Micro signal extraction and analytics for digital forensics

SpeakerChau-Wai Wong
Organization Univ. of Maryland, College Park (UMD)
Location3002 Engineering Building II
Start Date June 30, 2017 10:00 AM
End Date June 30, 2017 11:30 AM

 Digital Forensics cluster candidate seminar

A class of challenging problems in digital forensics involves the extraction and utilization of micro signals. In these problems, the signals to be extracted have smaller magnitudes---typically by one order of magnitude or more---than the dominating signals, or have a smaller topological scale than what conventional algorithms can resolve. To extract a micro signal reliably and protecting it from being corrupted by other dominant or nuisance signals, we design extraction algorithms by closely calibrating the modeling assumptions with the behavior of the multimedia data.

We will show three examples of micro signal extraction by synergistically applying and adapting theories and techniques in signal processing and computer vision. First, we explore the use of mobile imaging to extract microscopic features of paper surfaces as an unique identifier for counterfeit detection. Second, we examine the possibility of extracting a signature of the power grid from an individual photo captured by digital camera to shine light on the geographic area where the photo was taken. Last, we present the monitoring of heart-rate from fitness exercise videos using the subtle facial color change that is invisible to naked eyes.

Chau-Wai Wong received the B.Eng. degree with first class honors in 2008, and M.Phil. degree in 2010, both in electronic and information engineering from The Hong Kong Polytechnic University, and the Ph.D. degree in electrical and computer engineering from the University of Maryland, College Park (UMD), in 2017. He is currently a data scientist at Origin Wireless, Inc., in Maryland. His research interests include multimedia forensics, signal and image processing, data analytics, and video coding.

Dr. Wong was the recipient of multiple scholarships and awards, including a Student Paper Award (for a coauthored paper applying data analytics to social science), Future Faculty Fellowship from UMD, HSBC Scholarship, Hitachi Scholarship, Chinese Government Award for Outstanding Students Abroad, two-year and four-year merit-based full scholarships for M.Phil. and B.Eng. studies, respectively. He is a member of the IEEE and the APSIPA, and served as general secretary of the IEEE PolyU Student Branch in 2006 - 2007. He contributed to organizing the IEEE Signal Processing Cup in 2016 on a location forensics topic involving power grid signature.

  June 2017
Sun Mon Tues Wed Thu Fri Sat