This repository contains a deep learning project focused on classifying distracted drivers using computer vision techniques. The goal of the project is to build a robust model that can accurately identify and classify various forms of driver distraction, such as texting, talking on the phone, eating, or interacting with the car's entertainment system.
The project utilizes the State Farm Distracted Driver Detection dataset, which provides a large collection of images captured from inside vehicles. The dataset includes labeled images of drivers engaged in different activities, spanning ten classes of distractions. These classes range from safe driving (Class 0) to various forms of distractions (Class 1 to Class 9).
The project explores different deep learning models for distracted driver classification. It starts with a baseline dense model and progressively builds more complex models using convolutional neural networks (CNNs) and transfer learning techniques. These models are trained on the dataset and evaluated to determine their performance and effectiveness in classifying driver distractions.