- We built a machine capable of recognizing handwritten digits using a camera module mounted on the DE2 Cyclone IV FPGA Board.
- We implemented a simple (pre-trained) Neural Network on the board, with about 93% accuracy when run in software.
- Empirical hardware tests suggest at least 80% accuracy on recognizing digits.
The final report for this project can be found at https://docs.google.com/document/d/1GDGmD4yhDDblcYEICTdyi9eqDBDxkWJCNs3Rulk2hdU/edit?usp=sharing
Our demo video for this project can be found at https://drive.google.com/file/d/0Bw2cgOTdZo47ZDNoM2wwQ2ZNbWc/view?usp=sharing