Interactive Gesture Chair

Abstract

A computer operator or an office worker now-a-days spend most his time sedentary. Technology is improving day by day and we are also becoming mechanic. Office employees have to stay in front of the PC almost all the day long and this sedentary behavior is not good for health. As a consequence people of all ages can suffer through health problems. Doctor suggested to make some movements during office time so that the probability of attacking by chronic diseases decreases. Therefore many proposals had been proposed to keep people moving during work time. However, for most office workers it is difficult to achieve a considerable reduction of the time spent seated within the office environment. To promote physical activity even in such sedentary situations, this work explores the possibilities of using an interactive office furniture to smoothly integrate physical activity into the daily working routine Chair is the most frequently used furniture by the office workers. We have made a system to interact with the PC using chair. By equipping motion sensing sensors with a chair the movement of the user can be detected which can be used as input device for PC. This way, the “Interactive Gesture Chair” becomes an input device that is ubiquitously embedded into the working environment, and provides an office worker with the possibility to use the movements of his body for rotating, tilting, or bouncing a chair to intuitively control the operations in Desktop Computer. In our thesis work we have used thresholding to define the chair gestures/movements. By analyzing the result we saw that threshold based gestures vary with the variety of weight of the people i.e. the people with different weights have different threshold value for same gesture. Then we tried machine learning algorithms to define gestures so that defined gestures should work for the people of different weights. First of all we tried Euclidian distance method to define gestures. Then we tried Dynamic Time Warping algorithm to define gestures and then we tried decision tree to find a universal threshold for gestures. These defined gestures can be used to control many application of PC. We defined these gestures to control Windows Multimedia Player.