Road Accident Severity Detection In Smart Cities

Abstract

Ensuring safety, in cities is a focus in the development of urban areas requiring new and creative methods for categorizing and managing accidents. Traditional approaches often face challenges in evaluating accident seriousness within changing city environments. This research utilizes Long Short Term Memory (LSTM) and Convolutional Neural Network (CNN) techniques to create a system that categorizes accidents into three severity levels; minor, moderate and severe. By leveraging learning capabilities, our method boosts the precision and efficiency of safety protocols in cities. The outcomes exhibit promising results in categorizing accident severity offering a tool for enhancing urban safety infrastructure. Through empowering cities to handle accidents, our model establishes a foundation for safety initiatives. In essence, this study contributes to enhancing safety standards in cities promoting resilience and sustainability, within settings.