A number of accidents at sea are primarily caused by low visibility due to sea fog. Therefore, it is important to estimate sea fog intensity and visibility in real-time in the ocean environment. Visibility meters utilize optical sensors rather than visional information, so that the measured visibility data occasionally includes errors. Moreover, visibility meters have significant costs so that it is not viable to install them at various locations. Therefore, this paper proposes an algorithm called RDCP (Reduced Dark Channel Prior), which provides reliable estimation at a low cost by processing images captured from cameras, since they receive identical information as human eyes. For the estimation, the RDCP algorithm firstly acquires dark channels from an ocean image, then applies an optimized threshold value to the dark channels and crops out the sky region in the ocean image. For the estimation performance evaluation, 320 raw images captured from cameras at four different ports in Republic of Korea are used. Considering facilities in oceans, which are usually operated in power-limited environment, the processing time performance of PDCP is also examined. The experiments demonstrate that the RDCP algorithm provides reliable estimation performance in real-time.

The sea has a very wide, irregular, and continuously changing surface and is usually a mixed sea composed of several wave systems. Each wave system is generated from different locations and conditions and has its own characteristics. The Fourier domain approach using sea wave spectra is an effective technique for the realistic simulation of sea surfaces in real time, but the conventional Fourier domain approach cannot independently simulate the characteristics of each wave system. In this paper, we propose a realistic and real-time simulation method of the mixed sea using multiple spectrum-based wave systems for maritime simulators. We recognize the importance of the visual and physical contributions of each wave system and faithfully reproduce all wave systems in the mixed sea. In order to simulate the mixed sea, our method generates and combines multiple spectrum-based wave systems using adaptive spectral sampling of the separated spectrum of the multi-peaked spectrum. The unique characteristics of each wave system can be set independently through spectral parameters, sampling number and range, wave direction and spread, and the shape factor of waves. The proposed method also supports the smooth transition between sea states, such as wind sea, swell, and mixed sea. Through the experiments, we verify that the proposed method effectively reflects sea wave spectra and the reproduced sea has very similar statistical characteristics to the actual sea. Experimental results also show that our approach can simulate the mixed sea, which has high-frequency wind sea and low-frequency swell.

The identification of maritime objects is crucial for ensuring navigational safety, enabling effective environmental monitoring, and facilitating efficient maritime search and rescue operations. Given its ability to provide detailed spectral information, hyperspectral imaging has emerged as a powerful tool for analyzing the physical and chemical properties of target objects. This study proposes a novel maritime object identification framework that integrates hyperspectral imaging with machine learning models. Hyperspectral data from six ports in South Korea were collected using airborne sensors and subsequently processed into spectral statistics and RGB images. The processed data were then analyzed using classifier and convolutional neural network (CNN) models. The results obtained in this study show that CNN models achieved an average test accuracy of 90%, outperforming classifier models, which achieved 83%. Among the CNN models, EfficientNet B0 and Inception V3 demonstrated the best performance, with Inception V3 achieving a category-specific accuracy of 97% when weights were excluded. This study presents a robust and efficient framework for marine surveillance utilizing hyperspectral imaging and machine learning, offering significant potential for advancing marine detection and monitoring technologies.

A number of accidents at sea are primarily caused by low visibility due to sea fog therefore, it is important to estimate sea fog intensity and visibility in real-time in the ocean environment. However, the existing estimation systems have significant costs so that it is not viable to install them at various locations, such as small vessels, minor ports, and a number of buoys. Therefore, this paper proposes an algorithm called RDCP (Reduced Dark Channel Prior) which provides the estimation at a low complexity and a low cost. For the estimation, the RDCP algorithm firstly acquires dark channels from an ocean image, then applies an optimized threshold value to the dark channels, and crops out the sky region in the ocean image. For the estimation performance evaluation, 320 raw images captured from cameras at four different ports in South Korea are used rather than artificially composed images. Considering facilities in oceans which are usually operated in the power limited environment, the processing time performance of PDCP are also examined. The experiments demonstrate that the RDCP algorithm provides reliable estimation performance in real-time.

When a policy trained with deep reinforcement learning (DRL) in simulation is deployed in the real world, its performance often deteriorates due to the Sim2Real gap. This study addresses this problem for Autonomous Surface Vessels (ASVs) by developing a robust collision-avoidance framework. We integrate a MATLAB-based ship dynamics model with ROS and Gazebo, and employ the Twin Delayed Deep Deterministic Policy Gradient (TD3) algorithm. To enhance robustness and generalization, we combine domain randomization and curriculum learning. As a result, the trained agent consistently achieved a high success rate of over 90% in unseen environments, significantly outperforming a baseline TD3 agent and a conventional PID controller. This demonstrates that the proposed Sim2Real methods are highly effective for creating robust control policies for ASVs. For future work, we plan to validate the learned policy through real-world experiments.

In applications such as environmental monitoring, algorithms and deep learning-based methods using synthetic aperture radar (SAR) and electro-optical (EO) data have been proposed with promising results. These results have been achieved using already cleaned datasets for training data. However, in real-world data collection, data are often collected regardless of environmental noises (clouds, night, missing data, etc.). Without cleaning the data with these noises, the trained model has a critical problem of poor performance. To address these issues, we propose the Clean Collector Algorithm (CCA). First, we use a pixel-based approach to clean the QA60 mask and outliers. Secondly, we remove missing data and night-time data that can act as noise in the training process. Finally, we use a feature-based refinement method to clean the cloud images using FID. We demonstrate its effectiveness by winning first place in the SAR-to-EO translation track of the MultiEarth 2023 challenge. We also highlight the performance and robustness of the CCA on other cloud datasets, SEN12MS-CR-TS and Scotland&India.