Finger Edge Contour Perimeter as a Biometric Based Identification System

Abstract

Generally, personal identification system is still done by conventional techniques, such as using ID cards, passwords or PIN numbers. These systems is easily lost, damaged or forgotten. In this paper we present a new method for identifying a person with a biometric based on palm of the hand. This system does not require a peg like an identification system based on existing hand geometry. Phase of development are: first, the palm image scanned in the RGB format is converted into a gray level. Then the image is converted to binary by giving the correct threshold value. From the binary image, we carry out the edge detection process to get the contour perimeter of the palm. From the contour perimeter of the palm, we measure the length of 8 edges contour perimeter of the fingers i.e. the right edge of the thumb, the left edge of the index finger, the right edge of the index finger, the left edge of middle finger, the right edge of middle finger, the left edge of pinkie, the right edge of pinkie, and the left edge of ring finger. The identification process is performed by minimum distance, which is the smallest euclidean distance value associated with the test image and database. The experimental results giving good performance with a success of up to 90%.