IncidentNet: Traffic Incident Detection, Localization and Severity Estimation with Sparse Sensing
- School of Computing and Augmented Intelligence
- Arizona State University
Abstract
Prior art in traffic incident detection relies on high sensor coverage and is primarily based on decision-tree and random forest models that have limited representation capacity and, as a result, cannot detect incidents with high accuracy. This paper presents IncidentNet - a novel approach for classifying, localizing, and estimating the severity of traffic incidents using deep learning models trained on data captured from sparsely placed sensors in urban environments. Our model works on microscopic traffic data that can be collected using cameras installed at traffic intersections. Due to the unavailability of datasets that provide microscopic traffic details and traffic incident details simultaneously, we also present a methodology to generate a synthetic microscopic traffic dataset that matches given macroscopic traffic data. IncidentNet achieves a traffic incident detection rate of 98%, with false alarm rates of less than 7% in 197 seconds on average in urban environments with cameras on less than 20% of the traffic intersections.
Important Figures
The plot of the vehicle counts for a 24-hour period from the Department of Transportation of Tempe for the 12 roads between the placed sensors of interest from the selected Tempe region.
Representation of averaged ground-truth vehicle counts and generated traffic flow model. To ensure variance in generated vehicle counts, we add a small deviation $alpha$.
Shows A Tempe, AZ region selected as the test area for our implementations. All the plotted points indicate the locations where cameras can be deployed for simulation. However, the deployed locations are highlighted in red to make the deployment of cameras similar to the real world.
The block diagram depicts IncidentNet’s architecture. The raw data from the simulator is transformed into processed data. For training, all data points are used for the incident detection model, and data points with positive incident labels are used for incident localization and severity estimation models. During the prediction phase, localization and severity estimation models depend on the incident detection model’s prediction.
The KS Statistic and the p-value obtained from the KS test for the four days of data made available by Tempe are shown. The p-value threshold is indicated as the red line.
Our approaches’ performance with consideration for different sparsity levels. Notice that the incident detection rate is still high for sparsity, as high as 93%. Tabnet performs better in incident detection with a low false alarm rate.
Paper Details
Paper Contributions
IncidentNet presents two main contributions, to address the challenge of detecting incidents between intersections:
- A repeatable approach for generating realistic fine-grain synthetic datasets using traffic flow data within a microscopic traffic simulator, facilitating researchers with more realistic data. Our method takes readily available coarse-grain public traffic flow data. It generates a synthetic dataset using traffic data within a simulator that closely matches the coarse-grain distributions of the public traffic flow real-world dataset.
- A novel technique that can detect and localize a traffic incident without the incident being directly in the field of view of a visual sensor. Localization of the incident is achievable without knowing the precise distance between sensors. This incident detection technique is also robust to sparse sensor placement in urban regions.
Bibtex
@article{peddiraju2024incidentnet,
title={IncidentNet: Traffic Incident Detection, Localization and Severity Estimation with Sparse Sensing},
author={Peddiraju, Sai Shashank and Harapanahalli, Kaustubh and Andert, Edward and Shrivastava, Aviral},
journal={arXiv preprint arXiv:2408.00996},
year={2024}
}