DocumentCode
3629053
Title
A hybrid method for image colorization
Author
Bekir Dizdaroglu
Author_Institution
Bilgisayar M?hendisli?i B?l?m?, Karadeniz Teknik ?niversitesi, Trabzon, Turkey
fYear
2008
fDate
4/1/2008 12:00:00 AM
Firstpage
1
Lastpage
4
Abstract
In this paper, a hybrid method is proposed to colorize grayscale images. First, the input image is colorized by the fast colorization method, which is based on the technique of color blending obtained effectively from a weighted distance function computed from the luminance channel. Second, the neighborhood filter approach, which restores a pixel by taking a weighted average of the values of neighboring pixels in chrominance channels, is carried out to the obtained result to improve colorization in object boundaries. A number of examples and comparisons with the existing colorization method show that the proposed method efficiently betters the results.
Keywords
"Image color analysis","Chromium","Gray-scale","Pixel","Encyclopedias","Distance measurement","Optimization"
Publisher
ieee
Conference_Titel
Signal Processing, Communication and Applications Conference, 2008. SIU 2008. IEEE 16th
ISSN
2165-0608
Print_ISBN
978-1-4244-1998-2
Type
conf
DOI
10.1109/SIU.2008.4632572
Filename
4632572
Link To Document