This paper presents a novel contact-free biometric identification system based on geometrical features of the human hand. The right hand images are acquired by a commercial modified webcam with a 320×240 pixels resolution. The hand is illuminated by an infra-red light to solve segmentation problems in a real environment. The geometrical features are obtained from the binarized images and consist in normalized measures of the index, middle and ring fingers. A Support Vector Machines is used as verifier. A decision level fusion has been used for the final recognition with an EER of 3.4%