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

Learning to predict gender from iris images

NCJ Number
305674
Author(s)
Vince Thomas; Nitesh V. Chawla; Kevin W. Bowyer; Patrick J. Flynn
Date Published
2007
Length
5 pages
Annotation

This paper reports on the use of machine learning techniques to develop models that predict gender based on the iris texture features.

Abstract

Although there is a large body of research that explores biometrics as a means of verifying identity, there has been very little work done to determine whether biometric measures can be used to determine specific human attributes. If it is possible to discover such attributes, they would be useful in situations where a biometric system fails to identify an individual that has not been enrolled, yet still needs to be identified. The iris was selected as the biometric to analyze for two major reasons: (1) quality methods have already been developed to segment and encode an iris image, and (2) current iris encoding methods are conducive to selecting and extracting attributes from an iris texture and creating a meaningful feature vector. (Publisher abstract provided)

Downloads