Project Overview
Donut Blink is an AI-powered application designed to help users monitor their blinking behavior while using their computer. The app uses the computer's web camera to capture video of the user's face and analyze the video to detect when their eyes are open or closed. The app notifies the user with a red rectangle around a closed eye and a green rectangle around an open eye. The app is built using Python, TensorFlow, and Dlib library, making it a powerful tool for anyone interested in developing computer vision applications.
The AI model used in Donut Blink is based on the EfficientNetB0 pre-trained model as a feature extractor. The model has been trained on a large dataset of images, allowing it to identify patterns and features in the user's eye's states accurately. With this approach, Donut Blink can detect eye behavior with high precision and accuracy, making it a valuable tool for anyone interested in monitoring their eye health while using a computer. The app's intuitive interface and ease of use make it a great addition to anyone's toolbox, regardless of their technical expertise.
