A sparsity image inpainting algorithm combining color with gradient information

被引:0
作者
Li, Zhidan [1 ]
He, Hongjie [1 ]
Yin, Zhongke [1 ]
Chen, Fan [1 ]
机构
[1] Sichuan Key Laboratory of Signal and Information Processing, Southwest Jiaotong University, Chengdu
来源
Jisuanji Yanjiu yu Fazhan/Computer Research and Development | 2014年 / 51卷 / 09期
关键词
Gradient norm; Image inpainting; Local search strategy; Similarity; Structure sparsity;
D O I
10.7544/issn1000-1239.2014.20130071
中图分类号
学科分类号
摘要
In the existing sparsity-based image inpainting algorithms, only color information is used to measure the similarity between the exemplar to be filled and other exemplars, and the match patches of the exemplar to be filled are searched in the whole image. These decrease the structure connectivity and the neighborhood consistence of texture region, and increase the time complexity of these algorithm. To address these problems, the color-gradient distance between two exemplars is defined by the color and gradient norm information of them. Using the color-gradient distance, the new patch structure sparsity is constructed to determine the filling order, and the new match criterion is obtained to find the most similar patch. Furthermore, the size of local search region is adaptively decided by the patch structure sparsity values to decrease the time complexity of this algorithm. Also the weighting coefficients of color information and gradient information are different in different images, which is verified through experiments. Experimental results demonstrate that the proposed method has better ability to maintain the structure connectivity and the neighborhood consistence of texture area. The PSNR of repaired image by the proposed scheme is 1 dB higher than that by the existing algorithms. Additionally, the speed of the proposed scheme is about 4 to 11 times of that of the existing algorithms.
引用
收藏
页码:2081 / 2093
页数:12
相关论文
共 15 条
[1]  
Bertalmio M., Sapiro G., Caselles V., Et al., Image inpainting, pp. 417-424, (2000)
[2]  
Chan T.F., Shen J., Mathematical models for local nontexture inpaintings , SIAM Journal on Applied Mathematics, 62, 3, pp. 1019-1043, (2001)
[3]  
Wu J., Ruan Q., A curvature-driven image inpainting model based on Helmholtz vorticity equation , Journal of Computer Research and Development, 44, 5, pp. 860-866, (2007)
[4]  
Criminisi A., Perez P., Toyama K., Region filling and object removal by exemplar-based image inpainting , 13, 9, pp. 1200-1212, (2004)
[5]  
Sun J., Yuan L., Jia L., Et al., Image completion with structure propagation , 24, 3, pp. 861-868, (2005)
[6]  
Liu J., Zhang S.W., Yang W.Y., Et al., A fast image inpainting method based on hybrid similarity-distance , pp. 4432-4435, (2010)
[7]  
Chen Z., Liu L., Wang G., Exemplar-based image completion using global optimization , Journal of Computer Research and Development, 46, 1, pp. 144-150, (2009)
[8]  
Bugeau A., Bertalmio M., Caselles V., Et al., A comprehensive framework for image inpainting , 19, 10, pp. 2634-2645, (2010)
[9]  
Jemi F.D., Ebenezer J.S., Sadasivam V., Combined frequency and spatial domain-based patch propagation for image completion , Computers & Graphics, 35, 6, pp. 1051-1062, (2011)
[10]  
Lemeur O., Gautier J., Guillemot C., Exemplar-based inpainting based on local geometry , pp. 3401-3404, (2011)