Off-Line Arabic Signature Verification Using Geometrical Features

In this paper, it has been presented an off-line signature verification scheme using the combination of geometrical and grid features. First, the geometrical features use four border points in the direction of right and left, up and down in addition to two lengths between vertical and horizontal directions. The first two points are connected between left point and the right. The second two points are connected from highest point to the lowest point. The second test uses the grid features from the core of the signature image. The sample population for this study consists of 360 personal signatures collected from 60 Arabic native speakers and all the specimen signatures are stored in Oracle database. One specimen signature is fetched from the Oracle database for this study. The combination of two methods namely geometrical and grid features reduces the accuracy of the algorithm under False Rejection Rate (FRR) scale which is proved to be less than 0.16.