Article ID Journal Published Year Pages File Type
537968 Displays 2009 13 Pages PDF
Abstract

Thin Film Transistor (TFT) screens are replacing traditional Cathode Ray Tube (CRT) screens and are becoming more common in the workplace. Traditional light sensing pens do not work effectively on a TFT screen, as the TFT screen does not generate a scan-line. A novel prototype computer system has been created that depicts where a user is pointing. The new sensor may replace the need for a touch-screen, mouse or keyboard. The system flashes a kaleidoscope of colours onto a TFT screen. A tri-colour photodiode detects these colours. A problem with this solution was the limited number of colours that could reliably be detected by the photodiodes and the time that the colours needed to remain on the screen. This paper describes a new system that includes an Artificial Neural Network (ANN) to predict the future position of the colour sensor on a computer screen. This information was used to place the optimum colours closest to the sensor in order to increase the accuracy of the reported position.

Related Topics
Physical Sciences and Engineering Computer Science Hardware and Architecture
Authors
, ,