The main purpose of this project is to train a neural net to recognize and identify all cards that compose a Euchre deck in an image. This will be accomplished by leveraging existing tools such as TensorFlow and existing object recognition models, expanding them to recognize playing cards, and then further training to accurately identify the specific cards in any image. There are a number of advanced levels of ability we are hoping to reach, such as recognizing overlapping, partially obscured, or truncated images of cards. An application will be built to provide an interface with which to use the prediction model that is created as a result of the training.