Please use this identifier to cite or link to this item: https://hdl.handle.net/1959.11/62671
Title: The Application of Forensic Artists to Produce Facial Likenesses from Poor Quality Surveillance Images
Contributor(s): Spiteri, Vanessa Rose  (author); Porter, Glenn  (supervisor)orcid ; Bizo, Lewis  (supervisor); Wise, Jenny  (supervisor)orcid 
Conferred Date: 2023-02-14
Copyright Date: 2022
Handle Link: https://hdl.handle.net/1959.11/62671
Related DOI: 10.1108/JCRPP-08-2015-0039
Related Research Outputs: https://hdl.handle.net/1959.11/62594
Abstract: 

The presence of video surveillance systems within contemporary society as a crime prevention tool was founded to safeguard the community in response to increasing government and public concerns about crime and security. However, the sheer abundance and accessibility of visual evidence have provided their limitations, which have become evident in the arduous task of triaging countless hours of footage and the restrictions of poor image quality" these have impacted the potential of visual material sourced from surveillance assets. Although the automation of biometric database analysis has assisted with these limitations, problems are still evident with concerns about poor image quality. Therefore, the principal objective of this research was to investigate the potential of an intermediary step to mitigate these matters by applying the skills of forensic artists to produce the likeness of a person of interest from poor quality surveillance images. Adopting a mixed-method research design, this research involved a multidisciplinary approach to establish whether forensic artists possess the perceptual quality to interpret facial anatomy and recapture facial likeness from poor-quality images and whether these recaptures could then be searched within a face database to aid in the recognition of persons of interest. Several experiments were conducted to investigate these critical research questions. As a result, new knowledge has been developed for applying a method to assist in facial identification for forensic evidentiary and intelligence purposes. This has provided a framework for future work to improve the function of images previously deemed unusable in facial database analysis due to poor quality issues. The evidence from this research positively impacts the global need for understanding how facial identification and images can be applied and their impact on law enforcement investigations. The findings address this study's research questions, that artists can sketch facial likenesses from poor quality surveillance footage stimuli that are suitable for facial database analysis, thus underlining the importance of specialist support for developing forensic facial image analysis methods within law enforcement identification units and for forensic artists.

Publication Type: Thesis Doctoral
Fields of Research (FoR) 2020: 350305 Forensic intelligence
440216 Technology, crime and surveillance
490504 Forensic evaluation, inference and statistics
Socio-Economic Objective (SEO) 2020: 230403 Criminal justice
230404 Law enforcement
HERDC Category Description: T2 Thesis - Doctorate by Research
Description: Please contact rune@une.edu.au if you require access to this thesis for the purpose of research or study
Appears in Collections:School of Humanities, Arts and Social Sciences
School of Psychology
Thesis Doctoral

Files in This Item:
2 files
File Description SizeFormat 
Show full item record
Google Media

Google ScholarTM

Check


Items in Research UNE are protected by copyright, with all rights reserved, unless otherwise indicated.