Abstrakt

Human Age Estimation through Fingerprint

Basavaraj Patil G.V, Mohamed Rafi

Fingerprints are the most widely used proofs for identifying the individuals.In this work, we used human fingerprints as an evidence to determine the human age. Fingerprints are extensively used for identifying individual but age estimation is an emerging field. Encouraged by the fact that human fingerprint differs in width ranging from birth to middle age but patterns remain unchanged. The two methods 2D- Discrete Wavelet Transform (DWT) and Principal Component Analysis (PCA) are used in combination to extract the features of fingerprint, support vector machines(SVM) is used as classifier. The obtained fingerprint image goes through two steps of feature extraction process and results in separate feature vectors which are then combined to produce a final future vector. The SVM classifies the fingerprint image to a respected age class by comparing final future vector with the database fingerprints. This method can be useful in crime investigations to reduce the search space of suspects.