Image feature registration is a critical step for stitching and generating large-field images during the inland navigation of ships. To address the problems of sparse water surface feature points and low efficiency in traditional feature matching algorithms for image registration in inland navigation environments, this paper proposes a feature matching method based on image super-resolution reconstruction. Firstly, the input images are subjected to super-resolution reconstruction using generative adversarial networks to enrich image details and increase the number of image feature points. Secondly, the ORB operator and BEBLID algorithm are employed for feature point detection and description. Then, coarse matching is performed based on Hamming distance. Finally, an improved random sampling consistency algorithm is utilized to further eliminate gross errors and purify inliers, achieving robust matching results. The study conducts experiments using five sets of inland navigation environment images with challenges such as low visibility, varying lighting conditions, scale changes, blur, and rotation. The results demonstrate that the proposed approach, leveraging image super-resolution reconstruction for feature point extraction, achieves an increased number of feature points and outperforms comparative algorithms in terms of matching accuracy and speed. This method meets the requirements of high-precision and real-time image matching in inland navigation environments.

With the continuous increase in seaborne trade volume, ship traffic density in port areas is rising, and navigation conditions in port waters are becoming more complex. Short-term ship traffic prediction in port waters is playing an increasingly critical role in ship traffic control and navigation safety management. To address the limitation of low accuracy in aggregate models, this paper, based on ship Automatic Identification System (AIS) data, employs a disaggregate method to construct a hybrid prediction model. This model combines the Long Short-Term Memory (LSTM) network with ships' historical trajectories to calculate short-term ship trajectories in port waters. The counts of ships' trajectories intersecting with an approach channel section are used to predict the short-term ship flow across the section. A numerical example from Ningbo-Zhoushan Port during June to December 2020 demonstrates that the forecasting accuracy of the proposed model reaches up to 80%, significantly higher than that of traditional aggregate models. The model developed here provides a technical foundation for ports to implement ship traffic control methods and improve channel utilization rates.

A stochastic programming model is devised for the multi-base, multi-drone location and routing problem, considering the simultaneous movements of drones and ships as well as ship movement uncertainty. A decoding algorithm is developed to divide a sequence into sub-routes using ship-based and drone-based strategies. Furthermore, a bi-stage heuristic algorithm is proposed, combining a genetic algorithm and Tabu search. In the bi-stage algorithm, the first stage addresses ship movement uncertainty and employs Tabu search to solve the drone base station location problem. The second stage uses the genetic algorithm to route the drones for detection based on the location results. Numerical experiment results show that, in the same application scenario, the drone-based (D) strategy can optimize flying distance by 7% while reducing computing time by 50% compared to the ship-based (S) strategy. Considering ship movement uncertainty can reduce flying distance by 10% for the drone base station location solution. Flying distance is sensitive to the number of available drones. For example, in a scenario with two base stations and 3-5 drones, adding one drone may increase flying distance by 15%. Speeding up the drones by 5% may reduce flying distance by 5%. This method can effectively generate multi-UAV inspection paths that meet the requirements of moving ships, providing technical support for maritime supervision.

With the rapid development of computer technology, numerous new optimization methods and processes have emerged in the field of ship type optimization. However, there is still a lack of open-source and free optimization platforms in China that can efficiently integrate these optimization methods and processes. This article constructs an optimization platform based on the Grasshopper visual programming environment, integrating the fundamental steps of ship type optimization. By incorporating variable complexity methods into the optimization process, the platform addresses issues such as long optimization times and high computational costs, thereby enhancing its functionality and optimization capabilities. Using this platform, drag reduction optimization is performed on the KCS bulbous bow of container ships, and the newly designed ship form demonstrates superior drag performance compared to the original form. This verifies the correctness and feasibility of the platform and lays the foundation for further expansion of its functionalities.

To address the issues of the analytical-based F_QSD (Fuzzy Quaternion Ship Domain) model, such as the difficulty in defining its boundary, insufficient consideration of influencing factors, and challenges in practical application, this paper incorporates the human factor of ship navigator and establishes a D_QSD (Dynamic Quaternion Ship Domain) based on the ship navigator's state to enhance the analytical-based fuzzy quaternion ship domain. The shape parameter K value of the fuzzy quaternion ship domain is calculated according to the ship navigator's state, and the dynamic ship domain model under different ship navigator states is obtained through computer simulation. The results show that the shape and area of the ship domain dynamically change under different driving states. When the ship navigator's state is excellent, the ship domain shape is an irregular rhombus, and the domain area is the smallest. When the ship navigator's state is poor, the ship domain shape is approximately rectangular, and the domain area is the largest. Compared with the traditional analysis-based ship domain, the proposed model determines the fuzzy boundary of the ship domain according to the ship navigator's state, making the dynamic ship domain more flexible and adaptable.

A Vector Auto Regression (VAR) model is constructed by combining the China Containerized Freight Index of E/C America Service (CCFI E/C America Service) and the China Containerized Freight Index of W/C America Service (CCFI W/C America Service) with the Clarksons Container ship Port Congestion Index (CPCI) from January 2018 to February 2023, to quantitatively analyze the impact mechanism of port congestion on container freight rates. The model also incorporates a Vector Error Correction (VEC) model to study the long-run equilibrium relationship between the variables. The results show that: 1) Port congestion leads to the occupation of container capacity and port resources, as well as changes in the distribution of capacity and transportation strategies on the China-U.S. export container routes, which in turn causes different degrees of fluctuations in CCFI on the sub-routes; 2) The effect of port congestion on container freight rates persists for nearly three months; 3) Regardless of the U.S. East route or the U.S. West route, port congestion in the U.S. has a more significant impact on promoting the increase of the container freight index compared to port congestion in China. Meanwhile, this paper provides a new perspective for predicting CCFI by investigating the impact mechanism of port congestion on CCFI fluctuations.

Study on the location selection problem of multi-level offshore ship oil spill emergency equipment depots is of great guiding significance for the effective utilization of oil spill emergency resources, the improvement of regional oil spill emergency response capabilities, and the perfection of the oil spill emergency response system. According to the distribution of oil spill risk points and the predicted oil spill volume, combined with the characteristics of the emergency service radius and comprehensive removal and control capabilities of oil spill emergency equipment depots at different levels, a location selection model for multi-level offshore ship oil spill emergency equipment depots is established with the objectives of achieving the best coverage, the highest reliability, and the strongest timeliness for different risk waters. Then, the MOPSO algorithm and NSGA- Ⅱ algorithm are used to solve the model, and the Pareto optimal solution set is obtained. Starting from the perspective of coordinated rescue and based on the emergency service capabilities of equipment depots at different levels, the location selection model realizes the best multiple coverage of different risk waters, aiming to share emergency resources and further enhance the regional emergency response capabilities.

Complex meteorological sea conditions directly affect the safety of ship navigation, and the accuracy of the prediction of offshore wind speed, as a major factor in meteorological sea conditions, is of great significance to the navigation safety and trajectory planning. In order to effectively improve the accuracy of offshore wind speed prediction and overcome the limitations of a single prediction model, the offshore wind form data of Lianyungang station is used as an example study, and the Adaboost algorithm is used to integrate the advantages of multi-models to construct a combined prediction model of offshore wind speed. Four time series prediction models, including BP neural network, GA BPNN, long and short-term memory network and WOA-SVR, are used for wind speed prediction. Considering the prediction effect of a single model, Adaboost algorithm is applied to integrate the GA-BPNN model and WOA-SVR model to construct the combined offshore wind speed prediction model, and the integration accuracy is compared with that of Bagging algorithm. The results show that the root mean square error of the combined prediction model with the Adaboost algorithm is reduced by about 13% and the mean absolute error is reduced by about 16% compared with the single model, which effectively verifies the superiority of the combined prediction model in the prediction of offshore wind speed data, and it is of great significance for the enhancement of navigational safety and the optimization of the trajectory design.

All-electric tugboats (AETs) produce fewer carbon emissions, but their battery charging requirements can lead to ship delays, which may increase the overall carbon emission costs in ports. Therefore, it is essential to study the carbon emission reduction performance of AETs in ports and compare their effectiveness with that of diesel tugboats (DTs). This paper establishes scheduling optimization models for AETs and DTs, respectively, using the sum of tugboat carbon emissions and ship delay carbon emissions as the objective function. Taking a day's data from three port areas in Ningbo-Zhoushan Port as an example, the Gurobi solver is employed to find the optimal solution. A comparative study is conducted on the minimum tugboat carbon emission cost and total carbon emission cost when AETs and DTs with the same horsepower and quantity are used to handle the same ships. The results show that if only the carbon emission cost of tugboats is considered, AETs can reduce carbon emission costs by 9.5% to 22.9% compared to DTs. However, when considering the overall carbon emission cost of the port areas, AETs still demonstrate good carbon emission reduction effects when handling fewer than 8 ships. When the number of ships exceeds 8 and the port is busy, the charging requirements of AETs increase the overall carbon emission cost of the port by 24.68%, which is not conducive to port carbon emission reduction.

The use of diesel auxiliaries by ships during port calls causes a large amount of fossil energy consumption and pollutant emissions, and the use of shore power for energy supply is a good alternative. For the energy problem of shore power system, the article introduces a hybrid energy system composed of offshore wind turbines, shore power and hydrogen-based energy storage, and proposes a hydrogen-based energy storage planning model based on hybrid stochastic regularization-information gap decision theory. Aiming at the uncertainty of offshore wind turbine output, stochastic planning is used to get the time-sequence typical output scenario; for the difficulty of accurately portraying the probability distribution of shore power port call ship load and shore power price, IGDT is used to form a dual-objective model to deal with the uncertainty of the two as well as to introduce two different risky strategy planning models and analyze them by considering the seasonal factors. The results of the example show that the hybrid energy system can improve energy utilization and interaction, provide a planning basis for decision makers, and verify the effectiveness of the proposed model.