Evaluation of the Optical Laser Scanning System for Facial Identification

Gonzalez-Figueroa, America (1996) Evaluation of the Optical Laser Scanning System for Facial Identification. PhD thesis, University of Glasgow.

Full text available as:
Download (33MB) | Preview


Facial reconstruction is applied when there is no information available with regard to the possible identity of the deceased and no information such as antemortem records obtainable for comparison with the postmortem records of the human remains in question. The aim of the techniques involved in facial reconstruction is to produce a face, which can be recognized as belonging to a specific person, by relatives or friends of the missing person. Once such a recognition has been made, then the specialist can apply other techniques to confirm or refute identity. The reconstructed face can be publicized in newspapers or other mass media, in order to facilitate recognition. Facial reconstruction is based essentially on the data taken from measurements of soft tissue thickness, primarily on the cadaver head and face and on the relationships between facial features of the face and underlying bone. Recently measurements of soft tissue thickness have been carried out in living persons using ultrasound. Facial reconstruction has until now been carried out by the sculpting technique. There are two methods or a combination of both that can be used to achieve the reconstruction of facial features from a skull. The first is to use soft tissue thickness tables available to reconstruct the contour of the face at selected classical points. The depths of soft tissue are represented by markers placed directly on the skull which are then connected using bands of clay or plasticine or similar materials. The facial features are then formed using the same material. The second method involves reconstruction of the anatomical features of the face, applying the observation and comparison of sites of muscle attachments and other features to sculpt the facial muscles onto the skull using their bony attachments as indicators of size and extent, then applying more clay to the depths specified in tissue thickness tables to approximate the various structures, according to the anatomy of the face. The sculptor thus rebuilds the face anatomically as it would be found in life or from anatomical dissection. A combination of both methods can be used, as is the practice in Great Britain. These two major methods represent the two principal schools involved in the development of facial reconstruction; these are the American and Russian respectively. Application of the current techniques of facial reconstruction to individual identification has proven successful in obtaining personal identification in many forensic cases. However, according to the literature, the techniques on their own are still a long way from being accepted as definitive methods for identification. The latest investigations all agree that much research remains to be carried out to produce improvement in the reconstruction techniques. Problems remain unsolved which have a great impact on the final results, especially the relationship between the details of the facial features such as the eyes, nose, lips and ears for which the underlying bone does not provide information. Recently studies have shown the importance of these features as "good indicators" of facial recognition. This study presents a method of facial reconstruction using an optical surface laser scanner system with an evaluation of the system for facial identification. The comparative analysis was carried out using facial anthropometry. The study was performed on a sample of plaster casts of skulls exhumed from a mass grave from a south American country. Photographs of missing persons thought to be of persons from this grave were also supplied to the Facial Identification Centre. These samples were all examined in the Facial Identification Centre of the Department of Forensic Medicine and Science of Glasgow University. By collecting a set of measurements and calculation of proportion indices, using computerized facial anthropometry and photogrammetry, a comparison was made between the facial reconstructions and photographs and the results are presented in this study. The computer method involves initially digitising plaster casts of skulls using a laser scanner and video camera interfaced to a computer. An average male face from a databank, is then placed over the plaster casts of skulls as a mask and the soft tissue thicknesses are modified to conform with the underlying skull. The advantage of this technique is its speed and flexibility. Nevertheless the technique is not perfect. It shares the same problems when the reconstruction is performed by sculpting; i.e., the relationship between facial features such as eyes, nose, lips and ears. Results from the study have assessed the reliability of facial reconstruction using an optical surface laser scanner system. The system has some limitations but was able to produce a good resemblance between the finished reconstructed faces and the photographs of the missing persons. The morphological assessment was supported by facial anthropometry. Results from facial anthropometry were in turn strongly supported by statistical methods. The optical surface laser scanner in fact played an important role in the positive identification of sixteen cases of the sample studied. The identification, acting as supporting evidence for more positive techniques. The technique has been shown to be useful in personal identification, acting as supporting evidence for positive techniques.

Item Type: Thesis (PhD)
Qualification Level: Doctoral
Additional Information: Adviser: Peter Vanezis
Keywords: Medical imaging, Forensic anthropology
Date of Award: 1996
Depositing User: Enlighten Team
Unique ID: glathesis:1996-75411
Copyright: Copyright of this thesis is held by the author.
Date Deposited: 19 Nov 2019 20:12
Last Modified: 19 Nov 2019 20:12
URI: http://theses.gla.ac.uk/id/eprint/75411

Actions (login required)

View Item View Item


Downloads per month over past year