Title :
Gradient based threshold free color filter array interpolation
Author :
Pekkucuksen, Ibrahim ; Altunbasak, Yucel
Author_Institution :
Center for Signal & Image Process., Georgia Inst. of Technol., Atlanta, GA, USA
Abstract :
Color Filter Array (CFA) interpolation is an integral part of image processing pipeline for single sensor digital cameras. Many CFA algorithms have been proposed over the years to improve resulting image quality. One such algorithm is the highly successful Directional Linear Minimum Mean-Square Error Estimation (DLMMSE) method. We make several observations on this algorithm and propose a new method to address those points. The proposed method yields visually pleasing results and outperforms all CFA interpolation algorithms that are included in a recent survey paper in terms of PSNR.
Keywords :
cameras; gradient methods; image colour analysis; image segmentation; image sensors; information filtering; interpolation; least mean squares methods; sensor arrays; CFA interpolation; DLMMSE method; PSNR; demosaicing; directional linear minimum mean-square error estimation; gradient based threshold free color filter array interpolation; image processing pipeline; image quality; single sensor digital camera; Arrays; Color; Estimation; Image color analysis; Interpolation; PSNR; Pixel; Bayer color filter array; Demosaicing; color interpolation;
Conference_Titel :
Image Processing (ICIP), 2010 17th IEEE International Conference on
Conference_Location :
Hong Kong
Print_ISBN :
978-1-4244-7992-4
Electronic_ISBN :
1522-4880
DOI :
10.1109/ICIP.2010.5654327