The aim is to design and implement a Handwriting Recognition System.
This project is highly adaptive. With sophisticated algorithm, it should be able detect any patterns. For simpler implementation, Nearest Neighborhood Algorithm could be used. Requirements:
- Touch screen from a Palm Pilot m125
- A black and white TV
- A Mega32 micro-controller
- Data Acquisition - Touch screen was not that hard to use. The device driver, therefore, should be an easy thing to write.
- Recognition Algorithm - There are a lot of choices concerning the algorithm we can use to recognize patterns. Recognition algorithms fall under two categories. We can track the motions of the stylus for feature extraction for each pattern; or we can record positions for feature extraction. We choose the latter since the former will involve a lot more complicated implementations at the software level and will surely require more computational power as offered by our microcontroller.
- Video Generation - Careful investigation of various settings that gives us isolated readings of X, Y positions reveals that the touch screen is actually a very simple device that behaves exactly like a potentiometer with a little twist, that it has four pins instead of the usual three.
Assumptions are not defined for this project.
Duration is not defined for this project.
References are not defined for this project.