Title :
Predicting ethnicity and gender from iris texture
Author :
Lagree, S. ; Bowyer, K.W.
Author_Institution :
Dept. of Comput. Sci. & Eng., Univ. of Notre Dame, Notre Dame, IN, USA
Abstract :
Previous researchers have reported success in predicting ethnicity and in predicting gender from features of the iris texture. This paper is the first to consider both problems using similar experimental approaches. Contributions of this work include greater accuracy than previous work on predicting ethnicity from iris texture, empirical evidence that suggests that gender prediction is harder than ethnicity prediction, and empirical evidence that ethnicity prediction is more difficult for females than for males.
Keywords :
image texture; iris recognition; ethnicity prediction; gender prediction; iris texture; Accuracy; Decision trees; Detectors; Iris; Iris recognition; Training; Training data; ethnicity prediction; gender prediction; iris biometric; soft biometric; texture analysis;
Conference_Titel :
Technologies for Homeland Security (HST), 2011 IEEE International Conference on
Conference_Location :
Waltham, MA
Print_ISBN :
978-1-4577-1375-0
DOI :
10.1109/THS.2011.6107909