Title :
An adaptive matching algorithm for image inpainting
Author :
Xie, Zhen ; Zhang, Fan ; Zhang, Conggui
Author_Institution :
Sch. of Comput. Sci. & Technol., Zhejiang Univ. of Technol., Hangzhou, China
Abstract :
Image inpainting has a great value in many applications such as digital culture protection, film special effects production, virtual reality simulation and excrescent objects removal from video. This paper firstly compared four different inpainting algorithms, including linear interpolation, texture synthesis, edge inward expansion and radial basis function. Their different properties for image inpainting are analyzed and after that, this paper proposed a revised inpainting algorithm. The main idea is that the damaged image may be divided into different blocks and each block has corresponding characteristics. Then our adaptive matching algorithm will automatically select appropriate method to do inpainting to each block and get optimal inpainting result. Experiments show the adaptive algorithm can do image inpainting more effectively and efficiently.
Keywords :
image matching; image reconstruction; image texture; radial basis function networks; virtual reality; adaptive matching algorithm; appropriate method; edge inward expansion; image inpainting; linear interpolation; radial basis function; revised inpainting algorithm; texture synthesis; Digital images; Educational institutions; Head; Image color analysis; Image edge detection; Interpolation; Maintenance engineering; edge inward expansion; image inpainting; radial basis function; texture synthesis;
Conference_Titel :
Electronics, Communications and Control (ICECC), 2011 International Conference on
Conference_Location :
Zhejiang
Print_ISBN :
978-1-4577-0320-1
DOI :
10.1109/ICECC.2011.6067603