首页 > 论文 > 光学 精密工程 > 27卷 > 1期(pp:251-259)

结合梯度特征与色彩一致性的图像修复

Image inpainting using gradient features and color consistency

  • 摘要
  • 论文信息
  • 参考文献
  • 被引情况
  • PDF全文
分享:

摘要

针对现有的样本块修复算法置信度值会迅速衰减至0, 填充次序不稳定以及易产生误匹配等不足, 提出结合梯度特征与色彩一致性的图像修复算法。为获得更为稳定的填充次序, 一方面引入表征图像结构和纹理变化特征的平均梯度来计算优先权, 以保证优先填充结构部分, 并适度延伸纹理信息; 另一方面提出基于S型函数的置信度更新准则以抑制置信度项的快速衰减; 同时比较匹配块与待修复块的色彩一致性, 将其与颜色信息相结合寻找最佳匹配块, 以减轻误匹配现象。实验结果表明: 本文算法较对比算法在峰值信噪比上至少提高0.82 dB, 证明了本文算法的有效性。结果也显示: 提出的算法获得了更加稳定的修复次序, 能有效减少误匹配率, 使得修复后图像更加满足人眼的视觉需求。

Abstract

In existing exemplar-based algorithms, the confidence value tends to reach zero rapidly, filling order is unstable, and mismatch can easily occur easily. To address these problems, an image inpainting algorithm using gradient features and color consistency was proposed. To obtain a more stable filling order, mean gradient was introduced to represent the change characteristics of an image structure and texture. Mean gradient was also applied to calculate the priority to ensure that the structure was preferentially populated and the texture information was appropriately extended. A confidence update term based on an S-shaped function was also proposed to avoid rapid decay of the confidence term. Color consistency between the candidate patch and the target patch was also combined with color information to identify the most similar patch, and to reduce the false matching rate. Experimental results demonstrate that the peak signal-to-noise ratio of the proposed algorithm is at least 0.82 dB higher than that of existing algorithms, which indicates the validity of the proposed method. The results also show that the proposed algorithm can make the inpainting order more stable and can reduce the error matching rate, which brings the repaired images more in line with human visual requirements.

Newport宣传-MKS新实验室计划
补充资料

中图分类号:TP394.1;TH691.9

DOI:10.3788/ope.20192701.0251

所属栏目:信息科学

基金项目:国家自然科学基金资助项目(No.61601385, No.61603319)

收稿日期:2018-06-06

修改稿日期:2018-08-20

网络出版日期:--

作者单位    点击查看

李志丹:西南石油大学 电气信息学院, 四川 成都 610500
苟慧玲:西南石油大学 电气信息学院, 四川 成都 610500
程吉祥:西南石油大学 电气信息学院, 四川 成都 610500
谌贵辉:西南石油大学 工程学院, 四川 南充 637001

联系人作者:李志丹(dan.807@163.com)

备注:李志丹(1985-), 女, 河南周口人, 硕士生导师, 2008年于西南交通大学获得学士学位, 2015年于西南交通大学获得博士学位, 主要从事数字图像修复和图像处理方面的研究。

【1】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 C. A survey on digital image inpainting [J]. Journal of Image and Graphics, 2007, 12(1): 1-10. (in Chinese)

【2】纪强, 石文轩, 田茂, 等. 基于KL与小波联合变换的多光谱图像压缩[J]. 红外与激光工程, 2016, 45(2): 275-281.
JI Q, SHI W X, TIAN M, et al.. Multispectral image compression based on uniting KL transform and wavelet transform [J]. Infrared and Laser Engineering, 2016, 45(2): 275-281. (in Chinese)

【3】唐玲, 陈明举. 基于图像局部方向特性的自适应全变分去噪模型[J]. 液晶与显示, 2016, 31(5): 477-483.
TANG L, CHEN M J. Adaptive directional total variation denoising model based on image local direction [J]. Chinese Journal of Liquid Crystals and Displays, 2016, 31(5): 477-483. (in Chinese)

【4】BERTALMIO M, SAPIRO G, CASELLES V, et al.. Image inpainting [C]. Proceedings of the 27th Annual Conference on Computer Graphics and Interactive Techniques, New York, USA, 2000: 417-424.

【5】SHEN J H, CHAN T F. Mathematical models for local nontexture inpainting [J]. SIAM Journal of Applied Mathematics, 2001, 62(3): 1019-1043.

【6】CHAN T F, SHEN J H. Nontexture inpainting by curvature-driven diffusions [J]. Journal of Visual Communication and image Representation, 2001, 12(4): 436-449.

【7】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.

【8】ZHANG L, SHEN P Y, ZHANG SH E, et al.. Depth enhancement with improved exemplar-based inpainting and joint trilateral guided filtering [C]. Proceedings of the 2016 IEEE International Conference on Image Processing, Phoenix, AZ: ICIP, 2016, 4102-4106.

【9】WEI Y W, LIU SH G. Domain-based structure-aware image inpainting [J]. Signal Image and Video Processing, 2016, 10(5): 911-919.

【10】DENG L J, HUANG T ZH, ZHAO X L. Exemplar-based image inpainting using a modified priority definition [J]. Plos One, 2015, 10(10): e0141199.

【11】李志丹, 和红杰, 尹忠科, 等. 结合颜色和梯度信息的稀疏图像修复算法[J]. 计算机研究与发展, 2014, 51(9): 2081-2093.
LI ZH D, HE H J, YIN ZH K, et al.. A sparsity image inpainting algorithm combining with gradient information [J]. Journal of computer research and development, 2014, 51(9): 2081-2093. (in Chinese)

【12】KUMAR V, MUKHERJEE J, DAS MANDAL SK. Image inpainting through metric labeling via guided patch mixing [J]. IEEE Transaction on Image Processing, 2016, 25(11): 5212-5226.

【13】LI ZH D, HE H J, TAI H M, et al.. Color-direction patch sparsity based image inpainting using multi-direction features [J]. IEEE Transactions on Image Processing, 2015, 24(3): 1138-1152.

【14】WANG H X, JIANG L, LIANG R H, et al.. Exemplar-based image inpainting using structure consistent patch matching [J]. Neurocomputing, 2017, 269(SI): 90-96.

【15】李旭峰, 王静, 刘红敏, 等. 特征优先块匹配图像修复算法[J]. 计算机辅助设计与图形学学报, 2016, 28(7): 1131-1137.
LI X F, WANG J, LIU H M, et al.. Image inpainting using feature precedence and patch matching [J]. Journal of Computer-Aided Design & Computer Graphics, 2016, 28(7): 1131-1137. (in Chinese)

【16】WANG W L, JIA Y J. Damaged region filling and evaluation by symmetrical exemplar-based image inpainting for Thangka [J]. EURASIP Journal on Image and Video Processing, 2017, 77: 10807-10821.

【17】LEE J S, Wei K J, Wen K R. Image structure rebuilding technique using fractal dimension on the best match patch searching [J]. Multimedia Tools and Applications, 2017, 76(2): 1875-1899.

【18】何凯, 高俊俏, 卢雯霞. 基于改进置信度函数及匹配准则的图像修复算法[J]. 天津大学学报(自然科学与工程技术版), 2017, 50(4): 399-404.
HE K, GAO J Q, LU W X. Image inpainting algorithm based on improved confidence function and matching criterion [J]. Journal of Tianjin University (Science and Technology), 2017, 50(4): 399-404. (in Chinese)

【19】FAN Q, ZHANG L F. A novel patch matching algorithm for exemplar-based image inpainting [J]. Multimedia Tools and Applications, 2018, 77(9): 10807-10821.

【20】金炜, 王文龙, 符冉迪, 等. 联合块匹配与稀疏表示的卫星云图修复[J]. 光学 精密工程, 2014, 22(7): 1886-1895.
JIN W, WANG W L, FU R D, et al.. Satellite cloud image inpainting based on patch matching and sparse representation [J]. Opt. Precision Eng., 2014, 22(7): 1886-1895. (in Chinese)

【21】ZHANG X, HAMANN B, PAN X, et al.. Superpixel-based image inpainting with simple user guidance [C]. Proceedings of the 24th IEEE International Conference on Image Processing, Beijing, P.R. China: ICIP, 2017: 3785-3789.

【22】赵文昌, 李忠木. 融合改进人工蜂群和K均值聚类的图像分割[J]. 液晶与显示, 2017, 32(9): 726-735.
ZHAO W CH, LI ZH M. Image segmentation algorithm based on improved artificial bee colony and k-mean clustering [J]. Chinese Journal of Liquid Crystals and Displays, 2017, 32(9): 726-735. (in Chinese)

引用该论文

LI Zhi-dan,GOU Hui-ling,CHENG Ji-xiang,CHEN Gui-hui. Image inpainting using gradient features and color consistency[J]. Optics and Precision Engineering, 2019, 27(1): 251-259

李志丹,苟慧玲,程吉祥,谌贵辉. 结合梯度特征与色彩一致性的图像修复[J]. 光学 精密工程, 2019, 27(1): 251-259

您的浏览器不支持PDF插件,请使用最新的(Chrome/Fire Fox等)浏览器.或者您还可以点击此处下载该论文PDF