Biometric identity Authentication System Using Hand Geometry Measurements

Abstract

In recent years hand geometric dependent biometric system has shown to be the quite acceptable biometric trait and suitable for security applications. It has been recognized as an effective means of authenticating identity in a variety of commercial applications as a result of better hardware and improved algorithms. This paper purpose a hand recognition system that extract 21 features for the right hand to identify and authorize persons. The system has two main parts, the first contain the data collection, explains the basic pre-processing required and how hand geometry characteristics like fingers length, width, coordinates of the base of the fingers, and palm width are extracted to derive the features used for discrimination, While the second part include the training and testing of three artificial neural networks to perform the recognition. After features extraction, the system uses three kinds of artificial neural networks in performing the recognition process, which are feed forward back propagation NN, Elman NN, and the cascade forward neural network NN. The proposed system shows that the Recognition Rate RR for the neural networks after testing were 95%, 92%, 88% respectively.

A digital signature system based on hand geometry - Survey

“…People are classified and distinguished depending on their hand traits.Some of them demonstrated a multi-standard method based on the hand and other important metrics that improves the system's security. We also highlight the differences in accuracy rates amongst studies, with the lowest accuracy rate in [22] being about 88 percent. In [15] [9], the greatest accuracy rate is around 100 percent.…”

“…They have proposed in the future to develop a matching algorithm of hand geometry and palm print features. In the meantime, the article [22], introduced a reliable approach for biometric system based on measurements of hand geometry. Objective of this work is to increase the efficiency by enhanced reducing Magnitude of features like length of the finger, the finger width, and palm width.…”

“…The user is asked to put his right hand on the device board. The palm faces down and the pegs are being used as control points for setting the hand's proper position [22]. The biometric sample is most often obtained by a CCD camera or by a scanner.…”

“…White has the highest intensity while Black is the weakest. Using a Threshold, this process called Binarization is transformed to Binary Image using image thresholding procedure as follows [22]:…”

Different Biometrical Features for Detecting Human Intrusion Using Artificial Intelligence: Literature Review

“…One of these methods is to use that information as input parameters to the convolutional neural network in a back-propagation method. This idea was implemented by many researchers [12] [42]. Many authors use the same method, but by taking the measurements of the four fingers only and identifying the owner using Artificial Neural Network (ANN) and distance based nearest neighbor (DBNN) algorithms [39][41], and [68] which is used BPNN: backpropagation NN.…”