Smart Wireless Pedestrian Detection

Abstract

The World Health Organization aims to cut traffic-related accidents by 50% as part of its 2030 Agenda for Sustainable Development. It is possible to observe the movement of individuals across the pedestrian crossing by implementing the suggested localization method. Consequently, the smart city may establish a traffic flow that is both safe and effective. Ensuring that pedestrians have safe places to cross busy streets, like crosswalks and walkways that cross over occupied roads, along with various tactics, is very important. The world's major thoroughfares and pedestrian walkways are now busy places due to the constant advancement of technology and the need for more people to support it. As a result, using any type of roadway crossing place may make walking less safe and efficient. These days, one of the main features of street crossing zones is the button on each traffic light that doubles as a human detector, alerting the system to the presence of a pedestrian and prompting the system to request a "WALKING" signal as quickly as possible. Therefore, the goal of this study is to create a wireless, portable smart management system for pedestrian crossing areas that will automatically regulate traffic and enable pedestrians to cross the street securely and easily. To detect pedestrians, the system uses smart sensing, which automatically regulates the traffic lights at crosswalks. The sample images of pedestrians are taken from the Roboflow website. The detection of humans using Haar Cascade algorithm technique was done in Raspbian OS by using Python coding. From that, it will trigger the LED pedestrians traffic light to turn ON green light. Real-time object detection results yield a 78% accuracy rate for correct detections and a 22% rate for incorrect detections of 100 test images, with an average processing time of 0.3 milliseconds.