【作者】 黄淑英； 杨勇； 王国宇；

【Author】 HUANG Shu-ying~(1,2),YANG Yong~3,WANG Guo-yu~1 (1.Department of Electrical Engineering,Ocean University of China,Qingdao 266100,China; 2.School of Software and Communication Engineering, Jiangxi University of Finance and Economics,Nanchang 330032,China; 3.School of Information Technology,Jiangxi University of Finance and Economics,Nanchang 330032,China)

【机构】 中国海洋大学 电子工程系； 江西财经大学 软件与通信工程学院； 江西财经大学 信息管理学院；

【摘要】 基于重建的超分辨率方法中,图像求解是典型的不适定反问题,需借助有效的正则来稳定求解。针对二阶偏微分方程在重建图像中出现的块效应问题,引入四阶偏微方程的正则化处理,并用2-范数和四阶偏微分方程构造一种新的能量函数,提出一种新的超分辨率重建方法。通过仿真实验对比本方法与双三次插值和全变分方法重建高分辨率图像的差异。结果表明,该方法在抑制噪声的同时具有很好的保边缘能力。更多还原

【Abstract】 Super-resolution image reconstruction(SRIP) is a typical ill-posed inverse problem,which can be stabilized by using a valid regular.To overcome the blocky effects of the conventional used second-order partial differential equations (PDE),a novel fourth-order PDE based method for SRIP was proposed,which is formed by incorporating the fourth-order PDE regularization in the image reconstruction.The method was developed by defining a new energy function from 2-norm and fourth-order partial differential equations.The simulation experiments of the proposed method were compared with those of bi-cubic interpolation method and total variation algorithm.The results show that the proposed method has better performance in suppressing noise with preserving edge.更多还原