Understanding Biometric Entropy and Iris Capacity: Avoiding Identity Collisions on National Scales
This addresses the challenge of ensuring unique biometric identification on national scales, such as for national ID systems, though it is incremental as it builds on existing combinatorial theory and empirical data.
The paper tackles the problem of identity collisions in large-scale iris recognition by deriving a general solution to determine the population size at which collisions become likely, based on biometric entropy and decision thresholds. It shows that the entropy from a person's two iris patterns is sufficient for global uniqueness, referencing empirical data from 1.2 trillion comparisons in NIST trials.
The numbers of persons who can be enrolled by their iris patterns with no identity collisions is studied in relation to the biometric entropy extracted, and the decision operating threshold. The population size at which identity collision becomes likelier than not, given those variables, defines iris "capacity." The general solution to this combinatorial problem is derived, in analogy with the well-known "birthday problem." Its application to unique biometric identification on national population scales is shown, referencing empirical data from US NIST (National Institute of Standards and Technology) trials involving 1.2 trillion (1.2 x 10^(12) ) iris comparisons. The entropy of a given person's two iris patterns suffices for global identity uniqueness.