Fusion of iris & fingerprint biometrics for gender classification using neural network

The field of biometrics is tremendously gaining acceptance nowadays. Gender is a significant demographic attribute that can classify individuals. There are various biometric traits that have been used to classify gender. But the accuracy provided by a single trait is always less. Hence in this paper, fusion of two biometric traits viz., iris and fingerprint, is done to classify gender. Mean and standard deviation are the features extracted from an iris image, whereas Ridge Thickness to Valley Thickness Ratio (RTVTR) is extracted from a fingerprint image. The features extracted from both iris and fingerprint images are used to train a neural network. As a result, a suitable feature vector is formed which is used for classifying gender.