Comparing the face to the body, which is better for identification?

Abstract

As early as the nineteenth century, measurements of the face and body were used for forensic identification. It was believed that no two individuals had the exact same measurements. However, this was overtaken by fingerprint analysis because it was considered more reliable in court proceedings as the probabilities of finding matching individuals could be calculated. With the standardisation of photographs, identification primarily occurs from the face. With the ability to take measurements from photographs, why not use the body? The Army Anthropometry Survey (ANSUR) database contains anthropometric measurements of 3982 individuals. Eight facial and eight body measurements were compared to investigate whether or not there is enough information on the body to use for identification. Measurements were compared by adding one measurement to the other(s) in a stepwise approach until there were no duplicate cases where two or more individuals share the same combination of measurements. Results consistently show that less body measurements are needed to find no duplicates when compared to the face. The larger the range of each of the measurements, the less chance there is of finding a duplicate. With the combination of eight body measurements, it is possible to achieve a probability of finding a duplicate to the order of 10(-20) or 1 in a quintillion. These results are comparable with fingerprint analysis. The body is more variable than the face and should be used in identification. An advantage to using the body is that larger dimensions are easier to locate on images and not affected by facial expression.