Estimating Anthropometric Marker Locations from 3-D LADAR Point Clouds

An area of interest for improving the identification portion of the system is in extracting anthropometric markers from a Laser Detection and Ranging (LADAR) point cloud. Analyzing anthropometrics markers is a common means of studying how a human moves and has been shown to provide good results in determining certain demographic information about the subject. This research examines a marker extraction method utlizing principal component analysis (PCA), self-organizing maps (SOM), alpha hulls, and basic anthropometric knowledge. The performance of the extraction algorithm is tested by performing gender classification with the calculated markers.