Hand written digits recognition on STM32F4

Demo

Link to demo video : https://www.youtube.com/watch?v=_YbzxUagpbw

CNN MODEL

Simple CNN Model Is trained on MNIST dataset for handwriten digit recignition. Tensorflow 2.0 with Keras is used for training and saving the model. Link to the python code: https://github.com/nimaaghli/Mnist_TF2

STMCUBEMX.AI

STMCUBEMX.AI is used to generate C code from the trained model. Project is generated for SW4STM32 using STM32CUBEMX. SW4STM32 can be used to import and falsh the board or STM32AI_MNIST.ioc file can be loaded to CUBEMX to generate the code for diffrent softwares.

Hardware

This Demo runs on STM32F429I-DISC1 baord only, but can be easly modifed to be used on any STM32F4 board