(Work in Progress) card recognition project
I set out to gather and procure a dataset that I could use to train a neural network that I created to recognize different playing cards. This was my first time creating and training a model.
Steps
- Created a dataset from images of playing cards that I had taken.
- Performed augmentations on the dataset to greatly expand it.
- Trained a neural network from the expanded dataset.
- Created a script that used the model to recognize cards from camera input.
There's much more work to be done, but this was done in about an hour's time.
Libraries Used: Numpy, Tensorflow, PIL, CV2, IMGAUG