Usability Analysis of a Handwritten Signature Recognition System Applied to Mobile Scenarios Ramon Blanco-Gonzalo, Oscar Miguel-Hurtado, Raul Sanchez-Reillo University Carlos III of Madrid University Group for Identification Technologies (GUTI) Avda. Universidad, 30; 28911 Leganes (Madrid), SPAIN {rbgonzal, omiguel, rsreillo}@ing.uc3m.es Alfredo Gonzalez-Ramirez University Carlos III of Madrid Avda. Universidad, 30; 28911 Leganes (Madrid), SPAIN algonzalezramirez@gmail.com Abstract—Along with the necessity to solve the problems due to the misuse of biometric systems and thus the consistent increase in the final products error rates, a usability evaluation on handwritten signature recognition was carried out. Furthermore, according to the popularity of mobile devices and the market trends, the evaluation was performed signing in mobile scenarios with smart phones, tablets and other common mobile devices. This study reveals interesting results correlating habituation and preferences with better or worst results and it shows the need of involve the user more incisively in the development of biometric solutions, not only for comfort issues but for better systems throughput Keywords—handwritten signature; mobile; biometrics; usability; tablet; smartphone I. INTRODUCTION Mobile devices are playing a more significant role day by day, not only for communications but for entertainment, work and social relationships among other functions. Along with the high increase of the use of smart phones and mobile devices in general, the amount of sensitive data that they store also increases and the necessity of protect it is starting to be covered by biometrics [1]. According to the Handbook of Biometrics [2] “Biometrics is the science of recognizing the identity of a person based on the physical or behavioral attributes of the individual such as face, fingerprints, voice and iris”. Then, according to the smart phones features, the range of possibilities for integrating biometrics in them is promising including for instance: sign documents univocally, safely access websites, carry on administration procedures, etc. Nevertheless, the migration of biometrics to mobile environments supposes a big challenge due to several constraints such as less space and less CPU powerful. The literature related to this topic is very extensive [3] and will be growing with the mobile “revolution”. Almost in all the cases, in this migration process, researches are more focused on reach good error rates in algorithms than in develop systems that can be really useful and user friendly. This could lead to various drawbacks like worst error rates due to a misuse of the systems and it could end with users rejecting to use the technology and preferring other approaches. There are a few works in the line of improve the user experience interacting with biometric systems. For instance, researchers studying the HCI [4] and the NIST have been working on some usability aspects [5]. There is also a method, the HBSI (Human Biometric Sensor Interaction), [6] that is focused on the error rates produced by a bad biometric trait presentation to the system and studies the origin of the Failure to Acquire (FTA). The modality used in this evaluation, the handwritten signature recognition, is widely utilized for processes like ordinary civil proceedings and the signing action itself is an action that users consider as normal. In the literature, there are multiple researches related to handwritten signature recognition and it is addressed by two main approaches [7]: online signature (using dynamic attributes of the signature, such as time and pressure) and offline signature (utilizing only the image of the signature). This work is focused on online signature. In most of the works in the literature, the signature acquisition is done with specific devices, such as digitizer tablets. There are also a few works introducing smart phones and handwritten signature altogether [8]. Nevertheless, in our research the signature information is taken from common devices made not specially for signing but for typical smart phone functions, as well as the possibility of perform the signature with the finger tip instead using a stylus is introduced. Being the handwritten signature a biometric modality widely accepted by users and being the mobile environments a trending topic, in this work it is proposed a usability analysis, in order to improve the user experience and the error rates achieved. The system analyzed uses a DTW-based handwritten signature recognition algorithm [9] in mobile scenarios. The experiment was made under a scenario evaluation and the users were invited to sign in four different devices, in five different positions and in three sessions separated one week each. Signing with the finger tip in various devices is another significant novelty. The usability parameters were measured with satisfaction forms and notes taken by a guide responsible for the evaluation. During the acquisition process carried out not verification results were given to the user. This means that all signatures were stored in the data base, independently whether they were perform properly or not. Then, after obtaining results from the verification process and usability