Real – Time Accident Alerting System Using Convolutional Neural Network

The purpose of the research is to solve the problem by utilizing easily accessible devices and equipment in order to make a safety mechanism at a significantly reduced cost. The system offers one of its benefits, which includes a smart camera with wireless control, night vision, tracking, and human detection. This particular system utilizes night vision in the detection of minor and major traffic accidents on the highways. It then passes the information to the control room and transmits picture footage to the database for verification. The system is orientated to allow 360-degree sensing. It will be built with a time latency of at least 30 seconds in order to view the mishaps and establish that they were serious ones. This can be made to store the information onto the cloud without the problems of space mishaps of data loss. With the application of CNN's deep learning algorithm along with Python programming, it is possible to locate the accident location precisely and alert the emergency services immediately. This paper is focused on safety precautions taken once the accident has been detected. This means that there is a necessity to save precious human life by having this implemented in real life.