Centroid extraction from Hartmann-Shack images using swarm clustering approach

Image analysis of Hartmann-Shack Wavefront sensor is practically useful to extract complex refractive aberrations in the retina. However due to noises such as lens glares, reflections, occlusions or extreme aberrations, pinpointing centroid locations of these lenslets prove to be a challenging problem. In this paper we propose a novel automatic extraction of lenslet foci using Rapid Centroid Estimation (RCE) and Artificial Centroid Injection (ACI). Using this method we are able to extract as much as 86% of total lenslets´ centroids before injection and 97% of total lenslets´ centroids after injection on average based on experimental data on 20 Hartmann-Shack images of off-axis aberration in the human eye. Our technique advantage is that it does not require any prior fine tuning.