Evaluation of Correction Methods of Lateral Chromatic Aberration in Digital Eye Fundus Images

Abstract

This paper focuses on the lateral chromatic aberration correction to eliminate the chromatic aberration effect with image processing methods. Images were captured by the portable non-mydriatic eye fundus orbital camera that has no achromatic lenses. Image processing methods for correcting the lateral chromatic aberration effect correction try to scale the fringed colour channels so that all channels spatially overlap each other correctly in the final image. Classical algorithms are based on different patterns evaluation to derive an appropriate model for the correction. In this research, we have shown that these patterns can be successfully replaced with a map of detected eye fundus blood vessels. A comparison of the pattern (chessboard and circles) based method and the proposed method was accomplished using five different correction models: simple, affine, projective and two radial ones with one and two coefficients. Quality of images was measured and evaluated using Blur Metric, Chromatic Zipper and Achromatic Zipper metrics. The results have showed that the chromatic aberration correction using both methods have a significantly improved quality of the original images.