation Zhao Dan-feng Wang Bo Yang Da-wei (College of Information and Communication Engineering, Harbin Engineering University, Harbin 150001, China) Abstract: A novel effective image interpolation mechanism to enhance the enlarged image edges is proposed for the visual artifacts problem in traditional magnification methods and the high computational cost in some adaptive image interpolation methods. The proposed algorithm combines the edge-directed smoothness filter with the bilinear interpolation algorithm to obtain the higher quality of an interpolated image both in the edges and smooth areas. The simulation results show that the proposed algorithm gets a better visual effect. Compared with the edge-adaptive interpolation algorithm using covariance, the proposed algorithm is less complex with running time reduced 8.33 s on average and compared with the bi-cubic algorithm has higher quality with the PSNR is raised 0.30 dB on average. Key words: Image processing; Edge-directed; Smoothness filter; Bilinear; Interpolation 1 引言 甚至图像模糊现象。 图像插值技术在很多领域都有广泛应用，近年