Recognizing and quantifying human movement patterns through haptic-based applications

Abstract

Biometrics has been introduced recently to identify people by their behavior and physiological features. It offers a wide application scope to detect fraud attempts in organizations, corporations, educational institutions, electronic resources and even crime scenes. The field of biometrics can be divided into two main classes according to features that humans are born with, such as fingerprints or facial features, or behavioral characteristics of humans, like a handwritten signature or voice (J. Ortega-Garcia et al., 2004). The work presented in this paper pursues the latter class, specifically how a person reacts to using daily devices or tools. The fact that we can exploit people's habits in handling devices to identity individuals was the hypothesis that motivated this work. Among the many examples of the potential use of this class of biometrics is the particular force applied to the keys in a keyboard. There is also the time interval between each keypad when dialing a telephone number. Another example that can be extracted from the latter would be the map described by the fingers in navigating through solving maze operation. Extracting these features by using a haptic-based application and defining the subsequent individual pattern is the objective of this research. A framework that identifies behavioral patterns through physical parameters such as direction, force, pressure and velocity has been built. The set up for the experimental work consisted of a multisensory tool, using the Reachin system (Reachin Technologies, User's Programmers Guide and API).