Professors Devise Way to Detect Secret Data in Photos
A pair of professors at Iowa State University (ISU) are teaming with law-enforcement authorities in the state to create a better way to locate secret information concealed in photos that are transmitted over the Internet.
Professors Clifford Bergman and Jennifer Davidson have trained current software to determine whether child pornographers and terrorists have included secrets in harmless-looking photos found on the Internet. The project was financed by a Midwest Forensics Resource Center grant for $80,000.
The professors are currently awaiting approval to dispense the technology from the ISU Research Foundation, which takes care of intellectual property rights matters. The professors stated they hoped to hand out the software to law-enforcement personnel at no cost within the coming few months.
Bergman and Davidson trained the software to find the existence of secret information by placing concealed images in 1,200 photos in various ways until they had 10,000 images with different information included. They then introduced the vast database of images to the software, so it could tell the difference between a clean and an altered image.
The university’s new strategy should offer a more accessible and dependable way for authorities to discover whether or not a photo has a concealed image.
Although commercial software now exists for these applications, it is costly and hard to study, as the technology utilized is a trade secret, Bergman and Davidson claim.
Abstracted by the National Law Enforcement and Corrections Technology Center (NLECTC) from the Des Moines Register (IA), (11/19/06); Rossi, Lisa.