A ResNet1D-AttLSTM-Based Approach for Ship Trajectory Classification

To improve the feature extraction method for ship trajectories and enhance trajectory classification performance, this paper proposes a ship trajectory classification model that combines a one-dimensional residual network (ResNet1D) and an attention-based Long short-term memory network (AttLSTM). Th...

Full description

Saved in:
Bibliographic Details
Main Authors: Jiankang Ke, Faxing Lu, Yifei Liu, Bing Fu
Format: Article
Language:English
Published: MDPI AG 2025-03-01
Series:Applied Sciences
Subjects:
Online Access:https://www.mdpi.com/2076-3417/15/7/3489
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850184661490204672
author Jiankang Ke
Faxing Lu
Yifei Liu
Bing Fu
author_facet Jiankang Ke
Faxing Lu
Yifei Liu
Bing Fu
author_sort Jiankang Ke
collection DOAJ
description To improve the feature extraction method for ship trajectories and enhance trajectory classification performance, this paper proposes a ship trajectory classification model that combines a one-dimensional residual network (ResNet1D) and an attention-based Long short-term memory network (AttLSTM). The model aims to address the limitations of traditional methods in extracting feature patterns jointly represented by non-adjacent local regions in ship trajectories, optimized through the introduction of a self-attention mechanism. Specifically, the model first utilizes the ResNet1D module to progressively extract implicit motion pattern features from local to global levels, while the AttLSTM module captures temporal sequence features of ship trajectories. Finally, the fusion of these two types of features generates a more comprehensive and rich spatiotemporal motion feature representation, enabling accurate classification of five types of ship trajectories, including towing vessels, fishing vessels, sailing vessels, passenger ships, and tankers. Experimental results show that this model excels on extensive real-world trajectory datasets, achieving a classification accuracy of 89.7%, significantly outperforming models relying solely on single feature sets or lacking integrated attention mechanisms. This not only validates the model’s superior performance in ship trajectory classification tasks but also demonstrates its potential and effectiveness for practical applications.
format Article
id doaj-art-476f15f227704f33bdb0eed4b861e2dc
institution OA Journals
issn 2076-3417
language English
publishDate 2025-03-01
publisher MDPI AG
record_format Article
series Applied Sciences
spelling doaj-art-476f15f227704f33bdb0eed4b861e2dc2025-08-20T02:17:00ZengMDPI AGApplied Sciences2076-34172025-03-01157348910.3390/app15073489A ResNet1D-AttLSTM-Based Approach for Ship Trajectory ClassificationJiankang Ke0Faxing Lu1Yifei Liu2Bing Fu3Naval University of Engineering, Wuhan 430033, ChinaNaval University of Engineering, Wuhan 430033, ChinaNaval University of Engineering, Wuhan 430033, ChinaNaval University of Engineering, Wuhan 430033, ChinaTo improve the feature extraction method for ship trajectories and enhance trajectory classification performance, this paper proposes a ship trajectory classification model that combines a one-dimensional residual network (ResNet1D) and an attention-based Long short-term memory network (AttLSTM). The model aims to address the limitations of traditional methods in extracting feature patterns jointly represented by non-adjacent local regions in ship trajectories, optimized through the introduction of a self-attention mechanism. Specifically, the model first utilizes the ResNet1D module to progressively extract implicit motion pattern features from local to global levels, while the AttLSTM module captures temporal sequence features of ship trajectories. Finally, the fusion of these two types of features generates a more comprehensive and rich spatiotemporal motion feature representation, enabling accurate classification of five types of ship trajectories, including towing vessels, fishing vessels, sailing vessels, passenger ships, and tankers. Experimental results show that this model excels on extensive real-world trajectory datasets, achieving a classification accuracy of 89.7%, significantly outperforming models relying solely on single feature sets or lacking integrated attention mechanisms. This not only validates the model’s superior performance in ship trajectory classification tasks but also demonstrates its potential and effectiveness for practical applications.https://www.mdpi.com/2076-3417/15/7/3489ship trajectory classificationresidual neural network (ResNet)long short-term memory network (LSTM)self-attention mechanism
spellingShingle Jiankang Ke
Faxing Lu
Yifei Liu
Bing Fu
A ResNet1D-AttLSTM-Based Approach for Ship Trajectory Classification
Applied Sciences
ship trajectory classification
residual neural network (ResNet)
long short-term memory network (LSTM)
self-attention mechanism
title A ResNet1D-AttLSTM-Based Approach for Ship Trajectory Classification
title_full A ResNet1D-AttLSTM-Based Approach for Ship Trajectory Classification
title_fullStr A ResNet1D-AttLSTM-Based Approach for Ship Trajectory Classification
title_full_unstemmed A ResNet1D-AttLSTM-Based Approach for Ship Trajectory Classification
title_short A ResNet1D-AttLSTM-Based Approach for Ship Trajectory Classification
title_sort resnet1d attlstm based approach for ship trajectory classification
topic ship trajectory classification
residual neural network (ResNet)
long short-term memory network (LSTM)
self-attention mechanism
url https://www.mdpi.com/2076-3417/15/7/3489
work_keys_str_mv AT jiankangke aresnet1dattlstmbasedapproachforshiptrajectoryclassification
AT faxinglu aresnet1dattlstmbasedapproachforshiptrajectoryclassification
AT yifeiliu aresnet1dattlstmbasedapproachforshiptrajectoryclassification
AT bingfu aresnet1dattlstmbasedapproachforshiptrajectoryclassification
AT jiankangke resnet1dattlstmbasedapproachforshiptrajectoryclassification
AT faxinglu resnet1dattlstmbasedapproachforshiptrajectoryclassification
AT yifeiliu resnet1dattlstmbasedapproachforshiptrajectoryclassification
AT bingfu resnet1dattlstmbasedapproachforshiptrajectoryclassification