RI: Small: A Computational Framework for Marking Physical Objects against Counterfeiting and Tampering

点击翻译按钮获取中文摘要

The project?s goal is to create a science for embedding information in physical objects whose manufacturing process is inherently imprecise. The team will reach their objective through the investigation of a specific problem of significant economic importance: thwarting counterfeiting, and the related problem of physical tamper-detection. Counterfeiting is a growing economic problem that has been called the ?crime of the century? by a recent manufacturing industry report, and its cost is rapidly escalating (its yearly cost to the automotive industry alone is in the tens of billions of dollars and the loss of about 250,000 U.S. jobs for the legitimate manufacturers). In terms of scientific impact, the project has excellent potential for launching a significant new field. While the marking of digital objects is a well-explored area, the creation of algorithms for placing marks in physical objects is mostly unexplored territory. In terms of industrial impact, the project also has excellent potential for profoundly improving the current ?state of the practice?, which is all too easily defeated by sophisticated counterfeiters. This is because the project?s framework assumes an adversary with powerful capabilities, such as greater manufacturing prowess than the legitimate manufacturer, and full knowledge of the algorithms used to embed marks and to read marks (i.e., no ?security through obscurity?). The project?s only assumption is that the adversary does not know a secret key used to embed the mark. The work focuses on the development of the computational algorithms necessary to resolve several difficult issues and tradeoffs about what information to embed in the object, and where/how to embed it. A solution must not increase manufacturing cost and must be usable with the current manufacturing pipeline. The approach is inherently multidisciplinary, combining information hiding, computer vision/graphics, and robust algorithms. Hence, students in the project will acquire a unique combination of skills.