Palmprint identification using Sobel operator

Biometric is the science of measuring human physiological and/ or behavioral characteristics for the purpose of authentication and identification. One of the human's physiological characteristics, palmprint, is gaining popularity among researchers because it is features rich. Palmprint is unique and easy to capture using a conventional digital camera. It does not change much across time. In this paper, ten right hand images from one hundred individuals are acquired using a digital camera. The hand image is pre-processed to obtain the palmprint image. The palmprint image is enhanced and resized to 60 times 60 pixels. Different Sobel operators are applied to the resized palmprint image. The Sobel image is thresholded and represented using feature vector. All of the feature vectors are compared using Hamming Distance similarity. An accuracy of 94.84 percent can be achieved using the proposed method.