VISUAL IMAGE BASED HAND RECOGNITIONS

Abstract

As the contact of present computer systems on everyday life increases, humancomputer communicationshas become more and more significant in our everyday lives. Hand area based recognitions systems exploit features on the human hand to perform identity verification.In fact, as the computing, communication and display technologies progress, the existing human computer interactionpolicies, such as keyboards, mousseslimit the boundary of the speed and openness of our communication and may be converted into a restricted access in the successful usage of computers. This paper investigates a new approach to achieve performance improvement for hand area geometry systems by recognition of area of finger and hand and also the relative area between finger and hand which overcomes the conventional hand geometry based personal verification systems offer limited performance and therefore suitable only for small scale applications. The system is in an unrestricted pose captured by a flatbed scanner based on the images of the right or left hands of human. In the pre-processing steps of the algorithm some image enhanced technique, contour detection technique and some small area removal technique are appliedof hand imagesand in the post processing steps the area of hand and finger and relative area are calculated and after that we compare the different images.