海港航道水域船舶异常行为检测

李高才, 张新宇, 蒋晨星, 连晓荣, 张辉辉, 王佳伟

doi:10.16411/j.cnki.issn1006-7736.2024.04.004

大连海事大学学报 >

2024 , Vol. 50 >Issue 4: 31 - 40

DOI: https://doi.org/10.16411/j.cnki.issn1006-7736.2024.04.004

海港航道水域船舶异常行为检测

展开

(1.集美大学航海学院，福建厦门 361021; 2.大连海事大学海上智能交通研究团队，辽宁大连 116026；3. 中国船级社质量认证有限公司深圳分公司，广东深圳 518052)

李高才（1992 — ），男，博士生，研究方向：海事数据挖掘与分析。张新宇*（1978 — ），男，博士，教授，博士生导师，E-mail：zhangxy@dlmu.edu.cn。蒋晨星（1999 — ），男，硕士生，研究方向：海事数据挖掘与分析。连晓荣（1999 — ），女，硕士生，研究方向：海事数据挖掘与分析。张辉辉（1977 — ），女，本科生，研究方向：海事数据挖掘与分析。王佳伟（2000 — ），男，硕士生，研究方向：海事数据挖掘与分析。

网络出版日期: 2024-12-24

基金资助

国家自然科学基金面上项目（52371359）；大连市科技创新基金项目（2022JJ12GX015）；集美大学科研启动金项目(ZQ2024110)

收起

Abnormal behavior detection of ships in harbor waterways

Expand

(1. Navigation College, Jimei University, Xiamen 361021, China; 2. Maritime Intelligent Transportation Research Team, Dalian Maritime University, Dalian 116026, China；3. Shenzhen Branch of CCSC, Shenzhen 518052，China)

Online published: 2024-12-24

Fold

摘要

鉴于目前VTS、AIS人工分析的效率难以满足日益严峻的水上交通监管形势要求，本文考虑船舶进出港航行特点，提出一种海港航道水域船舶异常行为检测方法。首先，考虑船舶类型、通航规则等影响因素，建立基于语义轨迹多维相似度的船舶轨迹聚类方法，以识别符合通航规则的船舶进出港交通模式；然后，构建语义转换模型，将交通模式轨迹数据转换为模式轨迹文本，并采用文本余弦相似度方法匹配目标船舶的交通模式；接着，利用核密度估计构建船舶异常行为检测模型。以天津港为例，从历史船舶轨迹数据中提取40种进出港交通模式，并以此构建船舶异常行为检测方法，通过航海模拟器仿真数据对方法进行验证。结果表明，该方法能够有效检测出船舶的异常行为，可对水上交通监管起到一定的辅助作用。

关键词： 船舶; 海港航道水域; 交通模式; 异常行为检测; AIS数据; 文本相似度; 核密度估计

本文引用格式

李高才, 张新宇, 蒋晨星, 连晓荣, 张辉辉, 王佳伟 . 海港航道水域船舶异常行为检测[J]. 大连海事大学学报, 2024 , 50(4) : 31 -40 . DOI: 10.16411/j.cnki.issn1006-7736.2024.04.004

Abstract

In view of current efficiency of VTS and AIS manual analysis cannot meet the increasingly severe situation of water traffic supervision, a method for detecting abnormal behaviors of ships in the waters of harbor waterways was proposed by considering the navigation characteristics of ships entering and leaving ports. Firstly, considering influence factors such as ship type and navigation rules, a ship trajectory clustering method based on semantic trajectory multi-dimensional similarity was established to identify traffic patterns of ships entering and leaving the port that comply with navigation rules. Secondly, a semantic transformation model was constructed to convert traffic pattern trajectory data into pattern trajectory text, while the text cosine similarity method was used to match the traffic pattern of the target ship. Furthermore, a ship anomaly behavior detection model was constructed by using kernel density estimation. Taking Tianjin Port as an example, 40 kinds of inbound and outbound traffic patterns were extracted from historical ship trajectory data to construct the ship abnormal behavior detection method, and validated by using simulated data from navigation simulator. Results show that the proposed method can effectively detect ship abnormal behaviors, providing assistance in waterway supervision.

Key words： ship; harbor waterway waters; traffic patterns; abnormal behavior detection; automatic identification system (AIS) data; text similarity; kernel density estimation

参考文献

［1］丁志国, 张新宇, 王程博, 等. 基于驾驶实践的无人船智能避碰决策方法［J］. 中国舰船研究, 2021, 16(1): 96-104.
DING Z G, ZHANG X Y, WANG C B, et al. Intelligent collision avoidance decision-making method for unmanned ships based on driving practice［J］. Chinese Journal of Ship Research,2021, 16(1): 96-104. (in Chinese)
［2］唐皇. 进出港船舶行为特征智能分析研究［D］. 大连：大连海事大学, 2020.
TANG H. Research on intelligent analysis of ship’s behavior characteristics of entering and leaving port［D］.Dalian: Dalian Maritime University, 2020. (in Chinese)
［3］马文耀. 船舶异常行为的一致性检测［D］. 大连：大连海事大学, 2018.
MA W Y. Conformal detection of anomalous behaviors of vessel［D］. Dalian: Dalian Maritime University, 2018. (in Chinese)
［4］RONG H, TEIXEIRA A P, SOARES C G. Data mining approach to shipping route characterization and anomaly detection based on AIS data［J］. Ocean Engineering, 2020, 198: 106936.
［5］魏照坤, 谢新连, 潘伟，等. 基于朴素贝叶斯算法的船舶异常行为监测［J］.交通运输系统工程与信息, 2017,17(6): 147-154.
WEI Z K, XIE X L, PAN W, et al. Ship abnormal behavior detection based on Nave Bayes［J］. Journal of Transportation Systems Engineering and Information Technology, 2017, 17(6): 147-154. (in Chinese)
［6］熊勇, 瞿小豪, 郭俊维，等. 基于AIS数据的渡轮异常运动模式监测［J］. 中国安全科学学报, 2016, 26(1): 100-103.
XIONG Y, QU X H, GUO J W, et al. Monitoring of abnormal movement patterns of ferry based on AIS data［J］. China Safety Science Journal, 2016, 26(1): 100-103. (in Chinese)
［7］黄亮, 刘益, 文元桥, 等. 内河渡船异常行为识别［J］.大连海事大学学报, 2017, 43(1): 8-13.
HUANG L, LIU Y, WEN Y Q, et al. Abnormal behavior recognition of inland river ferryboat［J］. Journal of Dalian Maritime University,2017,43(1):8-13. (in Chinese)
［8］俞庆英, 李倩, 陈传明, 等. 基于BP神经网络的异常轨迹检测方法［J］. 计算机工程, 2019, 45(7): 229-236.
YU Q Y, LI Q, CHEN C M, et al. Abnormal trajectory detection method based on BP neural network［J］. Computer Engineering,2019,45(7):229-236. (in Chinese)
［9］HU J, KAUR K, LIN H, et al. Intelligent anomaly detection of trajectories for IoT empowered maritime transportation systems［J］. IEEE Transactions on Intelligent Transportation Systems, 2022, 24(2): 2382-2391.
［10］NGUYEN D, VADAINE R, HAJDUCH G, et al. GeoTrackNet—a maritime anomaly detector using probabilistic neural network representation of AIS tracks and acontrario detection［J］. IEEE Transactions on Intelligent Transportation Systems, 2021, 23(6): 5655-5667.
［11］ZHEN R, JIN Y X, HU Q Y, et al. Maritime anomaly detection within coastal waters based on vessel trajectory clustering and Nave Bayes Classifier［J］. The Journal of Navigation, 2017, 70(3): 648-670.
［12］唐春华. 基于轨迹聚类和KDE的船舶轨迹异常实时检测［D］. 上海:上海海事大学, 2022.
TANG C H. Real-time detection of abnormal ship trajectory based on trajectory clustering and KDE［D］. Shanghai: Shanghai Maritime University, 2022. (in Chinese)
［13］LIU B, DE SOUZA E N, MATWIN S, et al. Knowledge-based clustering of ship trajectories using density-based approach［C］//2014 IEEE International Conferenceon Big Data. Washington :IEEE, 2014: 603-608.
［14］甄荣, 邵哲平, 潘家财. 基于AIS数据的船舶行为特征挖掘与预测:研究进展与展望［J］.地球信息科学学报, 2021, 23(12): 2111-2127.
ZHEN R, SHAO Z P, PAN J C. Advance in character mining and prediction of ship behavior based on AIS data［J］.Journal of Geo-information Science, 2021, 23(12): 2111-2127.

［15］FURTADO A S, KOPANAKI D, ALVARES L O, et al. Multidimensional similarity measuring for semantic trajectories［J］. Transactions in GIS, 2016, 20(2): 280-298.

［16］LIANG M, LIU R W, LI S, et al. An unsupervised learning method with convolutional auto-encoder for vessel trajectory similarity computation［J］. Ocean Engineering, 2021, 225: 108803.
［17］张博文, 刘智, 桑国明. 基于核密度波动的异常检测算法［J］. 计算机工程与应用, 2021, 57(12): 132-136.
ZHANG B W, LIU Z, SANG G M. Anomaly detection algorithm based on kernel density fluctuation［J］. Computer Engineering and Applications, 2021, 57(12): 132-136. (in Chinese)
［18］SILVERMAN B W. Density estimation for statistics and data analysis［M］. New York: Chapman and Hall, 1986.
［19］马洋, 赵旭俊, 苏建花, 等. 基于核密度估计的离群数据挖掘［J］. 太原科技大学学报, 2020, 41(6): 456-462.
MA Y, ZHAO X J, SU J H,et al. Outlier mining method based on kernel density estimation［J］. Journal of Taiyuan University of Science and Technology, 2020, 41(6): 456-462. (in Chinese)
［20］CHEN X, LIU Y, ACHUTHAN K, et al. A ship movement classification based on automatic identification system (AIS) data using convolutional neural network［J］. Ocean Engineering, 2020, 218: 108182.
［21］赵梁滨. 基于AIS数据和循环神经网络的船舶轨迹异常检测［D］. 大连：大连海事大学, 2019.
ZHAO L B. Ship trajectory outlier detection based on AIS data and recurrent neural network［D］. Dalian:Dalian Maritime University, 2019. (in Chinese)

Options

文章导航

模态框（Modal）标题

摘要

本文引用格式

Abstract

参考文献