Real-time road accident reporting system with location detection using cloud-based data analytics

With the increase of road vehicles, road accidents are an increasingly severe threat with incredible economic and social impacts. There has been a rising trend of fatalities and non-fatal casualties caused by road accidents in recent years. One of the main factors contributing to this is the slow dispatch of authorities during an accident as there is currently a lack of an automated accident reporting system. Thus, in this paper, the design of a real-time road accident reporting system was developed and presented. Computer vision was used in accident detection through the employment of a trained model via deep learning. The system then uses location detection and Global System for Mobile Communications (GSM) to accurately report accidents to the relevant authorities. Cloud-based data analytics is also used to store collected data in a designated cloud location for future reference. A working prototype was developed with on-the-road testing demonstrating successful hazard detection and distance measurement as well as location detection. The reporting system and cloud analytics system was able to show instantaneous dispatch and storage of data to the respective targets. Overall, a working prototype was successfully designed, developed, and tested with both simulated inputs and live on-the-road footage.