Richard Henry Edwin Matthews – Isolating lens aberrations within fixed pattern noise.
May 11 @ 4:00 pm - 5:00 pm
Title: Isolating lens aberrations within fixed pattern noise.
Sub-Title: Extending the image forensic fixed pattern noise model to allow for lens aberrations using Sony IMX219 CMOS system on a chip – pinhole photography for digital image forensics.
The forensic technic used to link photos to their source cameras involving fixed pattern noise has been used successfully to prosecute criminals involved in child exploitation material in Scotland. The method, first published by Lukas, Goljan and Friedrich in 2005, only took 4 years before Goljan and Farid used it to the forensic standard upheld in Daubert. When compared to other forensic techniques, (Fingerprints, 69 years; DAN 33 years)* this rapid progression from conception to first criminal prosecution gives cause for concern behind the science of the methodology first proposed.
In our work, we have replicated the methodology first proposed by Lukas et al. Building upon the previous work of Knight, Moschu and Sorell, we have then applied standard image processing theory and an understanding of the geometric properties of light to continue to isolate artefacts contained within the fixed pattern noise model previously discounted as being removed. Our new, updated method creates a 15x increase in correlation over previous reported work.
Richard Matthews received the BElec degree in electrical and electronic engineering at the University of Adelaide, South Australia in 2014 and is currently studying a PhD in Engineering with his research concerning applications of sensor pattern noise for forensic applications. He is a Councillor for the University of Adelaide, the President of the Council of Australian Postgraduate Associations Central region and currently serves as the President for the University of Adelaide Postgraduate Student Association.
Prior to the Academy, he served in the Royal Australian Air Force as an Officer rank where he was awarded the Australian Defence Medal before working on industrial automation for the automotive and power monitoring industries. He now tutors first year engineering students, conducts research in his interest areas including digital forensics, cyber security and additive manufacturing as well as publishing for the Conversation.