Robust Ship Research Articles

MLANet: A Robust Ship Segmentation Network Based on Multilevel Multiattention Feature Fusion for Complex Maritime Background Environments

MLANet: A Robust Ship Segmentation Network Based on Multilevel Multiattention Feature Fusion for Complex Maritime Background Environments

Visual Ship Image Synthesis and Classification Framework Based on Attention-DCGAN

To improving ship image generation and classification tasks, a deep convolution generative adversarial network based on attention mechanism (ADCGAN) model was constructed. The rectified linear unit (ReLU) activation function was adopted, and three Deconv layers and Conv layers were added to both the generator and discriminator. Subsequently, an attention mechanism was added to the generator, while spectral normalization (SN) was added to the discriminator. Mean squared error (MSE) was used as loss function to stabilize the training process. Furthermore, ship classification tasks were performed using the generated ship images by end-to-end training of the classification network, enabling ship data augmentation and co-learning with other tasks. Experimental results on the Ship700 and Seaship7000 datasets demonstrate that the ADCGAN model can generate clear and robust ship images, with PSNR, LIPIPS, MS-SSIM values of 20.279 and 27.523, 0.596 and 0.096, 0.781 and 0.947, respectively. The effectiveness of the proposed method in ship image classification tasks was also verified, providing a data foundation for other collaborative tasks.

SwinYOLOv7: Robust ship detection in complex synthetic aperture radar images

Using satellite-based SAR (Synthetic Aperture Radar) imagery to detect and track ships is a formidable challenge. However, accurate analysis is hampered by inherent difficulties such as obscured edges, multiple targets, varying dimensions and complex backgrounds. These factors contribute to a poor signal-to-noise ratio. Various artificial intelligence models have been developed to improve these problems, especially with YOLO-based models. To overcome the above challenges and achieve robust performance, this study presents a model called SwinYOLOv7, a novel fusion of the YOLOv7 framework with the Features Pyramid Network and the Swin Transformer using ground-breaking anchor-free detection algorithms. This innovative method aims to increase the accuracy of vessel detection while reducing the impact of background clutter. The proposed model improved by YOLOv7 examined three different datasets in detail: SRSDD-v1.0, HRSID, and SSDD to optimize the performance of the model. The training process was consistently verified using superior recall, precision, and F1-score values, which can be easily compared with previous studies. The results show that the model using the Swin Transformer attention mechanism and using an image size of 640×640 achieves the highest accuracy of 96.59%. Alternative attention mechanisms, including the Squeeze-and-Excitation Network (SEnet), the Convolutional Block Attention Module (CBAM), Channel Attention (CA) and Efficient Channel Attention (ECA), deliver poorer accuracy rates. The combination of YOLOv7 and Swin Transformer yielded encouraging results that enabled the proposed model to outperform the current benchmarking models. Therefore, the proposed model provides a compelling solution to ensure accurate vessel identification in complex search and rescue scenarios.

A Deep Learning Approach for Ship Detection Using Satellite Imagery

INTRODUCTION: This paper addresses ship detection in satellite imagery through a deep learning approach, vital for maritime applications. Traditional methods face challenges with large datasets, motivating the adoption of deep learning techniques.
 OBJECTIVES: The primary objective is to present an algorithmic methodology for U-Net model training, focusing on achieving accuracy, efficiency, and robust ship detection. Overcoming manual limitations and enhancing real-time monitoring capabilities are key objectives.
 METHOD: The methodology involves dataset collection from Copernicus Open Hub, employing run-length encoding for efficient preprocessing, and utilizing a U-Net model trained on Sentinel-2 images. Data manipulation includes run-length encoding, masking, and balanced dataset preprocessing.
 RESULT: Results demonstrate the proposed deep learning model's effectiveness in handling diverse datasets, ensuring accuracy through U-Net architecture, and addressing imbalances. The algorithmic process showcases proficiency in ship detection.
 CONCLUSION: In conclusion, this paper contributes a comprehensive methodology for ship detection, significantly advancing accuracy, efficiency, and robustness in maritime applications. The U-Net-based model successfully automates ship detection, promising real-time monitoring enhancements and improved maritime security.

MS3Net: a deep ensemble learning approach for ship classification in heterogeneous remote sensing data

ABSTRACT Maritime ship classification is essential for effectively monitoring oceanic activities but faces challenges when using heterogeneous remote sensing data. This research presents a novel dataset called Heterogeneous Ship Data, created by combining synthetic aperture radar (SAR) and optical satellite images of ships. The integration of multimodal data aims to harness the complementary advantages of SAR’s all-weather imaging capabilities and optical data’s higher interretability and shape discrimination. To leverage the heterogeneous dataset, this paper proposes an ensemble deep learning model called MS3Net (Multi-convolutional network using Spatial attention and Second-order pooling for Ship image classification). MS3Net integrates spatial attention mechanisms and second-order pooling. MS3Net has multiple convolutional branches, each tailored to extract distinct representations from the SAR and optical data through varied filter sizes and kernels. The spatial attention units focus the learning on the most informative parts of the image while suppressing irrelevant features. The second-order pooling captures higher-order interactions between feature elements to model complex relationships. A concatenation fusion strategy aggregates the diverse branch outputs into a joint multimodal feature representation for enhanced discrimination. Extensive experiments demonstrate MS3Net’s superior classification accuracy over prior state-of-the-art approaches. The proposed heterogeneous data representation coupled with the attention-based deep ensemble learning model contributes to maritime surveillance by enabling robust ship recognition across diverse sensing modalities.

MrisNet: Robust Ship Instance Segmentation in Challenging Marine Radar Environments

In high-traffic harbor waters, marine radar frequently encounters signal interference stemming from various obstructive elements, thereby presenting formidable obstacles in the precise identification of ships. To achieve precise pixel-level ship identification in the complex environments, a customized neural network-based ship segmentation algorithm named MrisNet is proposed. MrisNet employs a lightweight and efficient FasterYOLO network to extract features from radar images at different levels, capturing fine-grained edge information and deep semantic features of ship pixels. To address the limitation of deep features in the backbone network lacking detailed shape and structured information, an adaptive attention mechanism is introduced after the FasterYOLO network to enhance crucial ship features. To fully utilize the multi-dimensional feature outputs, MrisNet incorporates a Transformer structure to reconstruct the PANet feature fusion network, allowing for the fusion of contextual information and capturing more essential ship information and semantic correlations. In the prediction stage, MrisNet optimizes the target position loss using the EIoU function, enabling the algorithm to adapt to ship position deviations and size variations, thereby improving segmentation accuracy and convergence speed. Experimental results demonstrate MrisNet achieves high recall and precision rates of 94.8% and 95.2%, respectively, in ship instance segmentation, outperforming various YOLO and other single-stage algorithms. Moreover, MrisNet has a model parameter size of 13.8M and real-time computational cost of 23.5G, demonstrating notable advantages in terms of convolutional efficiency. In conclusion, MrisNet accurately segments ships with different spot features and under diverse environmental conditions in marine radar images. It exhibits outstanding performance, particularly in extreme scenarios and challenging interference conditions, showcasing robustness and applicability.

Ship dynamics model identification based on Semblance least square support vector machine

In recent years, with the rapid development of Maritime Autonomous Surface Ships (MASS) in the industry and academia, using system identification method to build the accurate ship dynamics model has become a hotspot in this area, which is critical to achieve precise and robust ship motion control for safety and adaptive navigation of MASS. Therefore, a non-parametric and robust two-phase system identification method is proposed. Firstly, the improved complete ensemble empirical mode decomposition is introduced to filter datasets collected from model-scale free-running model tests. Secondly, Semblance least square support vector machine (S-LS-SVM) is proposed for the ship dynamics modeling based on the LS-SVM with the state-of-the-art Semblance kernel function. Compared with the traditional method, the Root Mean Square Error (RMSE) of overall prediction on the test dataset improves by 20.82%, 7.68%, and 58.19% for surge, sway, and yaw speed of target ship model. According to maneuverability simulation, it is verified that this method has better generalization performance and higher accuracy for the prediction of the ship’s heading angle and trajectory.

Robust Ship Berthing Control with Wind Disturbance Compensation using Reachability Analysis

Automatic ship berthing under environmental disturbances is considered one of the most challenging control tasks. In particular, wind is a dominant disturbance that greatly impacts ships’ berthing operations. Feedforward control with disturbance observation is a common strategy to compensate for wind effects, however the disturbance observation is limited to mild low-frequency disturbances. Its prediction accuracy and the resultant control performance of wind-effect compensation degrade in severe wind conditions. This study proposes a new control strategy for wind compensation based on the reachable set reflecting the characteristics of the controller and bounded wind disturbance. State feedback linearization is used for berthing control and zonotope sets are employed for reachability analysis. Simulation results are presented to demonstrate the proposed wind compensation method's feasibility and performance.

Robust ship fleet deployment with shipping revenue management

The fleet deployment problem is a tactical planning problem in the shipping industry that seeks to efficiently assign ships to predetermined shipping services to meet the uncertain and seasonal changing demands caused by a highly fluctuating market. Once the fleet deployment plan is fixed, the allocated capacity on each service affects the possible flows transported and the profit obtained. This paper proposes a two-stage robust optimization model for ship fleet deployment and shipping revenue management of a liner shipping network under demand uncertainty. The randomness of demand in our model is represented by probability-free uncertain sets. A column-and-constraint generation based exact algorithm is designed to solve our model based on the analysis of its structural properties. To further accelerate the convergence of our algorithm, an M-tightening technique has been exploited. Finally, extensive computational experiments based on realistic instances are conducted to validate the effectiveness of our model and the efficiency of the algorithm; further, managerial insights are acquired based on the numerical results.

Ship Steering Adaptive CGS Control Based on EKF Identification Method

In recent years, marine autonomous surface vessels (MASS) have grown into a ship research issue to increase the level of autonomy of ship behavior decision-making and control while sailing at sea. This paper focuses on the MASS motion control module design that aims to improve the accuracy and reliability of ship steering control systems. Nevertheless, the stochastic sea and wind environment have led to the extensive use of filters and state observers for estimating the ship-motion-related parameters, which are important for ship steering control systems. In particular, the ship maneuverability Nomoto index, which primarily determines the designed ship steering controller’s performance, cannot be observed directly due to the model errors and the external environment disturbance in the process of sailing. Hence, an adaptive robust ship steering controller based on a closed-loop gain shaping (CGS) scheme and an extended Kalman filter (EKF) on-line identification method is explored in this paper. To verify the effectiveness of the proposed steering controller design scheme, the motor vessel YUKUN was taken as the control plant and a series of simulation experiments were carried out. The results show the advantages of the dynamic response performance of the proposed steering controller compared with the classical PD and traditional CGS controllers. Therefore, the proposed adaptive CGS steering controller would be a good solution for MASS motion control module design.

Robust Ship Detection in Infrared Images through Multiscale Feature Extraction and Lightweight CNN.

The sophistication of ship detection technology in remote sensing images is insufficient, the detection results differ substantially from the practical requirements, mainly reflected in the inadequate support for the differentiated application of multi-scene, multi-resolution and multi-type target ships. To overcome these challenges, a ship detection method based on multiscale feature extraction and lightweight CNN is proposed. Firstly, the candidate-region extraction method, based on a multiscale model, can cover the potential targets under different backgrounds accurately. Secondly, the multiple feature fusion method is employed to achieve ship classification, in which, Fourier global spectrum features are applied to discriminate between targets and simple interference, and the targets in complex interference scenarios are further distinguished by using lightweight CNN. Thirdly, the cascade classifier training algorithm and an improved non-maximum suppression method are used to minimise the classification error rate and maximise generalisation, which can achieve final-target confirmation. Experimental results validate our method, showing that it significantly outperforms the available alternatives, reducing the model size by up to 2.17 times while improving detection performance be improved by up to 5.5% in multi-interference scenarios. Furthermore, the robustness ability was verified by three indicators, among which the F-measure score and true–false-positive rate can increase by up to 5.8% and 4.7% respectively, while the mean error rate can decrease by up to 38.2%.

Investigation of robust ship course control based on auxiliary loop method

A novel robust control law is investigated on the problem of ship stabilization on the trajectory, which allows one to compensate perturbations in the parameters of the mathematical model of ship dynamics in cases of their changes caused by external conditions, namely: sea waves, currents, wind gusts, etc. To implement the proposed control law, only measured adjustable values are required such as the yaw angle and the control action that is the angle of the rudder of the ship. The synthesized ship course control system was investigated in MATLAB. The law of controlling the ship’s course with unknown parameters and external disturbances in the power supply is proposed. The design of the control law is based on a robust auxiliary loop algorithm and Khalil observers. The simulations illustrate the efficiency of the proposed control law.

LIFE-CYCLE CO2 EMISSIONS OF BULK CARRIERS: A COMPARATIVE STUDY

In order to maintain shipping capacity to serve seaborne trade, new ships have to be built to replace those scrapped. The cost of building, manning, operating, maintaining and repairing a ship throughout its life is borne by society at large through market mechanisms. Gratsos and Zachariadis (2005) had investigated through a cost/benefit analysis how the average annual cost of ship transport varies with the corrosion additions elected at the design stage. The results of that paper clearly indicated that ships built with sufficient corrosion allowances, truly adequate for the ship’s design life, have a lower life cycle cost per annum despite the fact that such ships would carry a slightly smaller quantity of cargo. Furthermore the safety and environmental benefits due to the reduced repairs and extended lifetime of such ships were briefly discussed. The debate of how “robust” a ship should be was also transferred to IMO in the context of Goal Based Standards following a submission by Japan which stated that the increased steel weight of a more robust ship will result in increased CO2 emissions due to a reduced cargo carrying capacity. Greece replied by submitting a summary of the aforementioned paper and preliminary estimations on Life cycle CO2 emissions disputing the Japanese contentions. However, taking onboard the challenge, an update is provided in the present paper, using the final Common Structural Rules (CSR) of the International Association of Classification Societies (IACS) bulk carrier corrosion margins and taking into account the major environmental implications of the heavier ship scantlings for two bulk carrier size brackets, Panamax and Handymax. The results show that the more robust ships would produce less CO2 emissions over their lifetime.

Deploying a shipping infrastructure to enable carbon capture and storage from Norwegian industries

This paper investigates the deployment over time of a shipping infrastructure to enable carbon capture and storage from Norwegian industries. In particular, optimal transport investments, routing, shipping portfolio, and transport conditions required cost-efficiently connect nine Norwegian and Swedish industrial facilities to the Northern Light Initiative are identified based on a multi-period mixed-integer model. Optimization of the shipping portfolio allows us to analyze the potential benefits of economics of scale and how sailing multi-stop routes can utilize the potential benefits.Analyses show that the transport cost can be reduced by 12% when the optimal shipping portfolio is implemented instead of operating an independent transport chain for each emission source. Shipping at 7-bar is expected to result in lower costs compared 15-bar shipping, but a significant drawback is the technological maturity of the 7-bar transport technology. The model proposes two transport chains run in parallel as cost optimal, the average transport cost of the 15-bar chain is 32.4 €/ton, and the 7-bar has an average cost of 25.4 €/ton. A net present value criterion is used to identify the conditions for when retrofitting of the 15-bar transport chain is economically sound. Break-even is achieved when the ship's re-sale price is 60% of the purchase price, and the cost of retrofitting a 15-bar conditioning plant to a 7-bar is 40% of the investment cost. While retrofitting may be an attractive solution, the decisions lead to a less robust shipping portfolio, and the retrofitting of a fully operating carbon capture and storage value chain is regarded as unlikely.

Robust Shipping Date Predictions in Manufacturing using High-level Petri Nets

ABSTRACT Machine failures, unavailability of the workforce, and missing raw materials are typical, frequent faults in complex manufacturing systems and cause unstable processes. Plans and schedules become volatile, predictability of shipping dates suffers and delayed delivery appears. However, for a customer reliable shipping can often be of higher benefit than promised short-term shipping with significant probability of delay and error. Inserting buffer times stabilises schedules by reducing the potential impact of disruptions. Therefore, computing appropriately sized buffer times for individual jobs in a flexible job shop manufacturing with setup times is a promising yet difficult mission for production research. In this paper, we discuss methods for buffer time calculation that are based on data collected directly from the underlying production processes. We show how a high-level Petri net model can be used to simulate stochastic variations of the manufacturing process. Using this simulation model, we estimate the effect of buffer times and produce predictions for the expected shipping dates together with probabilities for how likely the predicted dates will actually be met.

Adaptive modeling of maritime autonomous surface ships with uncertainty using a weighted LS-SVR robust to outliers

Maritime Autonomous Surface Ships (MASS) have become increasingly interesting for the commercial maritime sectors as an alternative to conventional ships. For the purpose of development of MASS operations such as motion control (e.g. collision avoidance, trajectory tracking), the ship dynamic model is of significant importance for technique tests, i.e. verification and validation. The ship dynamic model should be suitable for cases which means the sufficient balance between the complexity and the accuracy of the model. This contribution is aiming to develop a robust ship dynamics modeling approach by making efforts on determining a common ship dynamic model and designing a robust identification method. A response model, i.e. the first-order nonlinear Nomoto model widely-used in ship autopilot design is selected. A robust optimal identification method named optimal DW-LSSVR is proposed by taking advantages of least square support vector regression algorithm (LS-SVR), robust 3α principle (D), adaptive weight technique (W), and artificial bee colony algorithm (ABC for optimization). The robust 3α principle has the function of outlier detection. Adaptive weight technique can adaptively weight support vectors to improve the sparsity of LS-SVR. Furthermore, ABC is served as the structural parameters optimizer for LS-SVR. The ships studied in this work are considered as more realistic objects with uncertain parameters induced by unmodeled terms, variation of the loading condition as well as environmental disturbances, and measurement noises. In identifying the widely-used response model for ships, the optimal DW-LSSVR method is verified and validated on both experimental measurements and simulated data. In simulation tests, the effectiveness of the proposed method in identifying the response model has been demonstrated with the use of simulated data including ones with uncertainties which are generated by activating a response model with predefined parameter values by means of step inputs in Monte Carlo simulations. Besides the simulation tests, the experimental investigation on a real Unmanned Surface Vessel (USV) using experimental zigzag maneuvers also indicates the consistency of the proposed approach.

Precise and Robust Ship Detection for High-Resolution SAR Imagery Based on HR-SDNet

Ship detection in high-resolution synthetic aperture radar (SAR) imagery is a challenging problem in the case of complex environments, especially inshore and offshore scenes. Nowadays, the existing methods of SAR ship detection mainly use low-resolution representations obtained by classification networks or recover high-resolution representations from low-resolution representations in SAR images. As the representation learning is characterized by low resolution and the huge loss of resolution makes it difficult to obtain accurate prediction results in spatial accuracy; therefore, these networks are not suitable to ship detection of region-level. In this paper, a novel ship detection method based on a high-resolution ship detection network (HR-SDNet) for high-resolution SAR imagery is proposed. The HR-SDNet adopts a novel high-resolution feature pyramid network (HRFPN) to take full advantage of the feature maps of high-resolution and low-resolution convolutions for SAR image ship detection. In this scheme, the HRFPN connects high-to-low resolution subnetworks in parallel and can maintain high resolution. Next, the Soft Non-Maximum Suppression (Soft-NMS) is used to improve the performance of the NMS, thereby improving the detection performance of the dense ships. Then, we introduce the Microsoft Common Objects in Context (COCO) evaluation metrics, which provides not only the higher quality evaluation metrics average precision (AP) for more accurate bounding box regression, but also the evaluation metrics for small, medium and large targets, so as to precisely evaluate the detection performance of our method. Finally, the experimental results on the SAR ship detection dataset (SSDD) and TerraSAR-X high-resolution images reveal that (1) our approach based on the HRFPN has superior detection performance for both inshore and offshore scenes of the high-resolution SAR imagery, which achieves nearly 4.3% performance gains compared to feature pyramid network (FPN) in inshore scenes, thus proving its effectiveness; (2) compared with the existing algorithms, our approach is more accurate and robust for ship detection of high-resolution SAR imagery, especially inshore and offshore scenes; (3) with the Soft-NMS algorithm, our network performs better, which achieves nearly 1% performance gains in terms of AP; (4) the COCO evaluation metrics are effective for SAR image ship detection; (5) the displayed thresholds within a certain range have a significant impact on the robustness of ship detectors.

Enhanced Feature Extraction for Ship Detection from Multi-Resolution and Multi-Scene Synthetic Aperture Radar (SAR) Images

Independent of daylight and weather conditions, synthetic aperture radar (SAR) images have been widely used for ship monitoring. The traditional methods for SAR ship detection are highly dependent on the statistical models of sea clutter or some predefined thresholds, and generally require a multi-step operation, which results in time-consuming and less robust ship detection. Recently, deep learning algorithms have found wide applications in ship detection from SAR images. However, due to the multi-resolution imaging mode and complex background, it is hard for the network to extract representative SAR target features, which limits the ship detection performance. In order to enhance the feature extraction ability of the network, three improvement techniques have been developed. Firstly, multi-level sparse optimization of SAR image is carried out to handle clutters and sidelobes so as to enhance the discrimination of the features of SAR images. Secondly, we hereby propose a novel split convolution block (SCB) to enhance the feature representation of small targets, which divides the SAR images into smaller sub-images as the input of the network. Finally, a spatial attention block (SAB) is embedded in the feature pyramid network (FPN) to reduce the loss of spatial information, during the dimensionality reduction process. In this paper, experiments on the multi-resolution SAR images of GaoFen-3 and Sentinel-1 under complex backgrounds are carried out and the results verify the effectiveness of SCB and SAB. The comparison results also show that the proposed method is superior to several state-of-the-art object detection algorithms.

Robust Ship Tracking via Multi-view Learning and Sparse Representation

Conventional visual ship tracking methods employ single and shallow features for the ship tracking task, which may fail when a ship presents a different appearance and shape in maritime surveillance videos. To overcome this difficulty, we propose to employ a multi-view learning algorithm to extract a highly coupled and robust ship descriptor from multiple distinct ship feature sets. First, we explore multiple distinct ship feature sets consisting of a Laplacian-of-Gaussian (LoG) descriptor, a Local Binary Patterns (LBP) descriptor, a Gabor filter, a Histogram of Oriented Gradients (HOG) descriptor and a Canny descriptor, which present geometry structure, texture and contour information, and more. Then, we propose a framework for integrating a multi-view learning algorithm and a sparse representation method to track ships efficiently and effectively. Finally, our framework is evaluated in four typical maritime surveillance scenarios. The experimental results show that the proposed framework outperforms the conventional and typical ship tracking methods.

Robust Liner Ship Routing and Scheduling Schemes under Uncertain Weather and Ocean Conditions

Robust Liner Ship Routing and Scheduling Schemes under Uncertain Weather and Ocean Conditions