Title :
Adaptive Order Spline Interpolation for Edge-Preserving Colour Filter Array Demosaicking
Author :
Randhawa, Sharmil ; Li, Jim S Jimmy
Author_Institution :
Sch. of Comput. Sci., Eng. & Math., Flinders Univ., Adelaide, SA, Australia
Abstract :
When a single-sensor digital camera is used for full colour image capture, only one colour for each pixel is acquired through the colour filter array which is placed in front of the image sensor. CFA demosaicking is a process for the estimation of the other missing colour values in order to produce a full colour image. To preserve sharp colour edges and avoid colour artifacts, it is desirable not to interpolate across an edge. It has been shown that cubic spline interpolation has been applied successfully for CFA demosaicking. However, cubic spline interpolation utilises pixels as far as five pixels away from the point of estimation. If an image contains closer edges, cubic spline interpolation will inevitably include samples from the other side of the edge for interpolation, and a lower order spline interpolation should be applied so as to avoid pixel values from the other side of the edge. In this paper, we propose an adaptive method to apply various orders of spline interpolation depending on the location of an edge from the point of estimation. It has been shown that images with closer edges can now be demosaicked to produce more accurate outputs with less colour artifacts.
Keywords :
image colour analysis; image sensors; interpolation; optical filters; splines (mathematics); adaptive order spline interpolation; cubic spline interpolation; edge-preserving colour filter array demosaicking; full colour image capture; image sensor; single-sensor digital camera; Adaptive algorithms; Equations; Estimation; Image color analysis; Image edge detection; Interpolation; Spline;
Conference_Titel :
Digital Image Computing Techniques and Applications (DICTA), 2011 International Conference on
Conference_Location :
Noosa, QLD
Print_ISBN :
978-1-4577-2006-2
DOI :
10.1109/DICTA.2011.118