基于双十字TV模型的图像修复算法

翟东海; 段维夏; 鱼江

doi:10.3969/j.issn.1001-0548.2014.03.020

基于双十字TV模型的图像修复算法

1.
西南交通大学信息科学与技术学院成都 610036;
2.
西藏大学工学院拉萨 850000

基金项目:

国家社会科学基金(12EF119); 西藏自治区科技厅科技计划重点项目(Z2013B28G28/02); 国家级大学生创新创业训练计划(201210694019)

详细信息

作者简介:
翟东海(1974-),男,副教授,主要从事数字图像处理方面的研究.

中图分类号: TP301.6
计量
- 文章访问数: 4481
- HTML全文浏览量: 135
- PDF下载量: 83
出版历程
- 收稿日期: 2013-03-17
- 修回日期: 2013-12-10
- 刊出日期: 2014-06-14

Image Inpainting Algorithm Based on Double Cross TV

1.
School of Information Science and Technology,Southwest Jiaotong University Chengdu 610031;
2.
Engineering School,Tibet University Lhasa 850000

Funds:

Supported by national social science foundation of China(12ef119); Key projects in the science & technology prograrn of science and technology department of Tibet autonomous region (Z2013B28G28/02); National undergraduate training programs for innovation and entrepreneurship (201210694019).

摘要

摘要: 当前的各种TV(total variation)算法均只利用待修复点及其邻域的4个点的信息进行修复，由于所提供的参考信息有限，使得修复后的图像精确度欠佳. 该文提出的双十字TV算法利用原始的TV算法，将待修复点邻域中的8个点分为两组，分别利用每组4个点的参考信息计算待修复点的像素值，然后将这两个像素值进行加权平均得到最终的修复值. 实例验证结果表明，在不增加时间复杂度的情况下，双十字TV算法有效提高了修复后图像的精确度.
- 双十字TV算法 /
- 图像修复 /
- 修复精度 /
- 邻域信息
Abstract: The various total variation (TV) algorithms currently reconstruct the lost or deteriorated parts of images by related information of damaged pixel and its 4 neighborhood pixels. Their inpainting accuracy is low because of finite related information. The double cross TV algorithm proposed in this paper divides 8 neighborhood pixels into 2 groups, and computes the pixel value of damaged pixel by using 4 neighborhood pixels in each group respectively. Therefore, the weighted mean of these two pixel values is the final inpainting pixel value. The Experiments show an improvement in PSNR and less iteration number compared with the original TV.
- double cross TV algorithm /
- image inpainting /
- inpainting accuracy /
- neighborhood information