High-precision habitat mapping can contribute to the identification and quantification of the human footprint on the seafloor. As a representative of seafloor habitats, seabed sediment classification is crucial for marine geological research, marine environment monitoring, marine engineering construction, and seabed biotic and abiotic resource assessment. Multibeam echo-sounding systems (MBES) have become the most popular tool in terms of acoustic equipment for seabed sediment classification. However, sonar images tend to consist of obvious noise and stripe interference. Furthermore, the low efficiency and high cost of seafloor field sampling leads to limited field samples. The factors above restrict high accuracy classification by a single classifier. To further investigate the classification techniques for seabed sediments, we developed a decision fusion algorithm based on voting strategies and fuzzy membership rules to integrate the merits of deep learning and shallow learning methods. First, in order to overcome the influence of obvious noise and the lack of training samples, we employed an effective deep learning framework, namely random patches network (RPNet), for classification. Then, to alleviate the over-smoothness and misclassifications of RPNet, the misclassified pixels with a lower fuzzy membership degree were rectified by other shallow learning classifiers, using the proposed decision fusion algorithm. The effectiveness of the proposed method was tested in two areas of Europe. The results show that RPNet outperforms other traditional classification methods, and the decision fusion framework further improves the accuracy compared with the results of a single classifier. Our experiments predict a promising prospect for efficiently mapping seafloor habitats through deep learning and multi-classifier combinations, even with few field samples.

Seafloor habitat mapping plays an important role in marine environment monitoring and marine geological research. Optic and acoustic remote sensing are becoming common survey tools in seafloor habitat mapping. However, a single acoustic or optic technique may have a limited detection range and be more susceptible to the impact of image quality. Additionally, it is challenging to satisfy the requirements for accurate detection since single-source data cannot fully reflect the substrate distribution characteristics. This paper developed a method for detecting coastal seafloor habitats through the fusion of multi-scale optics and acoustics data. First, the original feature set was composed of multispectral satellite data and bathymetric data by multibeam echosounder (MBES) and airborne light detection and ranging (LiDAR) at different scales, which improved the capacity to represent feature information. Then, a ReliefF–mRMR method was implemented to select optimal features with appropriate scales and remove redundant features. Finally, the optimal features were employed in model training and classification of several supervised classifiers to verify the effectiveness of the strategy. The developed method was applied to the Ganquan Island survey in the South China Sea. The results demonstrated that, after integrating multisource data, the accuracies were up to 3.31% and 17.28% higher than those obtained using multispectral data or bathymetric data alone, respectively. ReliefF–mRMR exhibited better performance than other feature selection methods. The average coral coverage in the study area was estimated to range from 70.85% to 80.33%. This research highlights the greater potential of multisource data for precisely detecting seafloor habitats.

Acoustic imaging systems are essential for underwater target recognition and localization, but forward-looking sonar (FLS) imagery faces challenges due to seabed variability, resulting in low resolution, blurred images, and sparse targets. To address these issues, we introduce RCF-YOLOv8, an enhanced detection framework based on YOLOv8, designed to improve FLS image analysis. Key innovations include the use of CoordConv modules to better encode spatial information, improving feature extraction and reducing misdetection rates. Additionally, an efficient multi-scale attention (EMA) mechanism addresses sparse target distributions, optimizing feature fusion and improving the network’s ability to identify key areas. Lastly, the C2f module with high-quality feature fusion (C2f-Fusion) optimizes feature extraction from noisy backgrounds. RCF-YOLOv8 achieved a 98.8% mAP@50 and a 67.6% mAP@50-95 on the URPC2021 dataset, outperforming baseline models with a 2.4% increase in single-threshold accuracy and a 10.4% increase in multi-threshold precision, demonstrating its robustness for underwater detection.

As sound speed is a fundamental parameter of ocean acoustic characteristics, its prediction is a central focus of underwater acoustics research. Traditional numerical and statistical forecasting methods often exhibit suboptimal performance under complex conditions, whereas deep learning approaches demonstrate promising results. However, these methodologies fall short in adequately addressing multi-spatial coupling effects and spatiotemporal weighting, particularly in scenarios characterized by limited data availability. To investigate the interactions across multiple spatial scales and to achieve accurate predictions, we propose the STA-ConvLSTM framework that integrates spatiotemporal attention mechanisms with convolutional long short-term memory neural networks (ConvLSTM). The core concept involves accounting for the coupling effects among various spatial scales while extracting temporal and spatial information from the data and assigning appropriate weights to different spatiotemporal entities. Furthermore, we introduce an interpolation method for ocean temperature and salinity data based on the KNN algorithm to enhance dataset resolution. Experimental results indicate that STA-ConvLSTM provides precise predictions of sound speed. Specifically, relative to the measured data, it achieved a root mean square error (RMSE) of approximately 0.57 m/s and a mean absolute error (MAE) of about 0.29 m/s. Additionally, when compared to single-dimensional spatial analysis, incorporating multi-spatial scale considerations yielded superior predictive performance.

The occurrence characteristics of hydrates in the Shenhu area reflect a typical inhomogeneity in terms of spatial distribution. It is difficult to accurately describe the petrophysical properties of a reservoir using a petrophysical model considering a single cementation factor parameter. According to the analysis of a mathematical model and the estimation results of and , the unique structure of foraminiferal sediment particles provides opportunities for forming a diversified hydrate occurrence in the foraminiferal area. In areas where hydrates are thin and interbedded, hydrate reservoirs are generally three-phase media, with obvious thermoelastic properties. Therefore, the parameters of the three characteristic models of the pore-filling model, particle cementation model, and thermodynamic elastic model are all included in the correction model. The weights of the influence factors are then changed to realize an accurate description of the petrophysical characteristics of the correction model in different drilling areas and at different formation depths, reducing the limitations of using a single petrophysical model to describe the petrophysical characteristics of heterogeneous regions under the influence of multiple factors.