अमूर्त

Design of Gesture Controlled Wireless Mouse Using Microcontroller

Suneeta V.Budihal, Anant R Choudhari, Kripa S Patil, Nayana P Desai.

This paper presents an advanced design of a Human Computer Interface (HCI) using wireless and accelerometer technology, for better interaction facility with computer and serves to be of great use to physically challenged people. To ensure wearable computing to be more widely accepted, the associated Human-Computer Interface will have to move past today's keyboard, keypad, touch screen, and any other bulky hand-held interfaces to allow a user to specify their input through their fingers without taking their eyes and attention off their immediate focus and without disturbed attention. As a result, this leads to invention of a wearable system to track hand gestures with the help of an accelerometer. The primary goals of the system are to have a comfortable and easily wearable device without interfering with other everyday activities while tracking particular movements of hand that could be used to control a wearable mouse or assist in interaction with ubiquitous or other wearable devices. The technique of Gesture Recognition provides an efficient Human-Computer interface for interactive and intelligent computing. This technique uses a single 3-axis accelerometer for data acquisition and comprises of two main stages: 1) Training stage and 2) Testing stage. During training, the system employs dynamic time warping as well as affinity propagation to create exemplars for each gesture while during testing, the system projects all candidate traces and also the unknown trace onto the same lower dimensional subspace for recognition. A dictionary of gestures is defined and a database of traces is created. The hardware for the same has been implemented and tested using ATMega 64.