Digital image processing techniques: an application for people identification through the frontal sinus

EMILCE MOLER; JUAN IGNACIO PASTORE

MECANICA COMPUTACIONAL

Asociación Argentina de Mecánica Computacional

Lugar: Buenos Aires; Año: 2010 p. 1 - 10

The identification of a person from the comparison frontal sinuses radiographs of skulls is admitted as a valid method of positive identification whereby the pre-mortem radiograph is compared with the post-mortem one. However, considering the position and distance from which these radiographs are taken, this comparison cannot be carried out by a mere visual inspection. Digital Image Processing techniques constitute an important tool in this discipline, as they provide a quantitative description of the frontal sinus. This paper presents a method of comparison by means of the analysis of certain Shape Factors, invariant to scaling, rotation and translation. The work was carried out with real digital radiographs and software-simulated images. A numerical value that indicate the similarities between two closed curves was obtained.