Gestures as Interface for a Home TV Digital Divide Solutions through Inertial Sensors

Seniors are the fastest growing segment of populations not only in many parts of Europe, but also in Japan and the United States. ICT technologies are not very popular among many elderly and also are not designed around their cultural necessities and ergonomic needs. The risk is that in the very near future this growing segment will be digitally isolated, in a society that is more and more based on ICT as infrastructure for service, and communications. Easy Reach Project proposes an ergonomic application to break social isolation through social interaction to help the elderly to overcome barrier of the digital divide. This paper focuses its attention on the development of the technology and algorithms used as Human Computer Interface of the Easy Reach Project, that exploits inertial sensors to detect gestures. Many experimental algorithms for gesture recognition have been developed using inertial sensors in conjunction with other sensors or devices, or by themselves, but they have not been thoroughly tested in real situations, they are not devoted to adapt to the elderly and their way of executing gestures. The elderly are not used to modern interfaces and devices, and – due to aging – they can face problems in executing even very simple gestures. Our algorithm based on Pearson index and Hamming distance for gestures recognition has been tested both with young and elderly, and was shown to be resilient to changes in velocity and individual differences, still maintaining great accuracy of recognition (97.4% in user independent mode; 98.79% in user dependent mode). The algorithm has been adopted by the Easy Reach consortium (2009-2013) to pilot the human machine gesture-based interface.