首页 | 本学科首页   官方微博 | 高级检索  
     检索      

利用分裂Bregman方法的图像恢复组合模型
引用本文:张万涛,李维国,邢丽丽.利用分裂Bregman方法的图像恢复组合模型[J].计算数学,2011,33(3):313-320.
作者姓名:张万涛  李维国  邢丽丽
作者单位:中国石油大学(华东)数学与计算科学学院, 山东青岛 266555
基金项目:国家自然科学基金,中国石油大学(华东)研究生创新基金
摘    要:ROF模型是图像恢复中的经典模型,具有保留图像边缘的优点,但同时也存在梯子现象.而利用二次范数fΩ|▽u|2 dxdy的模型可以避免梯子现象,但容易使图像变得模糊.针对两种方法的优缺点,提出了一种新的通过设置边缘检测开关函数的组合模型,在图像平坦区利用二次范数模型处理,而在强边缘处利用ROF模型处理,而且应用分裂的Br...

关 键 词:图像恢复  ROF模型  梯子现象  分裂Bregnan方法
收稿时间:2010-10-10;

A COMPOUND MODEL OF IMAGE RESTORATION USING SPLIT BREGMAN METHOD
Zhang Wantao,Li Weiguo,Xing Lili.A COMPOUND MODEL OF IMAGE RESTORATION USING SPLIT BREGMAN METHOD[J].Mathematica Numerica Sinica,2011,33(3):313-320.
Authors:Zhang Wantao  Li Weiguo  Xing Lili
Institution:School of Mathematics and Computational Sciences, China University of Petroleum, Qingdao 266555, Shandong, China
Abstract:The ROF model is one of the typical models in image restoration which can recover images with edges. However, there is staircase when using it to recover images. The quadratic norm model can avoid the effect, but it may make the images blurry. Due to the a dvantages and shortcomings of the two models, we propose a new compound model by using edge indicator switch function, which recover images using the quadratic norm model in the flat regions and ROF model in the strong edge regions, and solve it using the split Bregman method. Also, we prove the convergence of the new model. The numerical experiments show the new model has the advantages of the two models mentioned above and our algorithm is efficient.
Keywords:image restoration  ROF model  staircase  split Bregman method
本文献已被 万方数据 等数据库收录!
点击此处可从《计算数学》浏览原始摘要信息
点击此处可从《计算数学》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号