AUTHENTICATION OF RANDOM PATTERNS BY FINDING A MATCH IN AN IMAGE DATABASE

This paper describes an approach for finding the exact match of a rand om texture pattern in an image database. It is assumed that the random pattern may be misregistered relative to its representation in the da tabase and it may have missing parts. The specific application of the proposed algorithm is authentication of legal documents, bank notes or credit cards, where thin fibers are embedded randomly into the docume nt medium during medium fabrication. Random patterns guarantee extreme ly high uniqueness, i.e. low probability of two documents being identi cal, and thus offer very effective anti-counterfeiting measures. Image matching is achieved by a three step hierarchical procedure, where ea ch step uses a separate database. The preliminary tests of the authent ication algorithm included experiments with over 10,000 synthetic imag es and 50 real images. In all cases the algorithm has given correct re sults. This paper details the algorithm, considerations on document un iqueness, and testing procedures.