Lane-Wise Highway Anomaly Detection

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Qiu, Mei, Reindl, William Lorenz, Chen, Yaobin, Chien, Stanley, Hu, Shu
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866912361425666048
author Qiu, Mei
Reindl, William Lorenz
Chen, Yaobin
Chien, Stanley
Hu, Shu
author_facet Qiu, Mei
Reindl, William Lorenz
Chen, Yaobin
Chien, Stanley
Hu, Shu
contents This paper proposes a scalable and interpretable framework for lane-wise highway traffic anomaly detection, leveraging multi-modal time series data extracted from surveillance cameras. Unlike traditional sensor-dependent methods, our approach uses AI-powered vision models to extract lane-specific features, including vehicle count, occupancy, and truck percentage, without relying on costly hardware or complex road modeling. We introduce a novel dataset containing 73,139 lane-wise samples, annotated with four classes of expert-validated anomalies: three traffic-related anomalies (lane blockage and recovery, foreign object intrusion, and sustained congestion) and one sensor-related anomaly (camera angle shift). Our multi-branch detection system integrates deep learning, rule-based logic, and machine learning to improve robustness and precision. Extensive experiments demonstrate that our framework outperforms state-of-the-art methods in precision, recall, and F1-score, providing a cost-effective and scalable solution for real-world intelligent transportation systems.
format Preprint
id arxiv_https___arxiv_org_abs_2505_02613
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Lane-Wise Highway Anomaly Detection
Qiu, Mei
Reindl, William Lorenz
Chen, Yaobin
Chien, Stanley
Hu, Shu
Image and Video Processing
Machine Learning
This paper proposes a scalable and interpretable framework for lane-wise highway traffic anomaly detection, leveraging multi-modal time series data extracted from surveillance cameras. Unlike traditional sensor-dependent methods, our approach uses AI-powered vision models to extract lane-specific features, including vehicle count, occupancy, and truck percentage, without relying on costly hardware or complex road modeling. We introduce a novel dataset containing 73,139 lane-wise samples, annotated with four classes of expert-validated anomalies: three traffic-related anomalies (lane blockage and recovery, foreign object intrusion, and sustained congestion) and one sensor-related anomaly (camera angle shift). Our multi-branch detection system integrates deep learning, rule-based logic, and machine learning to improve robustness and precision. Extensive experiments demonstrate that our framework outperforms state-of-the-art methods in precision, recall, and F1-score, providing a cost-effective and scalable solution for real-world intelligent transportation systems.
title Lane-Wise Highway Anomaly Detection
topic Image and Video Processing
Machine Learning
url https://arxiv.org/abs/2505.02613