Abstrakt
Many living organisms exhibit rough symmetry about a plane. A method of quantified analysis of surface asymmetry is therefore useful for research areas such as biology and anthropology.
Currently it is very common to use 3D surface scanners for research which brings the need for statistical methods that work with triangular meshes that these scanners produce. We have implemented a method of detection and visualization of individual asymmetry as well as its decomposition to directional and fluctuating asymmetry.
The calculation of individual asymmetry involves the search for corresponding vertices and is implemented using the advanced non-rigid registration method Coherent point drift. For comparison we also include the results obtained using a simpler registration method that works only with landmarks.
We note an improvement of the computed asymmetry map most notably in areas that are distant to all landmarks.