参照四邻域裁剪样本的图像修复算法

doi:10.11772/j.issn.1001-9081.2017082033

计算机应用 ›› 2018, Vol. 38 ›› Issue (4): 1111-1116.DOI: 10.11772/j.issn.1001-9081.2017082033

• 虚拟现实与多媒体计算 • 上一篇下一篇

参照四邻域裁剪样本的图像修复算法

孟红月¹, 翟东海^1,2, 李梦雪¹, 曹大命¹

1. 西南交通大学信息科学与技术学院, 成都 610097;
2. 西藏大学工学院, 拉萨 850000

收稿日期:2017-08-21 修回日期:2017-11-01 出版日期:2018-04-10 发布日期:2018-04-09
通讯作者: 孟红月
作者简介:孟红月(1993-),女,河南永城人,硕士研究生,主要研究方向:数字图像处理;翟东海(1974-),男,山西芮城人,副教授,博士,主要研究方向:数字图像处理、海量数据挖掘;李梦雪(1992-),女,重庆潼南人,硕士研究生,主要研究方向:数字图像处理;曹大命(1990-),男,河北沧州人,硕士研究生,主要研究方向:数字图像处理。
基金资助:
国家自然科学基金资助项目（61461048）。

Image inpainting algorithm based on pruning samples referring to four-neighborhood

MENG Hongyue¹, ZHAI Donghai^1,2, LI Mengxue¹, CAO Daming¹

1. College of Information Science and Technology, Southwest Jiaotong University, Chengdu Sichuan 610097, China;
2. College of Engineering, Tibet University, Lhasa Tibet 850000, China

Received:2017-08-21 Revised:2017-11-01 Online:2018-04-10 Published:2018-04-09
Supported by:
This work is partially supported by the National Natural Science Foundation of China (61461048).

摘要/Abstract

摘要： 针对结构纹理信息较复杂、破损尺度较大的图像修复问题，提出一种既能保持图像特征又能提高修复速度的参照四邻域裁剪样本的修复算法，将图像修复问题转化为最佳样本的检索过程。首先，提取图像结构信息，并对图像进行区域划分以缩小样本的裁剪与检索范围；其次，为了改进离差平方和（SSD）方法对块的结构信息匹配的忽视，在像素块匹配计算中引入结构对称匹配约束，有效避免了误匹配，提高了图像块匹配精度及样本搜索效率；然后，通过引入结构因子和置信度，结合传统的优先权计算，得到突出结构作用的优先级公式；最后，利用目标块与四邻域块间的重叠区域计算四邻域参照优先级，并根据四邻域提供的可靠参照信息，依据改进的块匹配方法裁剪样本集并检索最佳样本块，直至所有目标块都检索匹配到最佳样本，完成修复。实验结果表明，该算法可以很好地解决纹理模糊和结构错位等问题，在提高图像修复速度的同时，所提算法修复效果的峰值信噪比（PSNR）比其他对比算法平均提高了0.5~1 dB，使得修复后的图像更好地满足视觉连通性，同时能高效地修复一般区域，具有更好的普适性。

关键词: 图像修复, 结构对称匹配约束, 结构因子, 四邻域参照优先级, 裁剪样本

Abstract: To inpaint the image with large damaged region and complex structure texture, a new method based on neighborhood reference priority which can not only maintain image character but also improve inpainting speed was proposed, by which the problem of image inpainting was translated into the best sample searching process. Firstly,the structure information of target image was extracted, and the sample region was divided into several sub-regions to reduce the sample size and the search scope. Secondly, in order to solve the problem that Sum of Squares of Deviations (SSD) method ignores the matching of structure information, structure symmetry matching constraint was introduced into matching method, which effectively avoided wrong matches and improves sample matching precision and searching efficiency. Then, priority formulas which highlights the effect of structure was obtained by introducing structure weight and confidence and combining the traditional priority calculation. Finally,the priority of four-neighborhood was got by computing overlapping information between target block and neighborhood blocks patches, according to the reliable reference information provided by four-neighborhood and the improved block matching method, the samples were pruned and the optimal sample was retrieved. The inpainting was completed until all the the optimal samples for all the target blocks were retrieved. The experimental results demonstrate that the proposed method can overcome the problems like texture blurring and structure dislocations and so on, the Peak Signal-to-Noise Ratio (PSNR) of the improved algorithm is increased by 0.5 dB to 1 dB compared with the contrast methods with speeding up inpainting process, the recovered image is much continuous for human vision. Meanwhile, it can effectively recover common damaged images and is more pervasive.

Key words: image inpainting, structure symmetry matching constraint, structure factor, four-neighborhood reference priority, pruning sample

中图分类号:

TP317.4

孟红月, 翟东海, 李梦雪, 曹大命. 参照四邻域裁剪样本的图像修复算法[J]. 计算机应用, 2018, 38(4): 1111-1116.

MENG Hongyue, ZHAI Donghai, LI Mengxue, CAO Daming. Image inpainting algorithm based on pruning samples referring to four-neighborhood[J]. Journal of Computer Applications, 2018, 38(4): 1111-1116.

参考文献

[1] BERTALMIO M, SAPIRO G, CASELLES V, et al. Image inpainting[C]//SIGGRAPH 2000:Proceedings of the 27th Annual Conference on Computer Graphics and Interactive Techniques. New York:ACM, 2000:417-424.
[2] CHAN T F, SHEN J. Nontexture inpainting by curvature-driven diffusions[J]. Journal of Visual Communication and Image Representation, 2001, 12(4):436-449.
[3] CHAN T F, SHEN J. Mathematical models for local nontexture inpaintings[J]. SIAM Journal on Applied Mathematics, 2001, 62(3):1019-1043,.
[4] 张红英, 彭启琮.数字图像修复技术综述[J]. 中国图象图形学报, 2007, 12(1):1-10.(ZHANG H Y, PENG Q Z. A survey on digital image inpainting[J]. Journal of Image and Graphics, 2007, 12(1):1-10.)
[5] DRORI I, COHEN-OR D, YESHURUN H. Fragment-based image completion[J]. ACM Transactions on Graphics, 2003, 22(3):303-312.)
[6] CRIMINISI A, PEREZ P, TOYAMA K. Region filling and object removal by exemplar-based image inpainting[J]. IEEE Transactions on Image Processing, 2004, 13(9):1200-1212.
[7] SUN J, YUAN L, JIA J, et al. Image completion with structure propagation[J]. ACM Transactions on Graphics, 2005, 24(3):861-868.
[8] FELZENSZWALB P F, HUTTENLOCHER D P. Efficient belief propagation for early vision[J]. International Journal of Computer Vision, 2006, 70(1):41-54.
[9] KOMODAKIS N, TZIRITAS G. Image completion using efficient belief propagation via priority scheduling and dynamic pruning[J]. IEEE Transactions on Image Processing, 2007, 16(11):2649-2661.
[10] YANG Y, ZHU Y, PENG Q. Image completion using structural priority belief propagation[C]//MM 2009:Proceedings of the 17th ACM International Conference on Multimedia. New York:ACM, 2009:717-720.
[11] HSIN H-F, LEOU J-J, LIN C-S, et al. Image inpainting using structure-guided priority belief propagation and label transformations[C]//ICPR 2010:Proceedings of the 201020th International Conference on Pattern Recognition. Washington, DC:IEEE Computer Society, 2010:4492-4495.
[12] TANG C, HU X, CHEN L, et al. Sample-based image completion using structure synthesis[J]. Journal of Visual Communication & Image Representation, 2013, 24(7):1115-1123.
[13] 翟东海, 段维夏, 鱼江, 等.局部特征分类的自适应扩散模型[J]. 信号处理, 2014, 30(9):1039-1047.(ZHAI D H, DUAN W X, YU J, et al. Adaptive diffusion model based on local feature classification[J]. Journal of Signal Processing, 2014, 30(9):1039-1047.)
[14] 王猛, 翟东海, 聂洪玉, 等.邻域窗口权重变分的图像修复[J]. 中国图象图形学报, 2015, 20(8):1000-1007.(WANG M, ZHAI D H, NIE H Y, et al. Image inpainting with weight variation of neighborhood window[J]. Journal of Image and Graphics, 2015, 20(8):1000-1007.)
[15] HE K, SUN J. Image completion approaches using the statistics of similar patches[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2014, 36(12):2423-2435.

参照四邻域裁剪样本的图像修复算法

Image inpainting algorithm based on pruning samples referring to four-neighborhood

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	冯浪, 张玲, 张晓龙. 基于扩张卷积的图像修复[J]. 计算机应用, 2020, 40(3): 825-831.
[2]	王肖, 魏嘉旺, 袁玉波. 基于特征部位圆形域的人脸图像修复方法[J]. 计算机应用, 2020, 40(3): 847-853.
[3]	袁琳君, 蒋旻, 罗敦浪, 江佳俊, 郭嘉. 基于生成对抗网络的人像修复[J]. 计算机应用, 2020, 40(3): 842-846.
[4]	杨文霞, 王萌, 张亮. 基于密集连接块U-Net的语义人脸图像修复[J]. 计算机应用, 2020, 40(12): 3651-3657.
[5]	杨文霞, 张亮. 基于图像结构纹理分解及局部总变分最小化的图像修复模型[J]. 计算机应用, 2018, 38(8): 2386-2392.
[6]	杨文霞, 张亮. 基于对数函数的非局部总变分图像修复模型[J]. 计算机应用, 2018, 38(6): 1784-1789.
[7]	曹大命, 翟东海, 孟红月, 李梦雪, 冯炎. 基于先验约束和统计的图像修复算法[J]. 计算机应用, 2018, 38(2): 533-538.
[8]	林金勇, 邓德祥, 颜佳, 林晓英. 基于自适应相似组稀疏表示的图像修复算法[J]. 计算机应用, 2017, 37(4): 1169-1173.
[9]	邹玮刚, 周志辉, 王洋. 基于非降采样轮廓波变换的图像修复算法[J]. 计算机应用, 2017, 37(2): 553-558.
[10]	李梦雪, 翟东海, 孟红月, 曹大命. 划分特征子区域的图像修复算法[J]. 计算机应用, 2017, 37(12): 3541-3546.
[11]	王佳君, 喻强, 张晶晶. 基于先验置信传播的图像修复改进算法[J]. 计算机应用, 2016, 36(4): 1115-1119.
[12]	何埜, 李光耀, 肖莽, 谢力, 彭磊, 唐可. 基于深度信息的图像修复算法[J]. 计算机应用, 2015, 35(10): 2955-2958.
[13]	翟东海左文杰段维夏鱼江李同亮. 基于双十字曲率驱动扩散模型的图像修复算法[J]. 计算机应用, 2013, 33(12): 3536-3539.
[14]	韩志科王贵. 基于方向场的指纹图像偏微分方程修补模型[J]. 计算机应用, 2013, 33(10): 2886-2890.
[15]	翟东海肖杰鱼江李同亮. 基于自适应模板的图像修复算法[J]. 计算机应用, 2013, 33(10): 2891-2894.