A machine learning number classifier that recognizes handwritten digits powered by the Raspberry Pi.
The Raspberry Pi photographs images of handwritten digits and sends them to a web server that trains a neural network to classify each image. The neural network's prediction is then displayed on the 7 segment display.
- Handwritten digit classifier input image
- Handwritten digit classifier loading shown with Larson scanner
- Handwritten digit classifier output answer
- Hardware close up
- OpenCV
- TensorFlow
- Flask
- numpy
- os
- PiCamera
- threading
- requests
- RPi.GPIO
- os
- Raspberry Pi 3 B+
- Raspberry Pi Camera V2.1
- 7 segment display
- 74HC595 shift register
- potentiometer
- neural networks
- computer vision
- image processing
- multi-threaded programming (to run Larson Scanner)
- webserver communication
- bit shifting