This deep learning application in python recognizes alphabet through gestures captured real time on a webcam. The user is allowed to write the alphabet on the screen using an object-of-interest.
The code is in Python (version 3.6 or higher). You also need to install OpenCV(4.4.0 version) and Keras (2.1.3 version) and tensorflow (2.3.1) libraries.
A popular demonstration of the capability of deep learning techniques is object recognition in image data.
The "Extended Hello World" of object recognition for machine learning and deep learning is the EMNIST dataset for handwritten letters recognition. It is an extended version of the MNIST dataset.
A set of sample images is shown below.
A Multilayer Perceptron (MLP) model as well as a Convolutional Neural Network (CNN) model using Keras library. The predictions of both the models are shown on the screen in real time.
The Test accuracies were as follows:
->MLP Test Accuracy: 91.7%
->CNN Test Accuracy: 93.1%
#Execution
Execute mlp_model_builder.py for bulding the Model architecture using Multilayer Perceptron Network
Execute cnn_model_builder.py for building the Model architecture using Convolutional Neural Network.
Execute alpha_recognition.py to run the Alphabet detection code.
Use a Blue bottle cap to Draw the letters on the Screen