U.S. flag

An official website of the United States government, Department of Justice.

NCJRS Virtual Library

The Virtual Library houses over 235,000 criminal justice resources, including all known OJP works.
Click here to search the NCJRS Virtual Library

El Libro de Referencia de las Huellas Dactilares - Chapter 6: Sistema Automatizado de Identificacion de Huellas Dactilares (AFIS)

NCJ Number
250979
Author(s)
Kenneth R. Moses; Peter Higgins; Michael McCabe; Salil Probhakar; Scott Swann
Date Published
August 2017
Length
10 pages
Annotation

This chapter reviews the evolution, features, and associated technology of automated fingerprint identification systems (AFISs) throughout the world and suggests future directions for such systems. 

Abstract

The first live scan fingerprint readers introduced in 1988 were difficult to use and had many problems when compared to the inexpensive and relatively small sensors currently available. Within the past few decades, research and the practice of fingerprint matching and indexing have also evolved in the understanding of individuality, the information accessible in fingerprints, and efficient ways of processing this information. The declining cost of computing power and fingerprint sensors, along with the demand for security, efficiency, and convenience have made automatic fingerprint algorithms viable for daily use in a large number of applications. Still, there are a number of challenges yet to be addressed in designing a completely automatic and reliable fingerprint individualization system, especially when it comes to processing fingerprint images of poor quality. The design of automated systems still fall short of duplicating the complex data processing of a well-trained fingerprint expert in making decisions regarding the matching of individual fingerprints, especially latent prints. Automatic matching systems lack the ability to engage in such sophisticated decisionmaking mostly because of the unavailability of complex modeling and image-processing techniques that can reliably and consistently extract detailed features in the presence of "noise." There may be a need to explore radically different features of fingerprint discriminatory information, robust methods of fingerprint matching, and more ingenious methods for combining fingerprint matching and classification that are amenable to automation. 43 references