Efficient image inpainting based on region segmentation and varying exemplar

LIU Yang; WANG Hao-jing; TIAN XIAO-jian; YIN Yu-mei

doi:10.3788/OPE.20101812.2656

您当前的位置：

首页 >

文章列表页 >

Efficient image inpainting based on region segmentation and varying exemplar

Article | 更新时间：2020-08-12

- Efficient image inpainting based on region segmentation and varying exemplar
- Optics and Precision Engineering Vol. 18, Issue 12, Pages: 2656-2664(2010)
- 作者机构：
  
  1. 中国科学院长春光学精密机械与物理研究所, 吉林长春 130033
  2. 中国科学院研究生院,北京 100039
  3. 吉林大学电子科学与工程学院,吉林长春 130012
- 作者简介：
- 基金信息：
- DOI：10.3788/OPE.20101812.2656
  CLC： TP391.41
- Received：01 April 2010，
  
  Revised：26 May 2010，
  
  Published Online：25 December 2010，
  
  Published：25 December 2010
- 稿件说明：
移动端阅览
刘洋, 王昊京, 田小建, 阴玉梅. 采用区域分割的变尺寸样本块高效图像修复[J]. 光学精密工程, 2010,18(12): 2656-2664

LIU Yang, WANG Hao-jing, TIAN XIAO-jian, YIN Yu-mei. Efficient image inpainting based on region segmentation and varying exemplar[J]. Editorial Office of Optics and Precision Engineering, 2010,18(12): 2656-2664
刘洋, 王昊京, 田小建, 阴玉梅. 采用区域分割的变尺寸样本块高效图像修复[J]. 光学精密工程, 2010,18(12): 2656-2664 DOI： 10.3788/OPE.20101812.2656.

LIU Yang, WANG Hao-jing, TIAN XIAO-jian, YIN Yu-mei. Efficient image inpainting based on region segmentation and varying exemplar[J]. Editorial Office of Optics and Precision Engineering, 2010,18(12): 2656-2664 DOI： 10.3788/OPE.20101812.2656.

摘要

针对现有的基于样本块的图像修复算法采用全局搜索法确定最佳匹配样本块时效率低

且易出现错误匹配块的问题

分析了影响算法效率和质量的原因

提出了基于区域分割的变尺寸样本块高效图像修复算法。首先

采用双线性收缩方法获得尺寸是待修复图像0.02~0.25倍的收缩图像

并在收缩图像中分割出预选区域作为源区域

使用自适应窗口尺寸调整规则确定修复窗口大小;然后

在预选区域中搜索最佳匹配修复块对图像进行修复。收缩图像修复完毕后

对收缩图像中修复不完全的区域采用分割子图像的方法进行修复;最后

将子图像填回到原始图像的修复区域

循环运行直到修复完毕。实验结果表明

采用本文提出的修复算法

修复效率约为现有算法的5~100倍

且具有较好的修复质量。

Abstract

As existed exemplar-based image inpainting algorithms have low efficiency and poor quality for searching the best match exemplar using global searching methods

this paper analyzes the reasons that effect on the efficiency and quality of these algorithms and proposes an image inpainting algorithm based on the regional segmentation and varying exemplar. Firstly

an original image is shrunken to a downscaling image with a size in 0.02-0.25 times that of an original one by the downscaling method

and the pre-selected regions in the downscaled image are segmented as the source regions. Then

the adjustment rule of adaptive window size is used to determine the fixed window size. By searching a best-exemplar from pre-selected region

the image is inpainted. For the inpainted downscaled image

the sub-image segmentation method is used to inpaint the regions of incomplete restoration again

and then fill them into the inpainted region of the original image. Iterating the above steps until the whole image inpainting is completed. Obtained results demonstrate that this method is 5~100 times the efficiency of the existed method

meanwhile it shows good image quality.

关键词

Keywords

references

何金海,李薇,屈磊,等. 一种改进的基于样本的图像修复方法[J]. 计算机工程,2008,34(14):182-184. HE J H, LI W, Q L, et al.. Improved method for exemplar-based image inpainting[J]. Computer Engineering, 2008,34(14):182-184. (in Chinese)[2] BERTALMIO M,SAPIRO G,CASELLES V,et al.. Image inpainting .Proceedings of SIGGRAPH,2000,New Orleans, USA, 2000,6(30):417-424.[3] CHAN T,SHEN J. Non-Texture inpainting by curvature-driven diffusions [J]. Visual of Communication and Image Representation, 2001,l2 (4):436-449.[4] CRIMINISI A, PEREZ P, TOYAMA K. Object removal by exemplar-based inpainting . Computer Vision and Pattern Recognition. USA: IEEE Computer Society Conference, 2003:721-728.[5] CRIMINISI A, PEREZ P, TOYAMA K. Region filling and object removal by exemplar-based image inpainting[J]. IEEE Trans. of Image Processing, 2004,13(9):1200-1212.[6] 朱霞,李宏,张卫. 一种基于颜色区域分割的图像修复算法[J]. 计算机工程, 2008,34(14):191-193. ZHU X, LI H, ZHANG W. Image inpainting algorithm based on color region segmentation [J]. Computer Engineering and Design, 2008,34(14):191-193. (in Chinese)[7] 王昊京,王建立,王鸣浩,等. 采用双线性插值收缩的高效图像修复方法[J]. 光学精密工程,2010,18(5):1234-1241. WANG H J,WANG J L,WANG M H,et al.. Efficient image inpainting method based on bilinear interpolation downscaling[J]. Opt. Precision Eng.,2010,18(5):1234-1241. (in Chinese)[8] 张红英.数字图像修复技术的研究与应用 . 成都:电子科技大学,2006. ZHANG H Y. Research and application on digital image inpainting . Chengdu: Electronic Science and Technology of China,2006.(in Chinese)[9] 何小海,张菊,陶青川,等. 基于各向异性Markov随机场的三维显微图像复原[J]. 光学精密工程,2008,16(4):733-739. HE X H,ZHANG J,TAO Q CH,et al.. 3-D microscopical image restoration based on anisotropic Markov random field in wavelet domain[J]. Opt. Precision Eng.,2008,16(4):733-739.[10] 耿则勋,王振国. 改进的天文斑点图像高清晰重建方法[J]. 光学精密工程,2007,15(7):1151-1156. GENG Z X, WANG Z G. Modified high definition reconstruction algorithm of astronomical speckle images[J]. Opt. Precision Eng.,2007,15(7):1151-1156.(in Chinese)

Views

478

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Using image smoothing structure information to guide image inpainting

Fast phase unwrapping algorithm based on region segmenting with mathematical morphology

Image inpainting using gradient features and color consistency

Region segmentation of point cloud data based on improved particle swarm optimization fuzzy clustering

Automatic alignment of multiplexed beams of excimer laser system based on fluorescence imaging

Related Author

ZHANG Huaikun

DONG Zilong

LIU Jizhao

LIAN Jing

ZHANG Jiajun

Qian LIU

Kai-Hua CUI

Jia-wei LU

Related Institution

School of Electronic and Information Engineering， Lanzhou Jiaotong University

School of Information Science and Engineering， Lanzhou University

Institute of Machinery Manufacturing Technology， China Academy of Engineering Physics

Department of Precision Instrument， Tsinghua University

College of Electrical Engineering and Information, Southwest Petroleum University

AI问答

Address：No.3888 Dong Nanhu Road, Changchun, Jilin, China Postal code：130033
Tel：0431-86176855 Email：gxjmgc@ciomp.ac.cn
Technical support is provided by Beijing Founder electronics co., LTD 吉ICP备11002662号-17 京公网安备11010802024621
It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.

⁰