Abstract
Computer forensics has emerged as an important area of research due to the increasing rise of criminal activities that involve the transmission and usage of images via the Internet. We present a method for mining images that contain objects and/or scenes that may be linked to criminal behaviour. The method allows investigators to construct these objects and scenes from components and their relationships, and to provide relevance feedback on search results in order to improve the accuracy of future search.
Brown, Ross and Pham, Binh and De Vel, Olivier (2005) Design of a Digital Forensics Image Mining System. In Proceedings The Workshop on Intelligent Information Hiding and Multimedia Signal Processing (IIHMSP05), Melbourne.
Brown, Ross and Pham, Binh (2005) Image Mining and Retrieval Using Hierarchical Support Vector Machines. In Chen, Yi-Ping, Eds. Proceedings 11th International Conference on Multi-Media Modeling, pages pp. 446-451, Melbourne, Australia. Brown, Ross and Pham, Binh and De Vel, Olivier (2003) A Grammar for the Specification of Forensic Image Mining Searches. In Lovell, Brian and Campbell, Duncan and Fookes, Clinton, Eds. Proceedings Eighth Australian and New Zealand Intelligent Information Systems Conference, pages pp. 139-144, Sydney, Australia.