Title :
Local correlation based CFA interpolation
Author :
Aleksic, Marko ; Lukac, Rastislav
Author_Institution :
Bell Canada Multimedia Lab., Univ. of Toronto, Ont., Canada
Abstract :
A new color filter array (CFA) interpolation scheme for single-sensor digital cameras is presented. The proposed method produces the restored full color image using edge adaptive interpolation and correction steps performed on the color-difference planes. Correction steps are applied to improve initially interpolated color components in regions that exhibit a correlation between the blue/green and red/green channels higher than a predetermined threshold. The scheme consistently outperforms well-known CFA interpolation algorithms, in terms of both objective and subjective image quality measures, and produces visually pleasing images.
Keywords :
cameras; computer vision; correlation methods; edge detection; image colour analysis; image restoration; interpolation; optical filters; CFA interpolation; blue/green channels; color filter array; color-difference planes; edge adaptive correction; image quality; local correlation; red/green channels; restored full color image; single-sensor digital cameras; Color; Digital cameras; Educational institutions; Gray-scale; Humans; Image quality; Image restoration; Interpolation;
Conference_Titel :
Electrical and Computer Engineering, 2004. Canadian Conference on
Print_ISBN :
0-7803-8253-6
DOI :
10.1109/CCECE.2004.1345233