Multi-Modal Traffic Analysis: Integrating Time-Series Forecasting, Accident Prediction, and Image Classification

TL;DR

The paper tackles multi-modal traffic analysis by integrating time-series forecasting, accident severity prediction, and image classification into a single modular framework. It combines ARIMA forecasting with $p=2$, $d=0$, $q=1$ for traffic volumes, XGBoost for severity prediction, and TrafficNet CNN for traffic-image classification, enabling real-time, holistic insights for smart cities. Key findings include $MAE=2.1$ for ARIMA forecasts, $Accuracy=1.0$ for accident severity on balanced data, and $Accuracy=0.92$ for image classification, with end-to-end latency under $30$ ms. This integrated approach provides interpretable, scalable, and deployment-ready capabilities for proactive traffic management and safety improvements in urban environments.

Abstract

This study proposes an integrated machine learning framework for advanced traffic analysis, combining time-series forecasting, classification, and computer vision techniques. The system utilizes an ARIMA(2,0,1) model for traffic prediction (MAE: 2.1), an XGBoost classifier for accident severity classification (100% accuracy on balanced data), and a Convolutional Neural Network (CNN) for traffic image classification (92% accuracy). Tested on diverse datasets, the framework outperforms baseline models and identifies key factors influencing accident severity, including weather and road infrastructure. Its modular design supports deployment in smart city systems for real-time monitoring, accident prevention, and resource optimization, contributing to the evolution of intelligent transportation systems.

Figures (11)

• Figure 1: Class distribution in accident severity dataset.
• Figure 2: Word Cloud
• Figure 3: ARIMA forecasting results
• Figure 4: Decomposition of traffic volume time series.
• Figure 5: Accident severity by weather conditions.