MatthiasKirchner_bwAbstract: Fragile Camera Sensor Noise Fingerprints

Digital camera sensor noise is now commonly used for forensic purposes, most importantly for digital camera identification. Sensor noise fingerprints have a number of desirable properties, among them high robustness against lossy JPEG compression. This talk looks at sensor noise from novel angle and focusses on noise components that are fragile to JPEG compression. We discuss various applications of fragile camera sensor noise fingerprints, including countermeasures against fingerprint-copy attacks, and mobile device authentication in image-based password replacement schemes.


Matthias Kirchner is an Assistant Professor at the State University of New York at Binghamton. His research interests focus on media security, and in particular on all aspects of media forensics and counter-forensics. Before joining Binghamton’s Electrical and Computer Engineering Department in Fall 2014, Dr. Kirchner was with the IT Security Research Group at the University of Münster, Germany, and a Post-Doctoral Fellow at the International Computer Science Institute (ICSI) in Berkeley, California. Dr. Kirchner holds a PhD (Dr.-Ing.) in Computer Science from TU Dresden, Germany.