Purdue University and an international coalition of researchers are developing technologies for the automated assessment of digital images through an end-to-end platform. If successful, the platform could help the U.S. military detect manipulations of open source images.
Visual media can be doctored by a wide range of sophisticated image- and video-editing applications, and authenticating the integrity of a potentially valuable image can be difficult.
"Many tools currently available cannot be used for the tens of millions of images that are out there on the Net," says Purdue professor Edward Delp. "They take too long to run and just don't scale up to this huge volume."
As part of a four-year project funded by the U.S. Defense Advanced Research Projects Agency, researchers are working to create an end-to-end system capable of gleaning useful data from the massive volume of media uploaded regularly to the Internet. Specialized machine-learning computers will be designed to automatically verify the authenticity of uploaded images and videos.
Delp says once the technique is fully developed, the system will be available to anyone in the field of media forensics, not just the intelligence community.
From Purdue University News
View Full Article
Abstracts Copyright © 2016 Information Inc., Bethesda, Maryland, USA