Functional Clustering on a Circle Using von Mises Mixtures

This paper addresses the question of clustering density curves around a unit circle by approximating each such curve mixture an appropriate number von Mises distributions. is done first defining distance between any two either via $$L^2$$ or symmetrized Kullback–Leibler divergence. We show that both these measures yield similar results. After demonstrating simulations proposed methods work successfully, they are applied on illustrative sample Optical Coherence Tomography data.