TECHNOLOGY

Image Categorization System for Digital Forensics Applications

Print Save as PDF

Overview

Computerized image categorization systems are used to speed the identification and categorization of digital images. Categorizations for these systems are used in law enforcement and security where the systems can be used to identify missing people or identify a scene such as a terrorist training camp.

Researchers at the University of Maryland have developed a new, more effective image categorization system based on machine-learning methods. The new system is trained through exposure to sample images that represent the categories for classification. These images are analyzed in segments for similarities such as backgrounds. Once the system has been trained, new images are categorized by comparing their segments to the known image segments that represent the various categories. Using this new method, the trained system is highly accurate. A patent application is pending.

Advantages
• The system allows users to categorize images automatically, which saves costs in manpower.

Applications
• This software has applications in digital forensics, security, and entertainment.

For additional information, please contact the Office of Technology Commercialization, University of Maryland College Park, via e-mail at [email protected] or phone at 301-405-3947.

Contact Info

UM Ventures
0134 Lee Building
7809 Regents Drive
College Park, MD 20742
Email: [email protected]
Phone: (301) 405-3947 | Fax: (301) 314-9502