Intra-Modal Biometric System Using Hand-Geometry and Palmprint Texture

This paper presents an identification-verification biometric system based on the combination of geometrical and palm-print hand features. It attempts to improve the performance of existing hand-geometry or palm-print systems which no combine both methods. 1440 hand images of 144 people with 10 samples each one have been acquired by a commercial scanner with 150 dpi resolution. 80 widths of fingers are obtained from binarized images. Wavelet transform, 20 Gabor filter and derivative method are used to extract the texture features from gray-scale images. Support Vector Machine (SVM) is the main classifier used as identifier and verifier. A feature, score and decision level fusion is implemented. An accuracy of 99.97% and an EER=0.0032% (Equal Error Rate) shown the final results of our system.