Image inpainting base on color differences and structure differences

Author

Yongsheng Xu ; Shuwen Wang

Author_Institution

Electr. Eng. Dept., Northwest Univ. for Nat., Lanzhou, China

fYear

2013

fDate

12-13 Oct. 2013

Firstpage

364

Lastpage

368

Abstract

The quality of the repaired image will be directly influenced by the error accumulation of initial repaired for the target block in the exemplar-based algorithm. According to analysis of the pixel´s surrounding image features, a new algorithm which joint structure characteristics and color differences is proposed in this paper. This algorithm considers the color differences and structure characteristics simultaneous, the experimental results show that the repaired image´s vision quality can be increased effectively.

Keywords

feature extraction; image colour analysis; color differences; error accumulation; exemplar-based algorithm; image inpainting; joint structure characteristics; pixel surrounding image features; repaired image vision quality; structure differences; Algorithm design and analysis; Computers; Educational institutions; Image color analysis; Maintenance engineering; Optimization; Signal processing algorithms; color differences function; image inpainting; structure differences function; the best sample block;

fLanguage

English

Publisher

ieee

Conference_Titel

Computer Science and Network Technology (ICCSNT), 2013 3rd International Conference on

Conference_Location

Dalian

Type

conf

DOI

10.1109/ICCSNT.2013.6967130

Filename

6967130