%0 Electronic Article %A Amir Soleimani %+ Machine Learning and Computational Modeling Lab, Control and Intelligent Processing Centre of Excellence, School of Electrical and Computer Engineering, University of Tehran, Tehran, Iran %A Kazim Fouladi %+ Machine Learning and Computational Modeling Lab, Control and Intelligent Processing Centre of Excellence, School of Electrical and Computer Engineering, University of Tehran, Tehran, Iran %A Babak N. Araabi %+ School of Cognitive Sciences, Institute for Research in Fundamental Sciences (IPM), Tehran, Iran %+ Machine Learning and Computational Modeling Lab, Control and Intelligent Processing Centre of Excellence, School of Electrical and Computer Engineering, University of Tehran, Tehran, Iran %K genuine signatures %K UTSig dataset %K minimum log likelihood ratio cost %K end points %K random forgeries %K signing period %K opposite-hand signatures %K equal error rate %K signature box size %K branch points %K University of Tehran Persian Signature %K Persian offline signature dataset %K data collection procedure %K public datasets %K observable signatures %K skilled forgeries %K writing instrument %K signature verification systems %X The pivotal role of datasets in signature verification systems motivates researchers to collect signature samples. Distinct characteristics of Persian signature demand for richer and culture-dependent offline signature datasets. This study introduces a new and public Persian offline signature dataset, UTSig (University of Tehran Persian Signature), that consists of 8280 images from 115 classes. Each class has 27 genuine signatures, 3 opposite-hand signatures, and 42 skilled forgeries made by 6 forgers. Compared with the other public datasets, UTSig has more samples, more classes, and more forgers. The authors considered various variables including signing period, writing instrument, signature box size, and number of observable samples for forgers in the data collection procedure. By careful examination of main characteristics of offline signature datasets, they observe that Persian signatures have fewer numbers of branch points and end points. They propose and evaluate four different training and test setups for UTSig. Results of the authors’ experiments show that training genuine samples along with opposite-hand samples and random forgeries can improve the performance in terms of equal error rate and minimum cost of log likelihood ratio. %@ 2047-4938 %T UTSig: A Persian offline signature dataset %B IET Biometrics %D January 2017 %V 6 %N 1 %P 1-8 %I Institution of Engineering and Technology %U https://digital-library.theiet.org/;jsessionid=c0a5jntkngj7m.x-iet-live-01content/journals/10.1049/iet-bmt.2015.0058 %G EN