Rahil Abouali - Biomedical engineering department Islamic Azad University Tehran, Iran
Kievan Maghooli - Biomedical engineering department Islamic Azad University Tehran, Iran

Identification by signature is a common biometric that can be used as a legal instrument for the individuals identification in the administrative and financial organizations. Since the signatures are affected by many factors such as signatory individual’s mental factors and conditions under which signature is done; and considering change of handwriting features as a result of the hand fatigue in previous studies, investigating the effect of hand fatigue caused by long writing on the dynamic and static features of handwritten signature seems necessary. In the present study the effect of fatigue on the dynamic and static features of signature is investigated. In order to investigate the signature changes before and after fatigue, forty individuals are recruited in this study. First, individuals signatures are collected before fatigue using Interlink Electronics signature pad and again their signature are stored after conducting the fatigue procedure. The generated data are given to three classifiers of SVM, MLP and KNN after conducting processing and the feature extraction. The accuracy, precision, sensitivity and specificity of three classifiers are calculated for each individual. Then, the significant change in the mean and standard deviation of features is discussed and investigated by conducting paired t-test in two fatigued and non-fatigued groups.

fatigue effects; handwritten signature; static feature;, dynamic features; on-line handwriting