CNN-based Cognitive Distraction Surveillance for Pedestrians using Phones and Headphones

The use of cell phones or headphones by pedestrians is a major concern for traffic safety. It exponentially raises the likelihood of an accident occurring. Since the frequency of pedestrian fatalities caused by mobile phone or headphone usage is on the rise, authorities restrict pedestrians from using cell phones or headphones. However, in order to execute the regulation, law enforcement officials are currently stationed at the roadside to visually assess pedestrians. This method is time-consuming and ultimately ineffective. The proposed CNN-based architecture is designed to detect pedestrians who are using phones or headphones while crossing the street. The ResNet-50 model has been fine-tuned for the task using transfer learning. This detection algorithm identifies offenders and highlights them using OpenCV. The proposed algorithm identifies people, who are breaking the rules by using phones or headphones while crossing the street, with the help of an efficient and cost-effective methodology.