Accident Prediction Model with Image Recognition

This repository contains a comparative analysis of accident prediction models using image recognition techniques as well as a corresponding research paper. The dataset, sourced from Kaggle, includes images related to accident scenarios. We explore three different models: CNN, KNN, and Random Forest Trees, evaluating their performance based on accuracy, precision, recall, and F1 score.

Project Overview

Models and Performance

1. Convolutional Neural Network (CNN):

   • Accuracy: 93%

2. K-Nearest Neighbors (KNN):

   • Accuracy: 91%

3. Random Forest Trees:

   • Accuracy: 96%