Reliable and User Friendly U.S Banknote Recognition Application for Visually Impaired Users on Android Smartphones
Zhuorui Yang, James M. Schafer, A. Ganz · International Journal of E-Health and Medical Communications · 2014
This paper introduces a reliable and user-friendly Android application that helps visually impaired users recognize U.S banknotes. The application relies only on an Android Smartphone and does not require any wireless data connection or back-end infrastructure. The application is proven to be robust and reliable under various environmental conditions which differ in lighting and background conditions. The overall banknote detection accuracy is over 94% in any lighting conditions (even in a dark room) or backgrounds with near real-time banknote detection time of 7 seconds. The main contribution of this paper is the use of a robust and reliable computer vision algorithm on the Android platform combined with a friendly “vision free” user interface.