A Feature Extraction Method Based on the Pattern Spectrum for Hand Shape Biometry

In this paper a novel feature extraction methodology based on the morphological pattern spectrum or pecstrum, for a hand-shape biometric system is proposed. The image of the right hand of a subject is captured in an unconstrained pose, with a commercial flatbed scanner. The invariance to rotation and position properties of the pecstrum allow the system to avoid a fixed hand position using pegs, as is the case in other reported systems. Identification experiments were carried out using the obtained feature vectors as the input to some recognition systems based on distance classifiers, neural networks, and support vector machines, for comparison purposes. The verification case was analyzed through an Euclidean distance classifier, obtaining the acceptance rate (FAR) and false rejection rate (FRR) of the system for some K-fold cross validation experiments. In average, an Equal Error Rate of 2.31 % was obtained. The results indicate that the pattern spectrum represents a good alternative of feature extraction for biometric applications.