Person Identification on the Basis of Footprint Geometry

With a number of emerging biometric applications there is a dire need of less expensive authentication technique which can authenticate even if the input image is of low resolution and low quality. Foot biometric has both the physiological and behavioral characteristics still it is an abandoned field. The reason behind this is, it involves removal of shoes and socks while capturing the image and also dirty feet makes the image noisy. Cracked heels is also a reason behind noisy images. Physiological and behavioral biometric characteristics makes it a great alternative to computational intensive algorithms like fingerprint, palm print, retina or iris scan [1] and face. On one hand foot biometric has minutia features which is considered totally unique. The uniqueness of minutiae feature is already tested in fingerprint analysis [2]. On the other hand it has geometric features like hand geometry which also give satisfactory results in recognition. We can easily apply foot biometrics at those places where people inherently remove their shoes, like at holy places such as temples and mosque people remove their shoes before entering from the perspective of faith, and also remove shoes at famous monuments such as The Taj Mahal, India from the perspective of cleanliness and preservation. Usually these are the places with a strong foot fall and high risk security due to chaotic crowd. Most of the robbery, theft, terrorist attacks, are happening at these places. One very fine example is Akshardham attack in September 2002. Hence we can secure these places using low cost security algorithms based on footprint recognition.