In this work we propose the use of heuristic features to guide an evolutionary approach for 3D range image registration in forensic anthropology. The aim of this study is assist to the forensic experts in one of their main tasks: the cranio-facial identification. This is usually done by the superimposition of the 3D model of the skull on a facial photograph of the missing person. In this paper we focus on the first stage: the automatic construction of an accurate model of the skull. Our method includes a pre-alignment stage, that uses a scatter-search based algorithm and a refinement stage. In this paper we propose a heuristic selection of the starting points used by the algorithm. Results are presented over a set of instances of real problems.
|Name||Advances in Soft Computing|
|Publisher||Springer Berlin Heidelberg|