Advanced Traffic Surveillance: YOLO-ARIMA Hybrid for Real-Time Monitoring

With the modern population boom in cities, traffic congestion has become an ever-present issue that worsens constantly. To tackle this critical issue, accurate predictions of traffic based on real-world trends are needed to anticipate traffic and take appropriate measures. To fulfil this, in this paper, we have used vehicle detection based on the YOLO model and Time-Series Forecasting based on the traffic detected over some time to identify the traffic trends and predict the future accordingly. By forecasting the traffic for the day on an intersection using traffic data extrapolated from the vehicle Detection model collected historically, we aim to provide high accurate Predictions of traffic congestion.