Ship Target Research Articles

A Transfer Learning and Optimized CNN Based Maritime Vessel Classification System

Deep learning has been used to improve intelligent transportation systems (ITS) by classifying ship targets in interior waterways. Researchers have created numerous classification methods, but they have low accuracy and misclassify other ship targets. As a result, more research into ship classification is required to avoid inland waterway collisions. We present a new convolutional neural network classification method for inland waterways that can classify the five major ship types: cargo, military, carrier, cruise, and tanker. This method can also be used for other ship classes. The proposed method consists of four phases for the boosting of classification accuracy for Intelligent Transport Systems (ITS) based on convolutional neural networks (CNNs); efficient augmentation method, the hyper-parameter optimization (HPO) technique for optimum CNN model parameter selection, transfer learning, and ensemble learning are suggested. All experiments used Kaggle’s public Game of Deep Learning Ship dataset. In addition, the proposed ship classification achieved 98.38% detection rates and 97.43% F1 scores. Our suggested classification technique was also evaluated on the MARVEL dataset. This dataset includes 10,000 image samples for each class and 26 types of ships for generalization. The suggested method also delivered an excellent performance compared to other algorithms, with performance metrics with an accuracy of 97.04%, a precision of 96.1%, a recall of 95.92%, a specificity of 96.55%, and a 96.31% F1 score.

Sentinel-2 Research on the Detection and Classification Methods of Maritime Ship Targets from Remote Sensing Images

There are problems such as low recognition accuracy and large classification error in the existing classification methods for ship identification based on optical remote sensing images. In this paper, we will analyze the characteristics of ships and determine the indicative factors for applying remote sensing to monitor ships in combination with optical remote sensing images. Using optical remote sensing image data, combined with U-Net and AttU-Net deep neural network models, we assist in extracting new remote sensing indices with strong generality and clear physical meaning, and establishing rules for monitoring ships, so as to establish a more general and clear physical meaning of the monitoring and identification method of remote sensing satellite images. The method is applied and evaluated with port optical remote sensing image data. The data show that compared with traditional machine learning methods, the accuracy of ship monitoring using U-Net and AttU-Net deep learning models in this paper reaches 89.04%, and the recall rate and accuracy rate are better than SVM. it shows that the model can detect ships effectively.

The Rest Is History

The Rest Is History

A Research on Autonomous Collision Avoidance under the Constraint of COLREGs

In this paper, a decision-making model suitable for the collision avoidance (CA) of numerous target ships (TSs) is proposed, based on the principle of ship collision avoidance geometry and the characteristics of numerous target ships’ collision avoidance at sea. To ensure that the collision avoidance behaviors of own-ship (OS) are subject to the International Regulations for Preventing Collisions at Sea (COLREGS), this paper gives full consideration to the requirements of COLREGS within the scope of CA action and the time of collision avoidance. A ship CA simulation is established based on the Mathematical Modeling Group (MMG) model. To optimize the CA decision-making model, the influence of hydrodynamic force on steering time required to reach the new course is integrated into the collision avoidance simulation system. The simulation results show that the method can quickly and effectively determine a collision avoidance decision under the complex situation of numerous target ships and static obstacles, and it can consider the unpredictable strategies used by other vessels.

Silicon energy bulk material cargo ship detection and tracking method combining YOLOv5 and DeepSort

Silicon widely contained in sand is expected to become a new energy material with environmental protection, safety and low cost. The intelligent management of the exploitation and transportation of such energy resources has become an urgent demand. Aiming at solving the problem of low detection rate of silicon energy bulk material cargo ship targets in river monitoring videos, this paper proposes a multi-ship detection and tracking method based on YOLOv5x combined with DeepSort algorithm. In order to improve the detector recognition efficiency, CIoU Loss is used as the target bounding box regression loss function instead of GIoU Loss to speed up the bounding box regression rate while improving the localization accuracy; NMS is replaced by DIoU-NMS to tackle the problem of missed detection when the targets are dense. The structure of DeepSort appearance feature extraction network is adjusted and trained on a self-built ship dataset to reduce the identity switching caused by target occlusion. The experimental results show that YOLOv5x trained on Pytorch framework makes the model converge fast and accurate with mAP reaching 95.6%, and then combined with DeepSort tracking can achieve the detection and tracking of multiple types of ships. This can provide effective technical support for the supervision of illegal mining and transportation of silicon energy bulk material.

Energy Management Optimization of Fuel Cell Hybrid Ship Based on Particle Swarm Optimization Algorithm

In order to optimize the energy management strategy and solve the problem of the power quality degradation of fuel cell hybrid electric ships, a particle swarm optimization algorithm based energy management strategy is proposed in this paper. Taking a fuel cell ship as the target ship, a system simulation model is built in Matlab/Simulink to verify the proposed energy management strategy. Through simulations and comparisons, the bus voltage curve of the optimized hybrid power system fluctuates more gently, and the voltage sag is smaller. The amplitude of the voltage fluctuation under maneuvering conditions is reduced by 55% compared with that of the original ship. The charging and discharging process of the composite energy storage system is optimized under maneuvering conditions, the power quality of the marine power grid is improved, and the use of the energy management strategy can extend the service life of the battery.

A distributed coordinated path planning algorithm for maritime autonomous surface ship

The anti-collision path planning is vital for establishing the decision support system for Maritime Autonomous Surface Ship (MASS). Numerous methods have been developed to solve the problem of collision avoidance with static and dynamic obstacles in the past two decades. However, most of the studies are conducted with certain limitations, such as the assumption of constant velocity of target ships, the simplification of ship’ dynamic properties during manoeuvering, the insufficient consideration of coordination among different ships' actions, etc. In this article, a distributed coordinated path planning algorithm especially for multi-ship encounter is proposed. Combined with constraints of apparent action, time dimension and ship's dynamic properties, a Three-dimension Generalized Velocity Obstacle (TGVO) algorithm is adopted to generate practical and COLREGs compliant collision-free velocities for vessels. Compared to traditional collision avoidance parameter calculation approach, the impact of ship manoeuvrability on avoidance performance is further discussed and analyzed to improve the accuracy and universality of the optimization solutions. In addition, a distributed multistage decision model aiming at solving the situation including multiple give-way vessels is introduced based on the priority analysis. Finally, simulation experiments show that this proposed scheme can work properly in various maritime environments with high degree of coordination and consistency with navigation practice.

Arbitrary-Oriented Ship Detection Method Based on Long-Edge Decomposition Rotated Bounding Box Encoding in SAR Images

Due to the limitations of the horizontal bounding boxes for locating the oriented ship targets in synthetic aperture radar (SAR) images, the rotated bounding box (RBB) has received wider attention in recent years. First, the existing RBB encodings suffer from boundary discontinuity problems, which interfere with the convergence of the model, and then lead to some problems, such as the inaccurate location of the ship targets in the boundary state. Thus, from the perspective that the long-edge features of the ships are more representative of their orientation, the long-edge decomposition RBB encoding has been proposed in this paper, which can avoid the boundary discontinuity problem. Second, the problem of the positive and negative samples imbalance is serious for the SAR ship images because only a few ship targets exist in the vast background of these images. Since the ship targets of different sizes are subject to varying degrees of interference caused by this problem, a multiscale elliptical Gaussian sample balancing strategy has been proposed in this paper, which can mitigate the impact of this problem by labeling the loss weights of the negative samples within the target foreground area with multiscale elliptical Gaussian kernels. Finally, experiments based on the CenterNet model were implemented on the benchmark SAR image dataset SSDD (SAR ship detection dataset). The experimental results demonstrate that our proposed long-edge decomposition RBB encoding outperforms other conventional RBB encodings in the task of oriented ship detection in SAR images. In addition, our proposed multiscale elliptical Gaussian sample balancing strategy is effective and can improve the model performance.

An Improved Method for Ship Target Detection Based on YOLOv4

The resolution of remote sensing images has increased with the maturation of satellite technology. Ship detection technology based on remote sensing images makes it possible to monitor a large range and far sea area, which can greatly enrich the monitoring means of maritime departments. In this paper, we conducted research on small target detection and resistance to complex background interference. First, a ship dataset with four types of targets (aircraft carriers, warships, merchant ships and submarines) is constructed, and experiments are conducted on the dataset using the object detection algorithm YOLOv4. The Kmeans++ clustering algorithm is used for a priori frame selection, and the migration learning method is used to enhance the detection effect of the YOLOv4. Second, the model is improved to address the problems of missed detection of small ships and difficulty in resisting background interference: the RFB_s (Receptive Field Block) with dilated convolution is introduced instead of the SPP (Spatial Pyramid Pooling) to enlarge the receptive field and improve the detection of small targets; the attention mechanism CBAM (Convolutional Block Attention Module) is added to adjust the weights of different features to highlight salient features useful for ship detection task, which improve the detection performance of small ships and improve the model’s ability to resist complex background. Compared to YOLOv4, our proposed model achieved a large improvement in mAP (mean Average Precision) from 77.66% to 91.40%.

Joint correction method for ionospheric phase pollution of high‐frequency sky‐surface wave radar based on adaptive optimal path

AbstractThe high‐frequency (HF) hybrid sky‐surface wave system is a new early warning radar. However, due to the influence of bistatic angle and ionospheric phase path disturbance, the echo spectrum will be widened, which seriously affects the ability of detecting slow targets of ships. In view of the overlapping of positive and negative Bragg broadening peaks caused by the large amplitude ionospheric phase path disturbance in the echo spectrum of HF sky‐surface wave radar, a joint correction method of ionospheric phase pollution based on adaptive optimal path is proposed in this paper. Firstly, the time‐frequency spectrum polluted by ionospheric phase is obtained by high resolution time‐frequency MUSIC transform, then the ionospheric phase disturbance curve in the time spectrum of polluted echo is extracted by the proposed adaptive optimal path method, and then the polluted echo is initially compensated to obtain the sharpened ocean echo spectrum. Finally, the residual phase of pollution curve is estimated by the Phase Gradient Autofocus method, and then the pollution echo is accurately corrected. The results of theoretical and measured data show that this method is effective in correcting the overlapping of positive and negative Bragg broadening peaks caused by serious ionospheric phase pollution.

MC-LRF based pose measurement system for shipborne aircraft automatic landing

Due to the portability and anti-interference ability, vision-based shipborne aircraft automatic landing systems have attracted the attention of researchers. In this paper, a Monocular Camera and Laser Range Finder (MC-LRF)-based pose measurement system is designed for shipborne aircraft automatic landing. First, the system represents the target ship using a set of sparse landmarks, and a two-stage model is adopted to detect landmarks on the target ship. The rough 6D pose is measured by solving a Perspective-n-Point problem. Then, once the rough pose is measured, a region-based pose refinement is used to continuously track the 6D pose in the subsequent image sequences. To address the low accuracy of monocular pose measurement in the depth direction, the designed system adopts a laser range finder to obtain an accurate range value. The measured rough pose is iteratively optimized using the accurate range measurement. Experimental results on synthetic and real images show that the system achieves robust and precise pose measurement of the target ship during automatic landing. The measurement means error is within 0.4° in rotation, and 0.2% in translation, meeting the requirements for automatic fixed-wing aircraft landing.Received 5 July 2022; revised 19 August 2022; accepted 27 September 2022.

Multi-scale ship target detection using SAR images based on improved Yolov5

Synthetic aperture radar (SAR) imaging is used to identify ships, which is a vital task in the maritime industry for managing maritime fisheries, marine transit, and rescue operations. However, some problems, like complex background interferences, various size ship feature variations, and indistinct tiny ship characteristics, continue to be challenges that tend to defy accuracy improvements in SAR ship detection. This research study for multiscale SAR ships detection has developed an upgraded YOLOv5s technique to address these issues. Using the C3 and FPN + PAN structures and attention mechanism, the generic YOLOv5 model has been enhanced in the backbone and neck section to achieve high identification rates. The SAR ship detection datasets and AirSARship datasets, along with two SAR large scene images acquired from the Chinese GF-3 satellite, are utilized to determine the experimental results. This model’s applicability is assessed using a variety of validation metrics, including accuracy, different training and test sets, and TF values, as well as comparisons with other cutting-edge classification models (ARPN, DAPN, Quad-FPN, HR-SDNet, Grid R-CNN, Cascade R-CNN, Multi-Stage YOLOv4-LITE, EfficientDet, Free-Anchor, Lite-Yolov5). The performance values demonstrate that the suggested model performed superior to the benchmark model used in this study, with higher identification rates. Additionally, these excellent identification rates demonstrate the recommended model’s applicability for maritime surveillance.

Long-Range Risk-Aware Path Planning for Autonomous Ships in Complex and Dynamic Environments

Abstract Path planning and collision avoidance are common problems for researchers in vehicle and robotics engineering design. In the case of autonomous ships, the navigation is guided by the regulations for preventing collisions at sea (COLREGs). However, COLREGs do not provide specific guidance for collision avoidance, especially for multi-ship encounters, which is a challenging task even for humans. In short-range path planning and collision avoidance problems, the motion of target ships is often considered as moving at a constant velocity and direction, which cannot be assumed in long-range planning and complex environments. The research challenge here is how to factor in the uncertainty of the motion of the target ships when making long-range path plans. In this paper, we introduce a long-range path planning algorithm for autonomous ships navigating in complex and dynamic environments to reduce the risk of encountering other ships during future motion. Based on the information on the position, speed over ground, and course over ground of other ships, our algorithm can estimate their intentions and future motions based on the probabilistic roadmap algorithm and use a risk-aware A* algorithm to find the optimal path that has low accumulated risk of encountering other ships. A case study is carried out on real automatic identification systems (AIS) datasets. The result shows that our algorithm can help reduce multi-ship encounters in long-term path planning.

A Spatial Cross-Scale Attention Network and Global Average Accuracy Loss for SAR Ship Detection

A neural network-based object detection algorithm has the advantages of high accuracy and end-to-end processing, and it has been widely used in synthetic aperture radar (SAR) ship detection. However, the multi-scale variation of ship targets, the complex background of near-shore scenes, and the dense arrangement of some ships make it difficult to improve detection accuracy. To solve the above problem, in this paper, a spatial cross-scale attention network (SCSA-Net) for SAR image ship detection is proposed, which includes a novel spatial cross-scale attention (SCSA) module for eliminating the interference of land background. The SCSA module uses the features at each scale output from the backbone to calculate where the network needs attention in space and enhances the features of the feature pyramid network (FPN) output to eliminate interference from noise, and land complex backgrounds. In addition, this paper analyzes the reasons for the “score shift” problem caused by average precision loss (AP loss) and proposes the global average precision loss (GAP loss) to solve the “score shift” problem. GAP loss enables the network to distinguish positive samples and negative samples faster than focal loss and AP loss, and achieve higher accuracy. Finally, we validate and illustrate the effectiveness of the proposed method by performing it on SAR Ship Detection Dataset (SSDD), SAR-ship-dataset, and High-Resolution SAR Images Dataset (HRSID). The experimental results show that the proposed method can significantly reduce the interference of background noise on the ship detection results, improve the detection accuracy, and achieve superior results to the existing methods.

Improvement and Optimization Configuration of Inland Ship Power and Propulsion System

Advances in power and propulsion and energy management improvements can significantly contribute to reducing emissions. The International Maritime Organization (IMO) Marpol regulations impose increasingly stringent restrictions on ship’s emission. According to the measured data of the target ship in typical working profiles, the power fluctuation, fuel consumption and emission data are analyzed, and the result represented that there are serious fuel consumption and pollution problems in the diesel engine power system. Based on the ship-engine propeller matching design theory, the ship-engine propeller model was built, and the new propulsion system power of the target ship was obtained by simulation. From the perspectives of power, economy and green, the performance and emission indexes of diesel engine and LNG engine are compared and analyzed, and the fuel cost advantage, green advantage and power performance disadvantage of LNG engine compared with diesel engine are determined. By comparing the topological structures of different hybrid propulsion forms, the new propulsion form of the ship is improved to be the gas-electric hybrid propulsion system based on the ESS (Energy Storage System), and the selection of the supercapacitors and lithium batteries is compared. Based on the low-pass filter strategy, the power distribution of the ultracapacitor and lithium battery is distributed. In order to determine the optimal ESS configuration, a capacity configuration model with investment cost, fuel cost and energy storage life as objective functions was established. NGSA-II algorithm was used to calculate the model and scheme selection was completed based on the scheme decision model. In this case, the optimal scheme significantly reduces pollutant emissions, it also reduces daily fuel costs by 38% and the result shows that we can complete the cost recovery in 1.28 years.

Bistatic SAR Maritime Ship Target 3-D Image Reconstruction Method Without Distortion in Local Cartesian Coordinate

Bistatic synthetic aperture radar (BiSAR) has been attracting worldwide attention because of its forward-looking imaging and high anti-interference. In harsh environment, it is vital for BiSAR to conduct extensive surveillance, imaging, and recognition of maritime ship targets. However, under the disturbance of sea waves, the ship target has an unknown and massive three-dimensional (3-D) rotation, so that its imaging projection plane (IPP) is also undetermined. Thus, high-dimensional random distortion appears in imaging results, making it difficult to recognize the target through two-dimensional distorted images effectively. To solve these problems, bistatic SAR maritime ship target 3-D image reconstruction method without distortion in local Cartesian coordinate (LCC) is proposed. In this paper, according to positions of scatterers and rotation parameters, significant differences of different scatterers of maritime ship targets have been found in bistatic range, Doppler centroid (DC), and Doppler frequency rate (DFR), which lays a solid foundation for the scatterer separation of ship targets. On this basis, a 3-D R-DC-DFR domain is constructed, and 2-D echoes of the maritime ship target are projected into R-DC-DFR domain to separate scatterers. Then, by remapping data of transmitter and receiver in R-DC-DFR domain to LCC, as well as evaluating their similarity metric, the optimal rotation parameters of the ship target can be obtained via the maximal similarity. Therefore, the image distortion caused by the unknown IPP has been removed, and 3-D image reconstruction of ship targets can be realized without distortion in the LCC. Furthermore, to evaluate performances of 3-D image reconstruction for different rotation parameters and bistatic configurations, 3-D reconstruction index is proposed and analyzed. Both point-targets and maritime ship targets are simulated to emphasize the effectiveness of the proposed method.

Inshore Dense Ship Detection in SAR Images Based on Edge Semantic Decoupling and Transformer

Synthetic aperture radar ship detection has recently received significant attention from scholars. However, accurately distinguishing between ships is challenging due to the significant overlap between inshore ship labels. In addition, some labeled boxes contain interference information, such as land areas, which can cause false alarms and confusion in ship feature learning. To address these challenges, this article creates an edge semantic decoupling (ESD) module, adds semantic segmentation branches, and introduces the edge semantic information of ships into the training process. As a result, the model can accurately distinguish between ship targets even when significant overlap exists between inshore labeled boxes. In addition, considering that transformer has the benefit of capturing global and contextual information, this article introduces it into the detection layer to construct a transformer detection layer (TDL) to limit the interference of land and other regions within the labeled box. Experimental results from the public SAR ship detection dataset show that the proposed ESD module and TDL detection layer effectively distinguish different ship targets in the inshore dense ship area, which is less affected by interference areas, such as land in the labeled box. The average precision improves to 96.72%, and both false alarms and miss detections inshore are reduced.

Monostatic-Equivalent Algorithm via Taylor Expansion for BiSAR Ship Target Imaging

It is well known that the bistatic synthetic aperture radar (BiSAR) can obtain high-resolution imaging in the forward direction of radar, which is not possible with the conventional monostatic SAR systems. However, most studies on BiSAR imaging mainly focus on stationary targets, and the research on moving targets and ship targets is relatively scarce. In this article, a monostatic-equivalent (ME) algorithm via Taylor expansion for BiSAR imaging of ship target is proposed. The monostatic equivalence method based on Taylor expansion is adopted to approximate BiSAR to monostatic model; besides, the method based on the midpoint of baseline is also analyzed for comparison. Then, the hybrid SAR-ISAR technology is utilized for ship target imaging under the ME model. Finally, the experimental results are given to substantiate the effectiveness and robustness of the proposed algorithm.

Estimation of Doppler Parameters Based on PSO for BFSAR Ship Target Imaging

Bistatic forward-looking synthetic aperture radar (BFSAR) can create a highly focused image of a target in the flight direction of the receiver. However, most research on BFSAR imaging is aimed at stationary targets, with relatively few studies on ship targets. We propose DPEP, an estimate of Doppler parameters, based on particle swarm optimization (PSO) for ship target imaging in the BFSAR system. The relationship between the Doppler parameters and those of the motion of a ship target is derived from the geometry of the BFSAR system. PSO is adopted to accurately estimate the Doppler parameters, so as to construct the functions for compression. The proposed algorithm is validated and compared to traditional algorithms through simulations.

Vision Transformer With Enhanced Self-Attention for Few-Shot Ship Target Recognition in Complex Environments

Ship target recognition is essential for maritime transportation, commercial trade, maritime security, and monitoring illegal activity. The majority of previous ship target recognition models are based on fully-supervised learning, which necessitates vast amounts of labeled training data. However, optical lenses capturing ship images in natural scenarios have numerous complex environments, making feature extraction difficult due to hardware, weather, light, waves, and other factors. Moreover, image annotation is also both expensive and time-consuming. How to effectively learn from unbalanced, limited, and sparsely annotated sample data is a significant challenge. In this paper, we propose a new vision transformer with enhanced self-attention(SAVT) that can locate ship targets in complex environments with just a few shots. SAVT consists of a novel enhanced self-attention(ESA) module and a novel measurement self-adjusting computation(MAC) module to differentiate unique ship classes in complex environments with few labeled samples. The ESA module focuses on the features of the target-region at each layer of the backbone network in order to ensure that the contour detailed features are extracted effectively from ship images with complex environments. The model is then enhanced in terms of speed and stability by utilizing the MAC module. CIB-ships and public MAR-ships datasets are used to evaluate the proposed method. In this paper, the results demonstrate that SAVT achieves good performance on few-shot labeled data under both 1-shot and 5-shot settings, while also providing valuable guidance for maritime ship target recognition with limited labeled samples.