RANK-BASED SCORE NORMALIZATION FRAMEWORK AND METHODS FOR IMPLEMENTING SAME

摘要

A rank-based score normalization framework that partitions matching scores into subsets and normalize each subset independently. Methods include implementing two versions of the framework: (i) using gallery-based information (i.e., gallery versus galleryscores), and (ii) updating available information in an online fashion. The methods improve the detection and identification rate from 20:90% up to 35:77% for Z-score and from 25:47% up to 30:29% for W-score.

主权项

1. A method for processing measures comprising the steps of:
obtaining a test sample, obtaining a set of indexed samples, comparing the test sample to each, some, or all of the indexed samples to produce first measures, computing first ranks for the first measures within each index, forming first subsets of all or any combination of current measures, where each first subset includes current measures of the same first rank, where the current measures for this step comprise the first measures, processing the current measures within each first subset that satisfy one condition or a plurality of conditions to form processed current measures, where the processing comprises one transformation or a plurality of transformations of the current measures, and outputting the current measures and the processed current measures.