Enhancing Railway Safety Through Human Activity Recognition

Intelligent Transportation System (ITS) is one of the evolving domains where Artificial Intelligence (AI) is creating endless opportunities. In recent days, for transit both passengers and freight prefer railway as the transportation medium. For this reason, the railway should provide the safety and uninterrupted services. These safety measures reduce the fatalities. In this paper, we present an Object Detection (OD) and Activity Recognition (AR) model for railway lines with the goal of reducing accidents, fatalities and improving safety. The proposed work uses the advantage of You Only Look Once Version 8 (YOLO-V8) to detect the objects and recognize the activities. This paper provides a detailed performance analysis of YOLO- V8 on training and validation sets. Also, this paper highlights the other performance indices such as classification accuracy and mean average precision with respect to the epoch number.