Title :
Colour Filter Array Demosaicking using Cubic Spline Interpolation
Author :
Li, Jim S. Jimmy ; Randhawa, Sharmil
Author_Institution :
Sch. of Inf. & Eng., Flinders Univ., Adelaide, SA, Australia
Abstract :
When a single image sensor is used to capture colour images, only one colour value is acquired at each pixel location. Colour filter array (CFA) demosaicking refers to the estimation of the other two missing colour values in order to produce a full colour image. In this paper, we have shown that cubic spline interpolation is suitable for CFA demosaicking for producing a smooth colour image due to the known fact that it can generate a smooth curve to fit a set of data points. In our approach to obtain a unique solution for cubic spline interpolation, the two required extra conditions were derived based on the correlation between colour planes. Our proposed method, by applying cubic spline interpolation in conjunction with a selector based on edge orientation, has been shown to outperform other existing demosaicking methods.
Keywords :
image colour analysis; image fusion; image segmentation; interpolation; optical filters; splines (mathematics); colour filter array demosaicking; cubic spline interpolation; edge orientation; image sensor; pixel location; smooth colour image; Australia; Color; Equations; Filters; Image sensors; Informatics; Interpolation; Pixel; Sensor arrays; Spline; colour filter array; cubic spline interpolation; demosaicking;
Conference_Titel :
Acoustics, Speech and Signal Processing, 2007. ICASSP 2007. IEEE International Conference on
Conference_Location :
Honolulu, HI
Print_ISBN :
1-4244-0727-3
DOI :
10.1109/ICASSP.2007.366045