Access to an Automated Security System using Gesture-based Passwords

ABSTRACT:

A primary goal of human gesture recognition research is to create systems which can identify specific human gestures and use them to convey information or for interactive control. Human gestures typically constitute a space of motions expressed by the body, face, or hands. Of these, hand gestures are often the most expressive and the most frequently used. Gesture recognition has potentially wide ranging applications such as recognizing sign language, medically monitoring patients' emotional states or stress levels, or allowing real­ time interactions with systems such as PCs, TVs, or mobile communication system.Recently, with the rapid development of MEMS technology, the micro-sensor’s application in gait research has become more and more widespread, relying on its small size, low cost, light weight, and high precision characteristics. We designed a gesture-based authentication system using a mobile phone equipped with an acceleration sensor. In the suggested system, users could generate passwords with a gesture input, instead of a traditional numeric keypad input. The passwords for authentication do not have to be in numeric format in the suggested system, instead they can be expressed in certain series of gesture motions with a mobile phone. Proposed gestures include tapping, tilting, flipping, and shaking of the mobile phone. The system developed in this study is an activity-recognition-based system by which gestures are categorized through a filtering and a probabilistic model we developed and generate personal passwords that are to be used for authentication. The classification model categorizes the gestures based on the difference of each signal in normalized probability distribution ranges on the five axes from the acceleration sensor. The gesture-based mobile authentication scheme can provide users with a convenient and reliable access process to many automated security systems.