ACCELEROMETER BASED DIGITAL PEN FOR HANDWRITTEN DIGIT AND GESTURE RECOGNITION

N the today’s world of fast changing technology, there is also rapid development of computer technology. Human computer interaction commonly known as HCI, technique is one of that which becomes an indispensable component of our daily life. HCI includes computer science, design, media studies, the term was popularized by Stuart k. card & Allen Newell in 1983. Here we are going to use simple pen as input devic e which track trajectories in 3 - D space by using sensor name as accelerometers & gyroscopes. An Error compensation algorithm which is based on zero velocity compensation is developed which help to reduce errors in acceleration to acquire accurate trajector y. This process is done in signal processing section which comprises of calibration, moving average filter, High pass filter which reduce high frequency noise & drift error, offset respectively. The output of signal processing is then fed to feature genera tion, feature selection & feature extraction to takes raw unstructured data & generate feature by removing low value words, to reduce set of feature & to create new feature from function of original feature respectively. Recently to improve its accuracy so me researcher starts more research to reduce the error of trajectory reconstruction