Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: Jan 1, 2026

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

Highway crash detection and risk estimation using deep learning.

Tingting Huang¹, Shuo Wang¹, Anuj Sharma¹

¹Department of Civil, Construction and Environmental Engineering, Iowa State University, United States.

Accident; Analysis and Prevention

|December 17, 2019

Summary

This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Improving crash data quality by identifying misclassified alcohol-involved crashes using NLP on narrative data.

Journal of safety research·2026

Same author

Gene expression and metabolic regulation of carotenoid and withanolide biosynthetic pathways in Withania somnifera with emphasis on FPPS-mediated regulation.

Protoplasma·2026

Same author

Optimizing hepatitis C diagnosis through reinforcement learning feature selection and multi-model machine learning evaluation.

Scientific reports·2026

Same author

A molecular dynamics simulation investigation on the material removal mechanism in nanoscale machining of aluminium.

Journal of molecular modeling·2026

Same author

Classifying Cognitive Decline in Older Drivers from Behavior on Adverse Roads Detected Using Computer Vision.

Journal of transportation technologies·2026

Same author

Effect of Hyperthermia on Prothrombin Time (PT), International Normalized Ratio (INR), and Activated Partial Thromboplastin Time (APTT).

Cureus·2026

Same journal

Differential sensitivity of self-reported driving and collision measures to aspects of shiftwork, sleep, and fatigue.

Accident; analysis and prevention·2026

Same journal

Delving into the visual attention of pedestrians during street crossing under time pressure: An eye-tracking approach.

Accident; analysis and prevention·2026

Same journal

Differentiating high-frequency and high-severity hotspots: A robust risk-evolution-volume (REV) framework.

Accident; analysis and prevention·2026

Same journal

Modeling takeover decisions in driving automation: a multilevel drift-diffusion model (MDDM) framework integrating human, system, and environmental factors.

Accident; analysis and prevention·2026

Same journal

The state-dependent causal effect of a V2X-based beyond-line-of-sight warning system on young driver response: a causal machine learning approach.

Accident; analysis and prevention·2026

Same journal

How conservative driving behavior increases crash risk: Understanding the systemic safety impacts of older drivers in mixed traffic flows.

Accident; analysis and prevention·2026

See all related articles

Deep learning models show improved crash detection and comparable crash risk prediction performance using real-world traffic data. Predicting crash risk accurately 10 minutes before an event remains challenging.

Area of Science:

Traffic Engineering
Data Science
Artificial Intelligence

Background:

Timely crash detection is crucial for traffic management and public safety.
Predicting crash risk helps prevent secondary incidents and improve highway safety.
Advancements in real-time data collection and machine learning offer new solutions for traffic safety.

Purpose of the Study:

To explore the feasibility of using deep learning models for crash detection and risk prediction.
To evaluate the performance of deep learning models against state-of-the-art shallow models.
To analyze the impact of data recency on crash risk prediction accuracy.

Main Methods:

Utilized real-world traffic data (Volume, Speed, Sensor Occupancy) from roadside radar sensors.

Keywords:

Crash detection Crash prediction Deep learning

Related Experiment Videos

Last Updated: Jan 1, 2026

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

Designed feature sets for deep learning models for crash detection and prediction.

Conducted a sensitivity analysis using data from 1, 5, and 10 minutes prior to crashes.

Main Results:

Deep learning models demonstrated superior performance in crash detection compared to shallow models.
Crash risk prediction performance of deep learning models was comparable to existing shallow models.
Predicting crash risk with high accuracy 10 minutes before an event proved difficult.

Conclusions:

Deep learning models are effective for real-time crash detection using traffic sensor data.
While promising, predicting crash risk remains a complex challenge, especially for longer prediction horizons.
Further research can optimize deep learning approaches for enhanced highway traffic safety systems.