多尺度并行坐标插值实时图像克隆算法
详细信息    查看全文 | 推荐本文 |
  • 英文篇名:Real Time Image Cloning Based on Multi-scale Parallel Coordinates Interpolation
  • 作者:沈晔湖 ; 蒋全胜 ; 汪帮富 ; 朱其新
  • 英文作者:Shen Yehu;Jiang Quansheng;Wang Bangfu;Zhu Qixin;College of Mechanical Engineering,Suzhou University of Science and Technology;
  • 关键词:图像克隆 ; 均值坐标 ; 并行加速 ; 图像金字塔
  • 英文关键词:image cloning;;mean-value coordinates;;parallel acceleration;;image pyramid
  • 中文刊名:SJCJ
  • 英文刊名:Journal of Data Acquisition and Processing
  • 机构:苏州科技大学机械工程学院;
  • 出版日期:2019-01-15
  • 出版单位:数据采集与处理
  • 年:2019
  • 期:v.34;No.153
  • 基金:国家自然科学基金(61501451,51875380)资助项目;; 江苏省产学研前瞻性联合研究资金(BY2016044-01)资助项目
  • 语种:中文;
  • 页:SJCJ201901014
  • 页数:11
  • CN:01
  • ISSN:32-1367/TN
  • 分类号:139-149
摘要
作为一种重要的数字图像合成编辑技术,图像克隆能够将源图像中的克隆区域平滑自然地嵌入目标图像的指定位置。传统的图像克隆技术通过分析图像的梯度域并求解泊松方程来获取图像信息,具有算法复杂度高、内存消耗大的缺点,从而限制了其在高分辨率图像中的实时应用。本文提出了一种改进的均值坐标计算方法,将图像克隆转化为高效且易于实现的图像插值流程。此外本文还采用多尺度和GPGPU并行计算技术进一步提升了算法的整体运算效率,从而能够实现对100万像素图像区域的实时克隆。
        As an important digital image manipulation technique,image cloning can naturally and smoothly embed the cloned region in the source image into the specified region in the target image. In traditional image cloning technique,image information can be obtained by analyzing the gradient domains and solving Poisson equations. Its shortcomings of great algorithm complexity and large memory consumption limit the real application of traditional algorithm in high-resolution image. An improved mean-value coordinates algorithm is proposed in this paper,which transforms image cloning into efficient image interpolation procedure easy to implement. Moreover,the multi-scale and GPGPU based parallel computation technique is applied to further improve the overall operation efficiency of the proposed algorithm. Experimental results show that the proposed algorithm can realize the real-time cloning for the 1 mega pixel image region.
引文
[1]Palmer S.Vision science:Photons to phenomenology[M].Cambridge:The MIT Press,1999.
    [2]Perez P,Gangnet M,Blake A.Poisson image editing[J].ACM Transactions on Graphics,2003,22(3):313-318.
    [3]Georgiev T.Photoshop healing brush:A tool for seamless cloning[C]//Workshop on Applications of Computer Vision(ECCV2004).Heidelberg:Springer,2004:1-8.
    [4]Agarwala A,Dontcheva M,Agrawala M,et al.Interactive digital photomontage[J].ACM Transactions on Graphics,2004,23(3):294-302.
    [5]Jia J,Sun J,Tang C,et al.Drag-and-drop pasting[J].ACM Transactions on Graphics,2006,25(3):631-637.
    [6]Agarwala A.Efficient gradient-domain compositing using quadtrees[J].ACM Transactions on Graphics,2007,26(3):94-98.
    [7]McCann J,Pollard N.Real-time gradient-domain painting[J].ACM Transactions on Graphics,2008,27(3):93-99.
    [8]Farbman Z,Hoffer G,Lipman Y,et al.Coordinates for instant image cloning[J].ACM Transactions on Graphics,2009,28(3):1-9.
    [9]Tanaka M,Kamio R,Okutomi M.Seamless image cloning by a closed form solution of a modified Poisson problem[C]//Proceedings of 5th ACM Conference and Exhibition on Computer Graphics and Interactive Techniques in Asia.New York:ACM,2012.
    [10]李贝,汪浩,于佳骏,等.颜色谐调的鲁棒式图像克隆[J].系统仿真学报,2015,27(4):723-730.Li Bei,Wang Hao,Yu Jiajun,et al.Robust image cloning with color harmonization[J].Journal of System Simulation,2015,27(4):723-730.
    [11]黄美玉,陈益强,纪雯.马尔科夫随机场化的光照一致图像合成方法[J].计算机辅助设计与图形学学报,2015,27(4):691-702.Huang Meiyu,Chen Yiqiang,Ji Wen.Markov random field based image composition with luminance consistence[J].Journal of Computer-Aided Design&Computer Graphics,2015,27(4):691-702.
    [12]程皓楠,王凯,刘世光.颜色风格自适应的图像克隆算法[J].图学学报,2017,38(5):700-705.Cheng Haonan,Wang Kai,Liu Shiguang.Adaptive color-style-aware image cloning[J].Journal of Graphics,2017,38(5):700-705.
    [13]Floater M.Mean value coordinates[J].Computer Aided Geometric Design,2003,20(1):19-27.
    [14]He K,Sun J.Fast guided filter[EB/OL].https://arxiv.org/abs/1505.00996,2015.
    [15]钟宝江,陆志芳,季家欢.图像插值技术综述[J].数据采集与处理,2016,31(6):1083-1097.Zhong Baojiang,Lu Zhifang,Ji Jiahuan.Review on image interpolation techniques[J].Journal of Data Acquisition and Processing,2016,31(6):1083-1097.
    [16]王平,全吉成,赵柏宇.基于双线性插值的图像缩放在GPU上的实现[J].微电子学与计算机,2016,33(11):129-132.Wang Ping,Quan Jicheng,Zhao Boyu.Realization of image zooming in GPU based on bilinear interpolation[J].Microelectronics&Computer,2016,33(11):129-132.
    [17]Storti D,Yurtoglu M.CUDA for engineers:An introduction to high-performance parallel computing[M].Boston:AddisonWesley Professional,2015.
    [18]Sonka M,Hlavac V,Boyle R.Image processing,analysis,and machine vision[M].3rd ed.Stanford:Cengage Learning,2007.

© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号

地址:北京市海淀区学院路29号 邮编:100083

电话:办公室:(+86 10)66554848;文献借阅、咨询服务、科技查新:66554700