Video-Based Source Camera Identification Via PRNU-Trained CNN
Video-based source camera identification (V-SCI) task in digital forensics relies on the extracted photo response non-uniformity (PRNU) patterns that deteriorate when the electronic image stabilization (EIS)---designed to reduce handshake motion of videos in modern cameras---is activated. We present a data-driven approach to exploit PRNU signals derived from EIS video via "device-specific" neural networks implemented with a novel PRNU image training and transfer learning strategy. Our technique is motivated by a discrepancy between how PRNU is used in real-world scenarios and existing methods' design of SCI. Specifically, a forensics examiner in the field often has possession of the device in question, which is a source of richer device-specific features besides just the reference PRNU used in current SCI methods. We benchmarked the proposed method using EIS video sequences from the well-known VISION dataset and UDAYTON22VSCI, a new dataset we collected containing video sequences from 53 modern smartphone cameras, which confirms the advantages of our approach over state-of-the-art SCI methods.
Date and Time
Location
Hosts
Registration
-
Add Event to Calendar
- Starts 20 December 2022 05:00 PM UTC
- Ends 21 December 2022 08:00 PM UTC
- No Admission Charge
Speakers
Nicholas Hopkins
Video-Based Source Camera Identification Via PRNU-Trained CNN
Biography:
Nicholas Hopkins is a lead engineer for the Air Force Research Laboratory, Materials and Manufacturing Directorate, Accelerated Materials and Processes Solutions branch (AFRL/RXNS) at Wright Patterson Air Force Base in Dayton, Ohio. In his role, Nicholas performs electrical system/component root cause failure analysis to support aerospace system mishap investigations with applicability to all aerospace vehicles owned and operated by the United States Department of Defense. Prior to Nicholas’ current assignment, he was a radar systems engineer for the Air Force Life Cycle Management Center. Other assignments have included performing missile system modeling and simulation for Booz Allen Hamilton, Bayesian inference electrical load modeling for NASA, and F-16 maintenance as an aircraft electrical and environmental systems craftsman while a Staff Sergeant with the United States Air Force. In 2018, Nicholas was a finalist for AFRL/RX’s Engineering Expertise Award and in 2012, was one of twenty college undergraduates in the nation to be awarded a prestigious NASA Aeronautics Scholarship. Nicholas holds bachelor’s and master’s degrees in electrical engineering and is currently on long term leave from AFRL/RXNS to complete his Ph.D. in machine learning-based digital forensics at the University of Dayton.
Email: