A Low Cost System for Dorsal Hand Vein Patterns Recognition Using Curvelets
Ricardo Janes, Augusto Ferreira Brandã Júnior · International Conference on Systems · 2014
Biometric identification systems based on the recognition of the hands vein pattern are better than the other systems such as fingerprint identification, face, iris, retina, and hand geometry features for using not visible images to the eye, having high acceptability to users and do not require physical contact. This paper presents an identification system based on the dorsal hand vein pattern recognition, using a low cost camera to capture images with near-infrared (NIR, curve let transforms for feature detection of images and random forest classification method. The proposed method was tested on a database of 1240 images captured by the authors and the system results presented classification about recognition with low Equal Error Rate (EER).