A Vehicle Trajectory Anomaly Detection Method Based on Noise Label Re-Weighting

SU YUE-yang; YAO Di; BI Jing-ping

doi:10.12263/DZXB.20230568

您当前的位置：

首页 >

文章列表页 >

A Vehicle Trajectory Anomaly Detection Method Based on Noise Label Re-Weighting

PAPERS | 更新时间：2025-12-08

- A Vehicle Trajectory Anomaly Detection Method Based on Noise Label Re-Weighting
- ACTA ELECTRONICA SINICA Vol. 53, Issue 1, Pages: 182-192(2025)
- 作者机构：
  
  1.中国科学院大学，北京 100049
  2.中国科学院计算技术研究所，北京 100190
- 作者简介：
- 基金信息：
  
  National Natural Science Foundation of China(62002343)
- DOI：10.12263/DZXB.20230568
  CLC： TP391;
- Received：20 June 2023，
  
  Revised：2024-01-25，
  
  Published：25 January 2025
- 稿件说明：
移动端阅览
苏越阳, 姚迪, 毕经平. 基于噪声标签重加权的车辆轨迹异常检测方法[J]. 电子学报, 2025, 53(01): 182-192.

SU YUE-yang, YAO Di, BI Jing-ping. A Vehicle Trajectory Anomaly Detection Method Based on Noise Label Re-Weighting[J]. Acta Electronica Sinica, 2025, 53(01): 182-192.
苏越阳, 姚迪, 毕经平. 基于噪声标签重加权的车辆轨迹异常检测方法[J]. 电子学报, 2025, 53(01): 182-192. DOI：10.12263/DZXB.20230568

SU YUE-yang, YAO Di, BI Jing-ping. A Vehicle Trajectory Anomaly Detection Method Based on Noise Label Re-Weighting[J]. Acta Electronica Sinica, 2025, 53(01): 182-192. DOI：10.12263/DZXB.20230568

摘要

车辆轨迹异常检测为各种位置信息服务提供了重要的安全保障，基于机器学习的方法作为主流检测方法已经被广泛地应用于交通、军事等各个领域. 然而受限于噪声标签问题，现有的异常检测方法在实际应用中性能不佳.为解决这个问题，本文提出了一种基于噪声标签重加权的车辆轨迹异常检测方法（noise label ReWeighting-based vehicle Trajectory Anomaly Detection，RW-TAD）.该方法采用自监督的方式构建样本权重估计模块，通过计算轨迹的生成概率评估给定标签的可信度.然后使用基于加权损失的检测模型判定异常轨迹.在训练过程中，RW-TAD模型使用基于双层损失的协同优化机制联合学习样本权重估计模块和异常检测模块.实验结果表明该方法可以有效缓解噪声标签样本对模型训练的干扰，取得了较好的性能.相比于已有的方法，RW-TAD在检测准确率和性能稳定性上都有很大的提升.

Abstract

Vehicle trajectory anomaly detection provides important security support for various location-based services. Machine learning-based methods

as the mainstream detection methods

have been widely applied in various fields such as transportation and military. However

due to the problem of noise labels

existing anomaly detection methods have poor performance in practical applications. To solve this problem

this paper proposes a vehicle trajectory anomaly detection method based on noise label re-weighting (RW-TAD). This method uses a self-supervised approach to construct a sample weight estimator

which evaluates the credibility of given labels by calculating the probability of trajectory generation. Then

a detector based on weighted loss is used to detect anomalous trajectories. During the training process

the RW-TAD model uses a collaborative optimization strategy based on a dual-layer loss to jointly learn the sample weight estimator and the detector. Experimental results show that this method can effectively alleviate the interference of noisy samples on model training and achieve good performance. Compared with existing methods

it has greatly improved in detection accuracy and performance stability.

关键词

Keywords

references

WANG S , BAO Z F , CULPEPPER J S , et al . A survey on trajectory data management, analytics, and learning [J ] . ACM Computing Surveys , 2021 , 54 ( 2 ): 1 - 36 .

BELHADI A , DJENOURI Y , LIN J C , et al . Trajectory outlier detection [J ] . ACM Transactions on Management Information Systems , 2020 , 11 ( 3 ): 1 - 29 .

SUKHBAATA S , FERGUS R . Learning from noisy labels with deep neural networks [EB/OL ] . ( 2015-04-10 )[ 2023-06-01 ] . https://arxiv.org/abs/1406.2080v4 https://arxiv.org/abs/1406.2080v4 .

CORDEIRO F R , CARNEIRO G . A survey on deep learning with noisy labels: How to train your model when you cannot trust on the annotations? [C ] // 2020 33rd SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI) . Piscataway : IEEE , 2020 : 9 - 16 .

ZHANG D Q , LI N , ZHOU Z H , et al . iBAT: Detecting anomalous taxi trajectories from GPS traces [C ] // UbiComp’11: Proceedings of the 2011 ACM Conference on Ubiquitous Computing . New York : ACM , 2011 : 99 - 108 .

LIU W , ZHENG Y , CHAWLA S , et al . Discovering spatio-temporal causal interactions in traffic data streams [C ] // Proceedings of the 17th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining . New York : ACM , 2011 : 1010 - 1018 .

ZHU J , JIANG W , LIU A , et al . Effective and efficient trajectory outlier detection based on time-dependent popular route [J ] . World Wide Web , 2017 , 20 ( 1 ): 111 - 134 .

LV Z J , XU J J , ZHAO P P , et al . Outlier Trajectory Detection: A Trajectory Analytics Based Approach [M ] // Database Systems for Advanced Applications . Cham : Springer International Publishing , 2017 : 231 - 246 .

TSIRIKOGLOU A , EILERTSEN G , UNGER J . A survey of image synthesis methods for visual machine learning [J ] . Computer Graphics Forum , 2020 , 39 ( 6 ): 426 - 451 .

NICHOLS J A , HERBERT CHAN H W , BAKER M A B . Machine learning: Applications of artificial intelligence to imaging and diagnosis [J ] . Biophysical Reviews , 2019 , 11 ( 1 ): 111 - 118 .

OTTER D W , MEDINA J R , KALITA J K . A survey of the usages of deep learning for natural language processing [J ] . IEEE Transactions on Neural Networks and Learning Systems , 2021 , 32 ( 2 ): 604 - 624 .

TORFI A , SHIRVANI R A , KENESHLOO Y , et al . Natural language processing advancements by deep learning: A survey [EB/OL ] . ( 2021-02-27 )[ 2023-06-01 ] . http://arxiv.org/abs/2003.01200v4 http://arxiv.org/abs/2003.01200v4 .

SONG L , WANG R J , XIAO D , et al . Anomalous trajectory detection using recurrent neural network [M ] // Advanced Data Mining and Applications . Cham : Springer International Publishing , 2018 : 263 - 277 .

FANG W , CHEN Y P , XUE Q Y . Survey on research of RNN-based spatio-temporal sequence prediction algorithms [J ] . Journal on Big Data , 2021 , 3 ( 3 ): 97 - 110 .

LIU Y D , ZHAO K Q , CONG G , et al . Online anomalous trajectory detection with deep generative sequence modeling [C ] // 2020 IEEE 36th International Conference on Data Engineering (ICDE) . Piscataway : IEEE , 2020 : 949 - 960 .

HAN X L , CHENG R , MA C H , et al . DeepTEA [J ] . Proceedings of the VLDB Endowment , 2022 , 15 ( 7 ): 1493 - 1505 .

SAHU M , DASH R . A survey on deep learning: Convolution neural network (CNN) [C ] // Intelligent and Cloud Co- mputing . Singapore : Springer , 2021 : 317 - 325 .

WU H , SUN W W , ZHENG B H , et al . A fast trajectory outlier detection approach via driving behavior modeling [C ] // Proceedings of the 2017 ACM on Conference on Information and Knowledge Management . New York : ACM , 2017 : 837 - 846 .

HAN B , YAO Q M , LIU T L , et al . A survey of label-noise representation learning: Past, present and future [EB/OL ] . ( 2021-02-20 )[ 2023-06-01 ] . http://arxiv.org/abs/2011.04406v2 http://arxiv.org/abs/2011.04406v2 .

SONG H , KIM M , PARK D , et al . Learning from noisy labels with deep neural networks: A survey [J ] . IEEE Transactions on Neural Networks and Learning Systems , 2023 , 34 ( 11 ): 8135 - 8153 .

DELANY S J , SEGATA N , MAC NAMEE B . Profiling instances in noise reduction [J ] . Knowledge-Based Systems , 2012 , 31 : 28 - 40 .

GAMBERGER D , LAVRAC N , DZEROSKI S . Noise detection and elimination in data preprocessing: Experiments in medical domains [J ] . Applied Artificial Intelligence , 2000 , 14 ( 2 ): 205 - 223 .

THONGKAM J , XU G D , ZHANG Y C , et al . Support vector machine for outlier detection in breast cancer survivability prediction [M ] //Advanced Web and Network Technologies, and Applications. Berlin : Springer , 2008 : 99 - 109 .

CERVANTES J , GARCIA-LAMONT F , RODRÍGUEZ-MAZAHUA L , et al . A comprehensive survey on support vector machine classification: Applications, challenges and trends [J ] . Neurocomputing , 2020 , 408 : 189 - 215 .

JIANG L , ZHOU Z Y , LEUNG T , et al . MentorNet: Learning data-driven curriculum for very deep neural networks on corrupted labels [EB/OL ] .( 2018-08-13 )[ 2023-06-01 ] . http://arxiv.org/abs/1712.05055v2 http://arxiv.org/abs/1712.05055v2 .

YU X R , HAN B , YAO J C , et al . How does disagreement help generalization against label corruption? [EB/OL ] . ( 2019-05-12 )[ 2023-06-01 ] . http://arxiv.org/abs/1901.04215v3 http://arxiv.org/abs/1901.04215v3 .

CHENG L L , ZHOU X Z , ZHAO L M , et al . Weakly supervised learning with side information for noisy labeled images [M ] // Computer Vision - ECCV 2020 . Cham : Springer International Publishing , 2020 : 306 - 321 .

XIA X B , HAN B , WANG N N , et al . Extended $T$T: Learning with mixed closed-set and open-set noisy labels [J ] . IEEE Transactions on Pattern Analysis and Machine Intelligence , 2023 , 45 ( 3 ): 3047 - 3058 .

HAN B , YAO J C , NIU G , et al . Masking: A new perspective of noisy supervision [EB/OL ] . ( 2018-11-31 )[ 2023-06-01 ] . http://arxiv.org/abs/1805.08193v2 http://arxiv.org/abs/1805.08193v2 .

YAO J C , WANG J J , TSANG I W , et al . Deep learning from noisy image labels with quality embedding [J ] . IEEE Transactions on Image Processing , 2019 , 28 ( 4 ): 1909 - 1922 .

ZHENG G Q , AWADALLAH A H , DUMAIS S . Meta label correction for noisy label learning [J ] . Proceedings of the AAAI Conference on Artificial Intelligence , 2021 , 35 ( 12 ): 11053 - 11061 .

DEHGHANI M , SEVERYN A , ROTHE S , et al . Learning to learn from weak supervision by full supervision [EB/OL ] . ( 2017-11-30 )[ 2023-06-01 ] . http://arxiv.org/abs/1711.11383 http://arxiv.org/abs/1711.11383 .

SHU J , XIE Q , YI L X , et al . Meta-weight-net: Learning an explicit mapping for sample weighting [EB/OL ] . ( 2019-09-27 )[ 2023-06-01 ] . http://arxiv.org/abs/1902.07379v6 http://arxiv.org/abs/1902.07379v6 .

DONG W , YAN D W , WANG P . Self-supervised node representation learning via node-to-neighbourhood alignment [J ] . IEEE Transactions on Pattern Analysis and Machine Intelligence , 2024 , 46 ( 6 ): 4218 - 4233 .

VASWANI A , SHAZEER N M , PARMAR N , et al . Attention is all you need [EB/OL ] .( 2017-06-12 )[ 2023-06-01 ] . http://arxiv.org/abs/1706.03762 http://arxiv.org/abs/1706.03762 .

MEDAK J , GOGOI P P . Review and analysis of single-source shortest path problem using fijkstra’s algorithm [J ] . IOSR Journal of Computer Engineering (IOSR-JCE) , 2018 , 20 ( 2 ): 10 - 15 .

MALHOTRA P , RAMAKRISHNAN A , ANAND G , et al . LSTM-based encoder-decoder for multi-sensor anomaly detection [EB/OL ] . ( 2016-07-11 )[ 2023-06-01 ] . http://arxiv.org/abs/1607.00148v2 http://arxiv.org/abs/1607.00148v2 .

YAO D , ZHANG C , ZHU Z H , et al . Trajectory clustering via deep representation learning [C ] // 2017 International Joint Conference on Neural Networks (IJCNN) . Piscataway : IEEE , 2017 : 3880 - 3887 .

NICHOLSON B , ZHANG J , SHENG V S , et al . Label noise correction methods [C ] // 2015 IEEE International Conference on Data Science and Advanced Analytics (DSAA) . Piscataway : IEEE , 2015 : 1 - 9 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Variable Horizon Multi-Directional Scanning Method for Time Series Anomaly Detection

Unsupervised Image Anomaly Detection Based on Constrained BidiRectional Distillation

A Survey of Network Attack Investigation Based on Provenance Graph

Improved Adaptive Model Pools for Online Anomaly Detection Algorithms

Broadband Wide-Beamwidth Phased Array System

Related Author

WEI Song-jie

HUANG Yu-zhe

GUAN Yong-yuan

TANG Jin-hui

XING Peng

LI Ze-chao

LI Bo

TIAN Zhi-hong

Related Institution

School of Computer Science and Engineering, School of Cyber Science and Engineering, Nanjing University of Science and Technology

School of Computer Science and Engineering, Nanjing University of Science and Technology

NSFOCUS Technologies Group Co., Ltd

Department of Computer Science and Engineering,Shanghai Jiaotong University

Peng Cheng Laboratory

⁰