计算机科学 ›› 2020, Vol. 47 ›› Issue (9): 175-184.doi: 10.11896/jsjkx.191000162

• 计算机图形学&多媒体 • 上一篇    下一篇

基于1DCNN-LSTM的船舶轨迹分类方法

崔彤彤, 王桂玲, 高晶   

  1. 大规模流数据集成与分析技术北京市重点实验室(北方工业大学) 北京100144
    北方工业大学信息学院 北京100144
  • 收稿日期:2019-10-24 发布日期:2020-09-10
  • 通讯作者: 王桂玲(wangguiling@ict.ac.cn)
  • 作者简介:ctt930814@163.com
  • 基金资助:
    北京市自然科学基金(4172018);国家自然科学基金(61832004,61672042);中电科海洋信息技术研究院有限公司高校合作课题(402054841879);北方工业大学毓优团队培养计划项目(107051360018XN012/020)

Ship Trajectory Classification Method Based on 1DCNN-LSTM

CUI Tong-tong, WANG Gui-ling, GAO Jing   

  1. Beijing Key Laboratory on Integration and Analysis of Large-scale Stream Data (North China University of Technology),Beijing 100144,China
    School of Information,North China University of Technology,Beijing 100144,China
  • Received:2019-10-24 Published:2020-09-10
  • About author:CUI Tong-tong,born in 1993,postgra-duate.Her main research interests include data processing technology and software services.
    WANG Gui-ling,born in 1978,Ph.D.associate professor,is a senior member of China Computer Federation.Her main research interests include large-scale streaming data integration and analysis,and service computing.
  • Supported by:
    This work was supported by Beijing Natural Science Foundation (4172018), National Natural Science Foundation of China (61832004,61672042),University of China Electric Ocean Information Technology Research Institute Co., Ltd (402054841879) and YuYou Team Training Project of North China University of Technology (107051360018XN012/020).

摘要: 由于监控设备视野有限、代价昂贵等问题,导致基于船舶图像或视频的船舶分类效果欠佳,改进船舶分类方法、提高船舶分类的准确率迫在眉睫。近几年,随着各类轨迹数据采集系统的兴起,通过船舶航行轨迹数据实现船舶类型的分类逐渐成为可能。针对使用传统二维卷积神经网络(Convolutional Neural Network,CNN)对船舶轨迹分类存在特征压缩和时序特征表达能力匮乏的问题,文中提出了一种一维CNN(One-Dimensional CNN,1DCNN)与长短期记忆网络(Long Short-Term Memory,LSTM)相结合的混合分类模型,对采集到的船舶自动识别系统(Automatic Identification System,AIS)数据进行船舶类型识别。首先,针对AIS采集到的船舶轨迹数据进行预处理,过滤噪声数据;然后,针对隐含在船舶轨迹信息中的特征对于1DCNN而言过于隐晦的问题,提出了一种针对大规模航舶轨迹数据的,且1DCNN能够识别的轨迹分布特征向量的构建算法,同时在此基础上提取了LSTM能够识别的时序特征向量;最后,将训练后的1DCNN模型与LSTM模型进行数据融合得到混合船舶分类模型。以渤海区域2016年6月的船舶AIS数据为基础,使用1DCNN与LSTM相结合的混合模型对渔船、客船、油船、集装箱船和散货船5类典型船舶的轨迹数据进行分类,并将其与使用一种神经网络如LSTM作为分类器的方法进行对比,结果表明所提方法具有明显的有效性,是一种有效的船舶轨迹分类方法。

关键词: 长短期记忆网络, 船舶轨迹分类, 分布特征向量, 时序特征向量, 一维卷积神经网络

Abstract: Due to the limited vision and cost of the monitoring equipment,the classification methods of ships based on images or videos are not very effective.So it is urgent to improve classification methods of the ships and the accuracy of those methods.In recent years,with the widelyused of various trajectory data acquisition systems,it has become possible to classify ship types through ship trajectory data.Based on the problem that the traditional two-dimensional convolutional neural network is lacking the ability of feature compression and temporal feature expression in ship trajectory recognition,this paper proposes a hybrid model which combines one-dimensional convolutional neural network (IDCNN) with long short-term memory (LSTM).This model can identify ship types by using the data collected from the automatic identification system (AIS).Firstly,this paper preprocesses the ship trajectory data collected by AIS to filter the noise data.Secondly,to solve the problem that the features hidden in the original ship trajectory information are over obscurity for 1DCNN,this paper proposes an algorithm for constructing the trajectory distribution feature vectors which can be accepted by 1DCNN for a large number of ship trajectory data.On this basis,the algorithm extracts the time series feature vectors which can be accepted by LSTM.Finally,this paper combines the trained 1DCNN model and LSTM model to get a hybrid ship classification model.Based on the AIS data of Bohai area on June 2016, the hybrid model combining 1DCNN with LSTM is used to classify five different typical ships including fishing ships,passenger ships,tanker ships,container ships and bulk-cargo ships.The experimental results show that compared with the method of using a neural network such as LSTM as classifier,the proposed method is obviously effective,and is an effective ship trajectory classification method.

Key words: Distribution feature vector, Long short-term memory(LSTM), One-dimensional convolutional neural network(1DCNN), Ship trajectory classification, Time series feature vector

中图分类号: 

  • TP183
[1] CHEN J H,LU F,PENG G J.Research progress on trajectory analysis of marine transportation vessels [J].China Navigation,2012,35(3):53-57.
[2] XU Y P.Research on the Trajectory Analysis of Marine Transportation Ships [J].China Water Transport (2nd Half),2017,17(2):16-17.
[3] XIAO W,SHAO Z P,PAN J C,et al.Ship trajectory clustering model based on AIS information and its application [J].China Navigation,2015,38(2):82-86.
[4] CAO W Q,LI Z X,WEI Q,et al.Trajectory Classification Me-thod Based on Probability Density Estimation of Regional Distribution [J].Computer Engineering,2018,44(4):262-267,286.
[5] LIU L,CHU X M,JIANG Z L,et al.Ship trajectory classification algorithm based on KNN [J].Journal of Dalian Maritime University,2018,44(3):15-21.
[6] ZHAO X Q,ZHANG L.SVM-based high-dimensional unba-.lanced data set classification algorithm[J].Journal of Nanjing University(Natural Science),2018,54(2):452-461.
[7] HOCHREITER S,SCHMIDHUBER J.Long short-term memory[J].Neural Computation,1997,9(8):1735-1780.
[8] ENDO Y,TODA H,NISHIDAK,et al.Deep feature extraction from trajectories for transportation mode estimation[C]//Paci-fic-Asia Conference on Knowledge Discovery and Data Mining.Springer,Cham,2016:54-66.
[9] LJUNGGREN H.Using Deep Learning for Classifying ShipTrajectories[C]//2018 21st International Conference on Information Fusion (FUSION).Cambridge,2018:2158-2164.
[10] ZHANG G H,LIU B.Research on time series classificationusing CNN and bidirectional GRU [J].Journal of Frontiers of Computer Science and Technology,2019,13(6):916-927.
[11] ZHENG Z T,ZHAO Z F,WANG G L,et al.Method for ship trajectory extraction for port stop area identification [J].Journal of Computer Applications,2019,39(1):113-117.
[12] ZHENG Y,LIU L,WANG L,et al.Learning transportationmode from raw gps data for geographic applications on the web[C]//Proceedings of the 17th International Conference on World Wide Web.ACM,2008:247-256.
[13] ZHU J,JIANG N,HU B.Application of Multiple Motion Parameters of Moving Objects in Trajectory Classification[J].Journal of Earth Sciences,2016,18(2):143-150.
[14] YE S N.Traffic pattern recognition based on deep convolutional network [D].Chengdu:Southwest Jiaotong University,2018.
[15] WANG J J.The safety of large tankers through the new,Ma Strait [J].China Navigation,2006 (3):27-30.
[16] SSEGEDY C,VANHOUCKE V,IOFFE S,et al.Rethinking the inception architecture for computer vision[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR).Las Vegas,USA:IEEE,2016:2818-2826.
[17] ZHANG L.Research on LSTM-based smartphone trajectory reco-gnition [D].Lanzhou:Lanzhou University of Technology,2018.
[18] QUAN B,YANG B C,HU K Q,et al.Ship track predictionmodel based on LSTM [J].Computer Science,2008,45(S2):126-131.
[19] GENG J L.Research on big data storage optimization and behavior recognition technology of fishing vessel trajectory [D].Hangzhou:Hangzhou University of Electronic Technology,2018.
[1] 王馨彤, 王璇, 孙知信.
基于多尺度记忆残差网络的网络流量异常检测模型
Network Traffic Anomaly Detection Method Based on Multi-scale Memory Residual Network
计算机科学, 2022, 49(8): 314-322. https://doi.org/10.11896/jsjkx.220200011
[2] 赵冬梅, 吴亚星, 张红斌.
基于IPSO-BiLSTM的网络安全态势预测
Network Security Situation Prediction Based on IPSO-BiLSTM
计算机科学, 2022, 49(7): 357-362. https://doi.org/10.11896/jsjkx.210900103
[3] 康雁, 徐玉龙, 寇勇奇, 谢思宇, 杨学昆, 李浩.
基于Transformer和LSTM的药物相互作用预测
Drug-Drug Interaction Prediction Based on Transformer and LSTM
计算机科学, 2022, 49(6A): 17-21. https://doi.org/10.11896/jsjkx.210400150
[4] 王飞, 黄涛, 杨晔.
基于Stacking多模型融合的IGBT器件寿命的机器学习预测算法研究
Study on Machine Learning Algorithms for Life Prediction of IGBT Devices Based on Stacking Multi-model Fusion
计算机科学, 2022, 49(6A): 784-789. https://doi.org/10.11896/jsjkx.210400030
[5] 高堰泸, 徐圆, 朱群雄.
基于A-DLSTM夹层网络结构的电能消耗预测方法
Predicting Electric Energy Consumption Using Sandwich Structure of Attention in Double -LSTM
计算机科学, 2022, 49(3): 269-275. https://doi.org/10.11896/jsjkx.210100006
[6] 刘嘉琛, 秦小麟, 朱润泽.
基于LSTM-Attention的RFID移动对象位置预测
Prediction of RFID Mobile Object Location Based on LSTM-Attention
计算机科学, 2021, 48(3): 188-195. https://doi.org/10.11896/jsjkx.200600134
[7] 刘奇, 陈红梅, 罗川.
基于改进的蝗虫优化算法的红细胞供应预测方法
Method for Prediction of Red Blood Cells Supply Based on Improved Grasshopper Optimization Algorithm
计算机科学, 2021, 48(2): 224-230. https://doi.org/10.11896/jsjkx.200600016
[8] 彭斌, 李征, 刘勇, 吴永豪.
基于卷积神经网络的代码注释自动生成方法
Automatic Code Comments Generation Method Based on Convolutional Neural Network
计算机科学, 2021, 48(12): 117-124. https://doi.org/10.11896/jsjkx.201100090
[9] 景丽, 何婷婷.
基于改进TF-IDF和ABLCNN的中文文本分类模型
Chinese Text Classification Model Based on Improved TF-IDF and ABLCNN
计算机科学, 2021, 48(11A): 170-175. https://doi.org/10.11896/jsjkx.210100232
[10] 赵佳琦, 王瀚正, 周勇, 张迪, 周子渊.
基于多尺度与注意力特征增强的遥感图像描述生成方法
Remote Sensing Image Description Generation Method Based on Attention and Multi-scale Feature Enhancement
计算机科学, 2021, 48(1): 190-196. https://doi.org/10.11896/jsjkx.200600076
[11] 张玉帅, 赵欢, 李博.
基于BERT和BiLSTM的语义槽填充
Semantic Slot Filling Based on BERT and BiLSTM
计算机科学, 2021, 48(1): 247-252. https://doi.org/10.11896/jsjkx.191200088
[12] 胡鹏程, 刁力力, 叶桦, 仰燕兰.
基于人工特征与深度特征的DGA域名检测算法
DGA Domains Detection Based on Artificial and Depth Features
计算机科学, 2020, 47(9): 311-317. https://doi.org/10.11896/jsjkx.191000118
[13] 吕亿林, 田宏韬, 高建伟, 万怀宇.
结合百科知识与句子语义特征的关系抽取方法
Relation Extraction Method Combining Encyclopedia Knowledge and Sentence Semantic Features
计算机科学, 2020, 47(6A): 40-44. https://doi.org/10.11896/JsJkx.190700042
[14] 陈晋音, 蒋焘, 郑海斌.
基于信噪比分级的信号调制类型识别
Radio Modulation Recognition Based on Signal-noise Ratio Classification
计算机科学, 2020, 47(6A): 310-317. https://doi.org/10.11896/JsJkx.190800073
[15] 黄虹玮,刘玉娇,沈卓恺,张少伟,陈志敏,高阳.
基于深度学习网络模型的端到端航迹关联
End-to-end Track Association Based on Deep Learning Network Model
计算机科学, 2020, 47(3): 200-205. https://doi.org/10.11896/jsjkx.190400037
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!