Arbitrary Ship Research Articles

YOLOv7oSAR: A Lightweight High-Precision Ship Detection Model for SAR Images Based on the YOLOv7 Algorithm

Researchers have explored various methods to fully exploit the all-weather characteristics of Synthetic aperture radar (SAR) images to achieve high-precision, real-time, computationally efficient, and easily deployable ship target detection models. These methods include Constant False Alarm Rate (CFAR) algorithms and deep learning approaches such as RCNN, YOLO, and SSD, among others. While these methods outperform traditional algorithms in SAR ship detection, challenges still exist in handling the arbitrary ship distributions and small target features in SAR remote sensing images. Existing models are complex, with a large number of parameters, hindering effective deployment. This paper introduces a YOLOv7 oriented bounding box SAR ship detection model (YOLOv7oSAR). The model employs a rotation box detection mechanism, uses the KLD loss function to enhance accuracy, and introduces a Bi-former attention mechanism to improve small target detection. By redesigning the network’s width and depth and incorporating a lightweight P-ELAN structure, the model effectively reduces its size and computational requirements. The proposed model achieves high-precision detection results on the public RSDD dataset (94.8% offshore, 66.6% nearshore), and its generalization ability is validated on a custom dataset (94.2% overall detection accuracy).

Formulation of Anchor and Chain Specifications for Establishing Subsea Cable Burial Depth

The penetration of ship anchor, which is a very important factor considered in the design stage of subsea cable burial, is highly dependent on the geometrical specifications and shapes of the anchor and chain. In order to facilitate predicting the anchor penetration depth by an arbitrary ship, formulations of the correlation between the ship mass and the existing anchor and chain specifications were attempted. In particular, for small ships with an equipment number of 40 (corresponding to DWT 90) or less, which were not covered by the existing classification data, a method for estimating the anchor mass, chain diameter, and mass per unit length was proposed.

KeyShip: Towards High-Precision Oriented SAR Ship Detection Using Key Points

Synthetic Aperture Radar (SAR) is an all-weather sensing technology that has proven its effectiveness for ship detection. However, detecting ships accurately with oriented bounding boxes (OBB) on SAR images is challenging due to arbitrary ship orientations and misleading scattering. In this article, we propose a novel anchor-free key-point-based detection method, KeyShip, for detecting orientated SAR ships with high precision. Our approach uses a shape descriptor to model a ship as a combination of three types of key points located at the short-edge centers, long-edge centers, and the target center. These key points are detected separately and clustered based on predicted shape descriptors to construct the final OBB detection results. To address the boundary problem that arises with the shape descriptor representation, we propose a soft training target assignment strategy that facilitates successful shape descriptor training and implicitly learns the shape information of the targets. Our experimental results on three datasets (SSDD, RSDD, and HRSC2016) demonstrate our proposed method’s high performance and robustness.

Near-Shore ship segmentation based on I-Polar Mask in remote sensing

ABSTRACT With the development of deep convolutional neural network, the performance of instance segmentation algorithms has been significantly improved. However, it is difficult to implement quick and accurate near-shore ship segmentation due to the complex background and arbitrary ship orientation, which makes segmentation challenging. To improve the segmentation efficiency of near-shore ships, this paper proposes a ship segmentation network based on Polar Mask, named I-Polar Mask. Specifically, we construct a Specific Polar Template (SPT) and Oriented Polar IoU (Intersection over Union) to better match the ship contour. Furthermore, oriented polar centre-ness was designed to reduce the weight of low-quality masks more effectively. In addition, a Context Information Extraction Module (CIEM) is built to reduce the influence of complex backgrounds and make the segmentation more accurate. To verify the effectiveness of the proposed algorithm, we collected 1015 images of near-shore areas from Google Maps and labelled them with Labelme to construct a ship instance segmentation dataset, called I-Ship. Extensive experiments on I-Ship show that the AP value of I-Polar Mask is improved by 5.9% compared with Polar Mask, which is a significant improvement. Compared with advanced methods, I-Polar Mask outperforms both quantitative and qualitative aspects.

MoLoc: Unsupervised Fingerprint Roaming for Device-Free Indoor Localization in a Mobile Ship Environment

Device-free indoor localization may play a critical role in improving passengers' safety in large vessels, particularly for scenarios without equipped radios. However, due to dynamic internal and external influences from the sailing ship such as changing sailing speed, the existing localization systems suffer huge accuracy degradation in a mobile ship environment. The challenges are mainly due to rich and arbitrary ship motions and the resulting complicated impacts on the indoor wireless channels. To address the challenges, in this article, we first propose a ship motion descriptor to extract discriminative latent representation from complex ship motions by leveraging deep-learning techniques. Based on this representation, we then design a novel fingerprint roaming model, i.e., MoLoc, to automatically learn the predictive fingerprint variation pattern and transfer the online fingerprint measurement to adapt to dynamic ship motions in real time. Furthermore, an unsupervised learning strategy is proposed to train the fingerprint roaming model using unlabeled onboard collected data which do not incur any labor costs. We have implemented and extensively evaluated MoLoc on real-world cruise ships, where experimental results demonstrate that MoLoc improves localization accuracy from 63.2% to 92.8% compared to the state-of-the-art localization methods, including Pilot, LiFS, SpotFi, and AutoFi while achieving a mean error of 0.68 m.

A Framework of A Ship Domain-Based Near-Miss Detection Method Using Mamdani Neuro-Fuzzy Classification

Abstract Safety analysis of navigation over a given area may cover application of various risk measures for ship collisions. One of them is percentage of the so called near-miss situations (potential collision situations). In this article a method of automatic detection of such situations based on the data from Automatic Identification System (AIS), is proposed. The method utilizes input parameters such as: collision risk measure based on ship’s domain concept, relative speed between ships as well as their course difference. For classification of ships encounters, there is used a neuro-fuzzy network which estimates a degree of collision hazard on the basis of a set of rules. The worked out method makes it possibile to apply an arbitrary ship’s domain as well as to learn the classifier on the basis of opinions of experts interpreting the data from the AIS.

Estimation of Weibull distribution for wind speeds along ship routes

In order to evaluate potential benefits of new green shipping concepts that utilize wind power as auxiliary propulsion in ships or of offshore wind energy harvest, it is essential to have reliable wind speed statistics. A new method to find parameters in the Weibull distribution is given. It can be used either at a fixed offshore position or along arbitrary ship routes. The method employs a spatio-temporal transformed Gaussian model for wind speed variability. The model was fitted to 10 years’ ERA-Interim reanalysis data of wind speed. The proposed method to derive Weibull distribution is validated using wind speeds measured on-board by vessels sailing in the North Atlantic and the west region of the Mediterranean Sea. For the westbound voyages in the North Atlantic, the proposed method gives a good approximation of the observed wind distribution along those ship routes. For the eastbound voyages, significant difference is found between the observed wind distribution and that approximated by the proposed method. The suspected reason is attributed to the ship routing decisions of masters and software. Hence, models that consider only the wind climate description need to be supplemented with a method to take into account the effect of wind-aware routing plan.

임의 선박 단면형상의 입수충격력에 관한 수치적 연구

The interaction between the hull of ship and free surface of water generates important loads during slamming motion. In the present study, the slamming load applied on the sectional surface of two-dimensional arbitrary bodies has been investigated under several falling velocities. This simulation has been done with the commercial CFD software ANSYS FLUENT ® . Through the conventional MARINTEK experiments for the benchmark of the simulation, we verified the impact pressure values between the experiments and simulation results. Two arbitrary ship bow section models, Panamax-like(with small convex bulb and flare) and Post panamax-like(with large convex bulb and flare) are also investigated. Simulation results show that a maximum impact pressure on the Post panama-like shape is higher than the Panamax-like shape. According to both a lump of water generated by arbitrary shape and various dead-rise angles of the shape, the pressure picks were enhanced in the simulation.

Interference effects on the Kelvin wake of a monohull ship represented via a continuous distribution of sources

Wave-interference effects on the wave pattern of a monohull ship that advances at constant speed along a straight path in calm water are considered. The flow around the ship is represented via a continuous distribution of sources over the hull surface, rather than via a point source at the bow and a point sink at the stern. A practical and realistic numerical method for determining the apparent wake angle ψmax, where the largest waves are found at high Froude numbers, for general ship hulls (including multihulls and/or distributions of pressure at the free surface) is considered. The method only involves elementary numerical computations that can be performed simply and efficiently. The method is therefore used here to perform systematic computations for seven ship hulls that correspond to broad ranges of main hull-shape parameters (beam/length, draft/length, beam/draft, waterline entrance angle) at ten Froude numbers. This numerical study shows that the apparent wake angle ψmax is only weakly influenced by the shape of the ship hull. A useful practical consequence of this numerical finding is that the wake angle ψmax can be estimated for general monohull ships via simple analytical relations. These relations provide relatively-accurate practical estimates (without computations) of the apparent wake angle ψmax(F) for arbitrary ship hull forms and Froude numbers. The explicit approximations for the wake angle ψmax(F) obtained here account for both longitudinal interference between the divergent waves created by the fore and aft regions of a monohull ship and lateral interference between the divergent waves created by the port and starboard sides of the hull. As expected, these relations yield an estimate of the wake angle ψmax(F) for a monohull ship that is more precise than the analytical estimate given by an elementary analysis of interference between the dominant waves created by the bow and the stern of a ship, which corresponds to a 2-point wavemaker approximation of the continuous hull-surface distribution of sources considered here.

A comparison study of water impact and water exit models

In problems of global hydroelastic ship response in severe seas including the whipping problem, we need to know the hydrodynamic forces acting on the ship hull during almost arbitrary ship motions. In terms of ship sections, some of them can enter water but others exit from water. Computations of nonlinear free surface flows, pressure distributions and hydrodynamic forces in parallel with the computations of the ship motions including elastic vibrations of the ship hull are time consuming and are suitable only for research purposes but not for practical calculations. In this paper, it is shown that the slamming forces can be decomposed in two components within three semi-analytical models of water entry. Only heave motion is considered. The first component is proportional to the entry speed squared and the second one to the body acceleration. The coefficients in these two components are functions of the penetration depth only and can be precomputed for given shape of the body. During the exit stage the hydrodynamic force is proportional to the acceleration of the body and independent of the body shape for bodies with small deadrise angles.

A more robust multiparameter conformal mapping method for geometry generation of any arbitrary ship section

The central problem of strip theory is the calculation of potential flowaround 2D sections. One particular method of solutions to this problem is conformal mapping of the body section to the unit circle over which a solution of potential flow is available. Here, a new multiparameter conformal mapping method is presented that can map any arbitrary section onto a unit circle with good accuracy. The procedure for finding the corresponding mapping coefficients is iterative. The suggested mapping technique is shown to be capable of appropriately mapping any chined, bulbous, and large and fine sections. Several examples of mapping symmetric and nonsymmetric sections are demonstrated. For symmetric and nonsymmetric sections, the results of the current method are compared against other mapping techniques, and the currently produced geometries display good agreement with the actual geometries.

Development of a Spectral Method and a Statistical Wave Model for Crack Propagation Prediction in Ship Structures

As a result of the wide use of high-tensile steel and the imperfect fabrication process in ship construction as well as some uncertainties in ship fatigue design, cracks could be initiated much earlier than expected. The presence of fatigue cracks greatly affects a ship's structural safety and serviceability. To ensure structural safety and perform reliable crack inspection and maintenance planning, it is important to know how fast cracks can grow in ship structures. In the current study, the principles of fracture mechanics are used for crack propagation analysis in ships. By taking into account the special properties of a ship's stress response, an efficient spectral method is proposed and validated for the prediction of crack propagation in ship structures. In this spectral method, structural stresses are assumed to be narrow-band Gaussian processes. Furthermore, for crack inspection and maintenance planning based on crack growth, it is essential to know the wave environments encountered in a ship's future operations. Therefore, a spatiotemporal statistical wave model based on both satellite and buoy measurements is briefly introduced. It is developed to generate wave environments along arbitrary ship routes. Finally, the deck longitudinal stiffener of a 2800TEU containership is used to demonstrate the application of the spectral method and the wave model for the prediction of crack propagation in ships. The route information and operating conditions are taken from full-scale measurements of this ship. In the case study, the scatter of crack propagation associated with the wave environments encountered is also investigated. The results from this investigation indicate possible potentials of crack inspection and maintenance optimization to enable more efficient ship operation.

Development of a Spectral Method and a Statistical Wave Model for Crack Propagation Prediction in Ship Structures

As a result of the wide use of high-tensile steel and the imperfect fabrication process in ship construction as well as some uncertainties in ship fatigue design, cracks could be initiated much earlier than expected. The presence of fatigue cracks greatly affects a ship's structural safety and serviceability. To ensure structural safety and perform reliable crack inspection and maintenance planning, it is important to know how fast cracks can grow in ship structures. In the current study, the principles of fracture mechanics are used for crack propagation analysis in ships. By taking into account the special properties of a ship's stress response, an efficient spectral method is proposed and validated for the prediction of crack propagation in ship structures. In this spectral method, structural stresses are assumed to be narrow-band Gaussian processes. Furthermore, for crack inspection and maintenance planning based on crack growth, it is essential to know the wave environments encountered in a ship's future operations. Therefore, a spatiotemporal statistical wave model based on both satellite and buoy measurements is briefly introduced. It is developed to generate wave environments along arbitrary ship routes. Finally, the deck longitudinal stiffener of a 2800TEU containership is used to demonstrate the application of the spectral method and the wave model for the prediction of crack propagation in ships. The route information and operating conditions are taken from full-scale measurements of this ship. In the case study, the scatter of crack propagation associated with the wave environments encountered is also investigated. The results from this investigation indicate possible potentials of crack inspection and maintenance optimization to enable more efficient ship operation.

Prediction of Ship Unsteady Maneuvering in Calm Water by a Fully Nonlinear Ship Motion Model

This is the continuation of our research on development of a fully nonlinear, dynamically consistent, numerical ship motion model (DiSSEL). In this study we will report our results in predicting ship motions in unsteady maneuvering in calm water. During the unsteady maneuvering, both the rudder angle, and ship forward speed vary with time. Therefore, not only surge, sway, and yaw motions occur, but roll, pitch and heave motions will also occur even in calm water as heel, trim, and sinkage, respectively. When the rudder angles and ship forward speed vary rapidly with time, the six degrees-of-freedom ship motions and their interactions become strong. To accurately predict the six degrees-of-freedom ship motions in unsteady maneuvering, a universal method for arbitrary ship hull requires physics-based fully-nonlinear models for ship motion and for rudder forces and moments. The numerical simulations will be benchmarked by experimental data of the Pre-Contract DDG51 design and an Experimental Hull Form. The benchmarking shows a good agreement between numerical simulations by the enhancement DiSSEL and experimental data. No empirical parameterization is used, except for the influence of the propeller slipstream on the rudder, which is included using a flow acceleration factor.

Simulation of Sloshing in LNG-Tanks

The purpose of this paper was to demonstrate the application of a procedure to predict internal sloshing loads on partially filled tank walls of liquefied natural gas (LNG) tankers that are subject to the action of sea waves. The method is numerical. We used a moving grid approach and a finite-volume solution method designed to allow for arbitrary ship motions. An interface-capturing scheme that accounts for overturning and breaking waves computed the motion of liquid inside the tanks. The method suppressed numerical mixing. Mixing effects close to the interface were buried in the numerical treatment of the interface. This interface, which was at least one cell wide, amounted to about 20–50 cm at full scale. Droplets and bubbles smaller than mesh size were not resolved. Tank walls were considered rigid. The results are first presented for an LNG tank whose motion was prescribed in accordance with planned laboratory experiments. Both two-dimensional and three-dimensional simulations were performed. The aim was to demonstrate that (1) realistic loads can be predicted using grids of moderate fineness, (2) the numerical method accurately resolves the free surface even when severe fragmentation occurs, and (3) long-term simulations over many oscillation periods are possible without numerical mixing of liquid and gas. The coupled simulation of a sea-going full-sized LNG tanker with partially filled tanks demonstrated the plausibility of this approach. Comparative experimental data were unavailable for validation; however, results were plausible and encouraged further validation.

Some Probabilistic Attributes of Inspection Policies in Alien Species Management

Inspections are a salient part of alien species management and, therefore, in contemporary times, there has been great interest in analyzing the role of inspections as an apposite regulatory tool. In this paper, we continue the study of inspections in a probabilistic environment and we shed light on three specific issues concerning inspection policies in the context of alien species management. First, given a particular seaport, we determine the likelihood that n ships will need to be inspected in a relevant time period. Second, we compute the probability distribution function of the amount of time it takes to inspect an arbitrary ship. Finally, we show how the probability distribution function of the total cost of inspections in the above time period can be calculated.

Nonlinear Ship–Wave Interaction Model, Part 2: Ship Boundary Condition

One of the main problems in modeling ship-wave hydrodynamics is solving for the forcing (pressure) at the ship boundary. With an arbitrary ship, singularities occur in evaluating the velocity potential and the velocities on the hull. Inaccuracies in the evaluation of the singular terms in the velocity potential result in discretization errors, numerical errors, and excessive computational costs. In this paper we present a new approach to evaluating the pressure on a ship: evaluating the pressure in the ship normal vector coordinate system. The main advantage of this approach is that p is a smooth function in the ship normal vector coordinate system, enabling us to obtain accurate results very efficiently (i.e., with much less CPU time) without encountering numerical difficulties arising from possible singularities.

Geometric Modeling for Fully Nonlinear Ship-Wave Interactions

Using the desingularized boundary integral method to solve transient nonlinear water-wave problems requires the solution of a mixed boundary value problem at each time step. The problem is solved at nodes (or collocation points) distributed on an ever-changing body surface. In this paper, a dynamic node allocation technique is developed to distribute efficiently nodes on the body surface. A B-spline surface representation is employed to generate an arbitrary ship hull form in parametric space. A variational adaptive curve grid generation method is then applied on the hull station curves to generate effective node placement. The numerical algorithm uses a conservative form of the parametric variational Euler-Lagrange equations to perform adaptive gridding on each station. Numerical examples of node placement on typical hull cross sections and for fully nonlinear wave resistance computations are presented.

Description of Ship Form and Evaluation of Ship Performance

Ship form has been expressed using the “OFF-SETS” and “PRINCIPAL PARTICULARS (PRINCIPAL DIMENSIONS)” with “LINES” for about 100 years since the beginning of the modern developments of ship engineering/science.In the field of ship design, ship performance has been estimated and evaluated by means of indirect parameters such as principal dimensions and coefficients. Since these parameters do not give an unique ship form, ship designer has utilize their experience and intuition during the procedure of performance design for improving and determining the ship form. However, how long are we able to pass on the skilful knowledge to the future generation ?In the present paper, the authors propose a method for describing ship form by means of Fourier series called Fourier Descriptors (FD). They will show that FD can determine an arbitrary ship section uniquely and can decompose/reproduce arbitrary 2-dimensional ship section form.As the 1 st step of the performance evaluation, they show the effects of section form on the damping and added mass coefficients of 2-dimensional oscillating bodies that are commonly used in the ship motion estimation with comparing results obtained by means of Lewis Form transformation.

The Detail Pipe Fitting Design Method based on Conversational Analyzing & Drafting System (CADS)

The detail pipe fitting design method is presented here utilizing versatile CADS which is the man-machine system conversational with mini-computer and large table of analyzing and drafting. By this method pipe fitting detail plan can be designed. So far computerization in the outfitting arrangement design field has been fallen behind.The characteristics of this system are as follows :1) Input Data can be generated by this system for the production design system, such as pipe piece drawing, pipe fabrication schedules, material lists etc., and all these data are connected to the large scale computer system in form of magnetic tape.2) This system is connected to the automatic piping arrangement design system (PPAC-IHI Program) to design pipe outline arrangement plan by the use of large scale computer, and also generated the data for practising pipe design process from outline pipe arranging to detail pipe fitting and piece. drawings.3) Several arbitrary ship's sections can be projected at on time on the wide table of CADS.4) The ship's design data can be generated from this system by in-putting minimal data, using the master data files.