Murdoch University Research Repository

Welcome to the Murdoch University Research Repository

The Murdoch University Research Repository is an open access digital collection of research
created by Murdoch University staff, researchers and postgraduate students.

Learn more

Facial biometrics for forensics

Vala, Hiralba (2021) Facial biometrics for forensics. Masters by Coursework thesis, Murdoch University.

[img]
Preview
PDF - Whole Thesis
Download (912kB) | Preview

Abstract

Facial biometrics works by verifying and identifying a person based on behavior and physical behavior. Biometric plays a significant role in today's safety environment. Face recognition technology stems from the use of providing secure access to Smartphones, identifying criminal suspects to identify criminal suspects from photographs considered as a tool of the justice system. Recently, facial biometrics have been added to this list and introduced one of the safest ways to verify authenticity. The techniques and methods of face biometrics are a face-to-face comparison system that verifies facial features Among the growing availability of observational evidence and expert comparisons based on experts is considered court approval, which ensures its effectiveness is very important. Face comparisons are often performed using morphological analysis (MA), a highly based method of the untested feature. Face recognition has become an important and common tool used by crime investigators. Compared with automatic facial recognition, forensic facial recognition is more demanding because it has to be able to handle images taken under adverse conditions and has a greater responsibility to follow legal procedures. Such systems assist users in collecting and analyzing biometric data, and support decision-making by human behavior pattern and / or validation.

The challenges and future trends of conventional in-depth learning methods are described in detail to assist the researcher. In addition, traditional and in-depth learning methods of various body-based systems are analyzed for evaluation. Advances in forensic facial recognition by research on facial aging, facial recognition, sketch recognition, face recognition in video, infrared proximity, and the use of soft biometrics will be discussed. Other studies examining facial comparisons have shown low accuracy and reliability in a variety of situations. This can happen with printed images, electronic images, and even similar appearance. A number of issues are related to court approval and the credibility of the system. To date, there is no application available that automatically compares the suspect's image with the mug shot database and gives a result that can be used in court. The fact that biometric face recognition in most cases can be used for forensic purposes is true but issues related to the integration of technology and the legal system of the court are still being resolved. There is a great need for research that is multidisciplinary in nature and that will integrate face recognition technology with existing legal systems.

Item Type: Thesis (Masters by Coursework)
Murdoch Affiliation(s): Medical, Molecular and Forensic Sciences
Supervisor(s): Chapman, Brendan
URI: http://researchrepository.murdoch.edu.au/id/eprint/63402
Item Control Page Item Control Page

Downloads

Downloads per month over past year