基于朴素贝叶斯算法的船舶异常行为监测

交通运输系统工程与信息 ›› 2017, Vol. 17 ›› Issue (6): 147-154.

基于朴素贝叶斯算法的船舶异常行为监测

魏照坤，谢新连*，潘伟，赵瑞嘉

大连海事大学综合运输研究所，辽宁大连116026

收稿日期:2017-06-20 修回日期:2017-08-30 出版日期:2017-12-25 发布日期:2017-12-25
作者简介:魏照坤(1987-)，男，山东青岛人，博士生.
基金资助:
中央高校基本科研业务费专项资金/Fundamental Research Funds for the Central Universities (3132016358)；国家重点研发计划/ National Program on Key Research Project(2017YFSF060139).

Ship Abnormal Behavior Detection Based on Naive Bayes

WEI Zhao-kun, XIE Xin-lian, PANWei, ZHAO Rui-jia

Integrated Transport Institute, Dalian Maritime University, Dalian 116026, Liaoning, China

Received:2017-06-20 Revised:2017-08-30 Online:2017-12-25 Published:2017-12-25

摘要/Abstract

摘要：

随着海事事故与海上违法行为的不断增多，智能的监控方法成为降低海事事故，打击海上违法行为的有效手段.同时，船舶自动识别系统(Automatic Identification System，AIS)的普及及船舶交通管理系统(Vessel Traffic Service，VTS)的扩建，又为智能监控提供了数据支持.鉴于此，利用船舶自动识别系统提供的数据，分析通航水域船舶信息的分布情况，根据其概率分布采用朴素贝叶斯算法，从船舶航速、航向及距航道边界距离 3 个方面，构建船舶异常行为检测模型.最后，以成山角通航水域为例，检验模型的有效性. 实验结果表明，构建的模型能够有效地完成异常行为监测，减少海事监管人员的工作强度，同时根据实验结果分析了成山角水域船舶航行的特点，并对成山角定线制提出合理化建议.

关键词: 交通工程, 船舶自动识别系统, 朴素贝叶斯算法, 异常行为监测

Abstract:

With maritime accidents and violating law increasing, intelligent surveillance as a kind of significant measure can be applied to decrease or avoid maritime accidents and violations. Meanwhile, Automatic Identification System (AIS) is widely used and Vessel Traffic Service (VTS) is being improved, which is beneficial to supply effective data for intelligent surveillance. On the bases of that, AIS data is used to analyze ship information on the sea area. According to probability distribution of ship information, Naive Bayes (NB) algorithm is applied to build abnormal detection model to monitor dynamic information such as speed, course and the distance between vessel position and limit of fairway. At last, data from Chengshan Jiao sea area is used to validate the validity of the algorithm. It is indicated from experimental results that the proposed method can effectively recognize ship abnormal behavior, which is conductive to decreasing workload. Moreover, the traffic characteristics in Chenshan Jiao sea area are analyzed and reasonable suggestions are proposed.

Key words: traffic engineering, automatic identification system (AIS), Naive Bayes (NB), abnormal behavior detection

中图分类号:

魏照坤，谢新连，潘伟，赵瑞嘉. 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 Zhao-kun, XIE Xin-lian, PANWei, ZHAO Rui-jia. Ship Abnormal Behavior Detection Based on Naive Bayes[J]. Journal of Transportation Systems Engineering and Information Technology, 2017, 17(6): 147-154.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: http://www.tseit.org.cn/CN/abstract/abstract19533.shtml

http://www.tseit.org.cn/CN/Y2017/V17/I6/147

参考文献

[1] UNCTAD. Review of maritime transport 2014[R]. American: United Nation Publication, 2014.

[2] MOU J M, VANDERTAK C, LIGTERINGEN H. Study on collision avoidance in busy Waterways by using AIS data[J]. Ocean Engineering, 2010(37): 483-490.

[3] HANSEN M G, JENSEN T K, LEHNSCHIOLER T, et al. Empirical ship domain based on AIS data[J]. Journal of Navigation, 2014, 66(6): 931-940.

[4] KRZYSZTOF M, Gucma L. Probabilistic model of minimal passing distances of vessels navigating in Polish coastal waters[C]. Psam 11 & Esrel, 2012.

[5] XIAO F, HAN L, GULIJK C V, et al. Comparison study on AIS data of ship traffic behavior[J]. Ocean Engineering, 2015, 95(3): 84-93.

[6] 魏照坤, 周康, 魏明, 等. 基于AIS数据的船舶运动模式识别与应用[J]. 上海海事大学学报, 2016, 37(2): 17-22. [WEI Z K, ZHOU K, WEI M, et al. Ship motion pattern recognition and application based on AIS data[J]. Journal of Shanghai Maritime University, 2016, 37(2): 17-22.]

[7] WANG Y, ZHANG J, CHEN X, et al. A spatialtemporal forensic analysis for inland- water ship collision using AIS data[J]. Safety Science, 2013(57): 187-202.

[8] LI S, MENG Q, QU X. An overview of maritime waterway quantitative risk assessment models[J]. Risk Analysis, 2012, 32(3): 496-512.

[9] GOERLANDT F, MONTEWKA J. Maritime transportation risk analysis: review and analysis in light of some foundational issues[J]. Reliability Engineering & System Safety, 2015(138): 115-134.

[10] MAZAHERI A, MONTEWKA J, KUJALA P. Modeling the risk of ship grounding-a literature review from a risk management perspective[J]. WMU Journal of Maritime Affairs, 2014, 13(2): 269-297.

[11] JIANG L, LI C, WANG S, et al. Deep feature weighting for naïve Bayes and its application to text classification[J]. Engineering Applications of Artificial Intelligence, 2016(52): 26-39.

[12] MARUCCI-WELLMAN H R, LEHTO M R, CORNS H L. A practical tool for public health surveillance: Semiautomated coding of short injury narratives from large administrative database using Naïve Bayes algorithms[J]. Accident Analysis and Prevention, 2015 (84): 165-176.

[13] ESCALANTE H J, MORALES E F, SUCAR L E. A naïve bayes baseline for early gesture recognition[J]. Pattern Recognition Letters, 2016(73): 91-99.

[14] ZHANG H, KANG Y Y, ZHU Y Y, et al. Novel naïve Bayes classification models for predicting the chemical ames mutagenicity[J]. Toxicology in Vitro, 2017(41): 56- 63.

[15] MUKHERJEE S, SHARMA N. Intrusion detection using naïve bayes classifier with feature reduction[J]. Procedia Technology, 2012(4): 119-128.

[16] FRANCESCO C, FRANCESCO A N P, VAHID B, et al. Abnormal vessel behavior detection in port areas based on dynamic bayesian networks[C]. 17th International Conference on Information Fusion, Indra: IEEE, 2014.

[17] SALMA Z, VALERIE R, NADIA M. Behavior analysis modulus for harbor security[C]. Harnessing the Power of the Ocean, United States: IEEE, 2012.

[18] SANG L Z, WALL A, MAO Z, et al. A novel method for restoring the trajectory of the inland waterway ship by using AIS data[J]. Ocean Engineering, 2015(110): 183- 194.

[19] 范明, 范宏建. 数据挖掘导论[M]. 北京: 人民邮电出版社, 2011. [FAN M, FAN H J. Introduction to data mining[M]. Beijing: The People's Posts and Telecommunications Press, 2011.]