Study of Individual Feature Extraction from Range Data of Human Nose
Akio Nozawa, and Hideto Ide
Department of Electronics and Electrical Engineering, College of Science and Engineering, Aoyama Gakuin University, 10-1 Fuchinobe 5-chome, Sagamihara, Kanagawa, 229-8558 Japan
We extracted individual and gender features from range data on the human nose measured by a three-dimensional digitizer. We propose extracting individual and gender features from range data measured for the nose based on a solid model, which gives feature vectors for volume, length between vertexes, and angles around vertex. We determined elements of feature vectors by statistical analysis and authentication tests. We achieved an 83.67% individual identification rate and a 98.01% gender identification rate, verifying the effectiveness of our proposed method.
Copyright© 2003 by Fuji Technology Press Ltd. and Japan Society of Mechanical Engineers. All right reserved.