| 0 | 0 | 48 |
| Downloads | Citas | Reads |
A key challenge in expressway anomaly handling is that existing methods usually identify local abnormalities from a single sensing signal, but struggle to jointly determine event type, cause, and risk. To address this issue, this paper proposes a multimodal large-model-based method for expressway anomaly event detection. This paper addresses the demand for intelligent expressway management by proposing a multimodal large-model-based anomaly detection method. It integrates heterogeneous data from multiple sources, including video surveillance, millimeter-wave radar, traffic flow, and me‑teorological information, through a “modality expansion-scenario fine-tuning-cross-modal attention Transformer fusion”approach. This process maps diverse features into the CLIP shared semantic space, enabling unified representation of heterogeneous data. Combining traffic knowledge graphs with prompt engineering guides the large model to deeply analyze the types, causes, and potential risks of abnormal events. Experimental results based on the DAIR-V2X dataset demonstrate that the proposed method significantly outperforms traditional approaches in both detection accuracy and comprehensive semantic understanding. This provides a new technical pathway and theoretical foundation for multidimensional perception on expressways.
[ 1] 2023 年 交 通 运 输 行 业 发 展 统 计 公 报[J]. 中 国 水 运 ,2024(8):31-35.
[ 2] 吴晓佩 . 基于多 SVM 分类器融合的高速公路异常事件检测方法[J]. 现代交通技术,2014,11(4):63-67.
[ 3] 隋 靓 ,党 建 武 . 基 于 运 动 目 标 轨 迹 的 高 速 公 路 异 常事 件 检 测 算 法 研 究[J]. 计 算 机 应 用 与 软 件 ,2018,35(1):246-252.
[ 4] 李旭,张秀杰 . 基于视频分析的高速公路异常事件检测[J]. 公路交通科技(应用技术版),2019(7):299-302.
[ 5] 孙 倩 ,郭 忠 印 . 基 于 离 散 时 间 信 号 相 关 性 的 交 通 事件 检 测 算 法[J]. 同 济 大 学 学 报(自 然 科 学 版),2018,46(11):1508-1513.
[ 6] WEIL R, WOOTTON J, GARCÍA-ORTIZ A. Traffic incident detection: sensors and algorithms[J]. Mathematical and Computer Modelling,1998,27(9-11):257-291.
[ 7] TENG H L, QI Y. Detection-delay-based freeway incident detection algorithms[J]. Transportation Research Part C: Emerging Technologies,2003,11(3-4):265-287.
[ 8] JIN J, RAN B. Automatic freeway incident detection based on fundamental diagrams of traffic flow[J]. Transportation Research Record: Journal of the Transportation Research Board,2009,2099(1):65-75.
[ 9] RITCHIE S G , CHEU R L. Simulation of freeway incident detection using artificial neural networks[J].Transportation Research Part C: Emerging Technologies,1993,1(3):203-217.
[10] DIA H, ROSE G. Development and evaluation of neural network freeway incident detection models using field data[J]. Transportation Research Part C: Emerging Technologies,1997,5(5):313-331.
[11] SRINIVASAN D, CHEU R L, POH Y P, et al. Development of an intelligent technique for traffic network incident detection[J]. Engineering Applications of Artificial Intelligence,2000,13(3):311-322.
[12] SHEHATA M S, CAI J, BADAWY W M, et al.Video-based automatic incident detection for smart roads: the outdoor environmental challenges regarding false alarms[J]. IEEE Transactions on Intelligent Trans‑
portation Systems,2008,9(2):349-360.
[13] BUCH N, VELASTIN S A, ORWELL J. A review of computer vision techniques for the analysis of urban traffic[J]. IEEE Transactions on Intelligent Transporta‑tion Systems,2011,12(3):920-939.
[14] CHEN S Y, WANG W. Decision tree learning for freeway automatic incident detection[J]. Expert Systems with Applications,2009,36(2):4101-4105.
[15] MA X L, DAI Z, HE Z B, et al. Learning traffic as images: a deep convolutional neural network for largescale transportation network speed prediction[J]. Sensors,2017,17(4):818.
[16] YAO Y, XU M Z, WANG Y C, et al. Unsupervised traffic accident detection in first-person videos[C]//Proceedings of IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE,2020:
273-280.
[17] XU Y J, HU H, HUANG C W, et al. TAD: a largescale benchmark for traffic accidents detection from video surveillance[J]. IEEE Access,2025,13:2018-2033.
Basic Information:
Citation Information:
[1]LIU Huaxue,ZHOU Jie, CUI Liang ,et al.Research on expressway anomaly event detection methods based on multimodal large models[J].Modern Transportation and Metallurgical Materials,2026(04):33-42.
Fund Information:
江苏省自然科学基金资助项目(BK20230964)
2026-04-27
2026
2026-07-15