Search Results (469)

Utilizing underwater vehicles for hydropower infrastructure inspection is increasingly vital. However, these GNSS-denied and confined environments pose significant navigation challenges: Inertial Navigation Systems (INSs) suffer cumulative drift, Doppler Velocity Logs (DVLs) face acoustic blind zones near walls, and visual navigation frequently fails in highly turbid waters. To address these issues, this paper proposes a tightly coupled multi-source (INS/acoustic/optical/vision) navigation algorithm leveraging prior wall geometry constraints. Developed within an Error-State Kalman Filter (ESKF) framework, the model seamlessly accommodates sensor spatiotemporal heterogeneity. To overcome optical failures, a structural surface constraint model is innovatively constructed using single-beam sonar ranging. The core contribution involves transforming sonar ranging data into 6-DOF spatial pose constraints based on the dam’s planar characteristics, effectively bounding the localization drift perpendicular to the surface. Field experiments at the hydropower station dam demonstrate that under extreme conditions with total visual failure, the proposed algorithm effectively constrains critical motion degrees of freedom. By maintaining the wall-tracking error within 0.08 m (Root Mean Square Error, RMSE)—which effectively represents the relative localization error given the known absolute position of the structural wall—this method significantly enhances the operational robustness and precision of close-wall inspections in extreme underwater environments. Full article

Underwater RGB–sonar object detection remains challenging due to severe optical degradation, strong sonar noise, and spatial misalignment between heterogeneous modalities. Existing multimodal detectors usually rely on simple feature aggregation or limited structural coupling, which cannot effectively model global cross-modal dependencies or address modality-specific degradation. To address these challenges, we propose Chrs-Net, a YOLOv12-based dual-stream framework for underwater RGB–sonar object detection. The proposed network integrates three key components: a Transformer-based Cross-Modal Communication Fusion module (C-mcf) for global cross-modal interaction and semantic alignment, a Multi-Layer Feature Enhancement module (MLFE) for degraded optical feature enhancement, and a Pinwheel-Shaped Convolution module (PConv) for sonar-side structural feature extraction. In addition, an RGB–sonar object detection dataset is constructed for experimental evaluation by relabeling part of the RGBS benchmark, combining simulator-collected samples, and introducing style-transfer-based augmentation to improve data diversity. Experiments on the constructed dataset yield 94.91% mAP@0.5 and 61.10% mAP@0.5:0.95 on the RGB branch, and 94.00% and 57.13% on the sonar branch, respectively, with an inference speed of 53.6 FPS. Compared with representative single-modality and multimodal detectors, Chrs-Net consistently yields superior detection accuracy and localization performance. These results demonstrate that the combination of global cross-modal communication and modality-specific enhancement is effective for robust underwater RGB–sonar object detection in complex environments. Full article

Underwater target detection via side-scan sonar is vital for defense and economy but hindered by sparse targets, high data costs, and feature extraction difficulties due to textureless acoustic data and limited samples. To overcome these limitations, particularly for few-shot, small-object detection, we propose a Spatial Distribution Probability-Guided Detection Framework to aid Unmanned Underwater Vehicles (UUVs) in precise localization and clustering. The framework features a novel module that leverages a pre-trained Vision Foundation Model (DINOv3) to generate spatial distribution probability maps, guiding a Transformer-based network for accurate detection with scarce data. Additionally, it incorporates a Target Position Calculation Module and a DBSCAN-based post-processing module to determine global geographic coordinates and cluster discrete points, respectively. Experiments were conducted on both a Public Mine Detection Dataset and a self-collected dataset containing simulated mines and buoys. Ablation studies and comparison experiments demonstrated that the proposed guidance mechanism significantly improves detection performance. Furthermore, two comb-search missions verified that the system could accurately locate and cluster targets, distinguishing real targets from false detections (noise). These results confirm the framework’s efficacy in enabling high-precision perception and autonomous operations for complex underwater inspection tasks. Full article

Autonomous Underwater Vehicles (AUVs) are increasingly used in deep-sea exploration, environmental monitoring, and marine engineering. Their operational safety and mission performance rely heavily on accurate and long-endurance underwater localization. However, both single-sensor localization methods and existing multi-sensor fusion approaches have inherent limitations, making it difficult to achieve high-precision localization during long-duration missions. To address this issue, this study develops a deep-learning-based multi-source sensor fusion framework for AUV localization. In the proposed framework, high-frequency data from the Inertial navigation system (INS) and Doppler velocity log (DVL) are used for continuous position propagation, while low-frequency absolute position observations from the Ultra-short baseline (USBL) system and Sonar are used to periodically correct the propagated results. Based on this framework, three instantiated models are developed using a Deep neural network (DNN), a Long short-term memory (LSTM) network, and a Bayesian semi-supervised mixed shallow-layer neural network (BSsMSLNN), respectively. Comparative experiments are conducted against the Extended Kalman filter (EKF) and Simultaneous localization and mapping system using Sonar, Visual, Inertial, and Depth sensor (SVIn2). The results show that the proposed framework effectively suppresses long-term error accumulation and significantly improves localization accuracy. Among the evaluated models, the BSsMSLNN-based method achieves the best performance in terms of trajectory fitting, root mean square error (RMSE), and coefficient of determination (R²). The proposed method provides a feasible solution for high-precision autonomous navigation of AUVs in GPS-denied environments. Full article

High-precision identification of acoustic convergence zones (CZs) and acoustic shadow zones (SZs) is a core prerequisite for deep-sea sonar performance prediction and long-range underwater target detection. However, in data-scarce marine environments, traditional acoustic identification methods suffer from high environmental sensitivity and significant computational costs, while pure data-driven deep learning methods face dilemmas such as a lack of physical consistency and poor generalization on small samples. To address these issues, a three-level cascaded recognition framework based on physics-prior-augmented deep learning is proposed in this paper, enabling accurate segmentation of CZs and intelligent classification of sound field types under data-scarce scenarios. In this framework, physical acoustic principles are incorporated exclusively as priors through a training dataset generated by a Gaussian beam acoustic propagation code (Bellhop) and through hand-crafted geometric features derived post hoc from the initial segmentation outputs. Taking a typical deep-sea area in the Northwest Pacific Ocean as the research object, a hybrid dataset comprising 5000 simulated transmission loss images and 500 simulated images from a geographically distinct sea area is constructed. The sound field is categorized into four types: strong convergence, usable convergence, weak convergence, and shadow zone. In the first stage, the ResNet-34 backbone is improved by integrating deformable convolution and a global statistical feature module, which, combined with a joint loss function, achieves high-precision pixel-level segmentation of CZs and SZs, with the regional gray contrast reaching 86.9%. In the second stage, a customized dual-channel VGG16 architecture is designed to fuse the extracted geometric priors and visual features, achieving a sound field classification accuracy of 89.91%. In the third stage, a hybrid data augmentation technique combining Mixup and convolutional autoencoder is adopted alongside a transfer learning strategy to mitigate the data scarcity under cross-domain conditions, boosting the small-sample classification accuracy to 84.45%. The experimental results demonstrate that the models in each stage of the proposed framework significantly outperform traditional methods and baseline networks. This study provides a novel methodology and technical support for intelligent sound field identification in data-scarce marine environments. Finally, the core contributions and current limitations are summarized, and future research directions, such as constructing a dynamic hydrological parameter feedback mechanism and identifying three-dimensional complex sound fields, are prospected. Full article

Background: Data extraction for systematic reviews is highly resource-intensive. This study evaluated four frontier large language models (LLMs) on complex structured metadata extraction from specialized neuroimaging artificial intelligence (AI) literature to determine their performance in automated evidence synthesis. Methods: We compared Google Gemini 3 Pro Preview, Anthropic Claude Opus 4.5, Perplexity Sonar Pro, and OpenAI GPT 5.2. Using a standardized prompt, each model extracted 22 variables from 91 peer-reviewed neuroimaging AI articles. The variables were stratified into low-, medium-, and high-complexity tiers. The performance was measured via the exact-match accuracy against a consensus-based expert ground truth. Results: The overall exact-match accuracy was moderate. Gemini 3 Pro Preview achieved the highest overall rate (56.4%), followed by Sonar Pro (52.1%), Claude Opus 4.5 (51.3%), and GPT 5.2 (46.5%). Gemini significantly outperformed all other models (p < 0.001). The performance declined dramatically as the variable complexity increased. Across models, the accuracy was 88.9–92.9% for low-complexity categorical fields, 47.0–63.3% for medium-complexity text extraction, and 2.7–15.5% for high-complexity variables requiring clinical judgment or multi-section synthesis. The most common type of error was misclassification. All four models scored 0% on the main performance metric, but this reflected a representational mismatch with the ground truth rather than extraction failure, indicating that the exact-match accuracy underestimates the true semantic performance. Conclusions: Frontier LLMs can effectively automate the retrieval of simple categorical data, but have serious difficulties with methodological variables that are complex. Although extraction can be fully automated for low-complexity fields, human review remains essential for context-dependent variables that require clinical judgment. Full article

Subsea Christmas trees are often deployed in turbid coastal waters or seabed environments. During manipulator operations on Christmas tree panels, conventional optical servoing is severely limited by rapid electromagnetic attenuation and strong scattering from suspended particles, resulting in reduced visibility. Forward-looking sonar (FLS) provides stable imaging, but its unique imaging geometry and low resolution make direct 6D pose estimation challenging. To address this issue, this paper proposes a 6D object pose estimation method for FLS images, in which conventional optical control-point-based pose estimation is restructured to resolve the mismatch between optical-centric network assumptions and acoustic imaging characteristics, and is further integrated with acoustic projection-based pose inversion. First, to address the limited diversity of target appearances and the scarcity of training data, we construct an FLS imaging model based on primary truncation for image simulation, providing data for model pretraining. Second, a multi-task acoustic control-point detection network, Acoustic-Yolo6D, is designed to mitigate localization degradation caused by heavy speckle noise, low boundary contrast, and resolution variations associated with polar-coordinate imaging, through heatmap regression, auxiliary object segmentation, and explicit range-bearing positional encoding. An Acoustic-n-Point (AnP) model is then used to recover the target 6D pose. Finally, simulation and water-tank experiments on the socket target verify the feasibility and robustness of the proposed method under limited-data conditions. The method achieves a 3.1 cm mean translation error, a 10.88° mean orientation error, and 52 FPS in real underwater acoustic environments. Full article

Forward-looking sonar (FLS) target detection is essential for autonomous underwater vehicles (AUVs), yet its effectiveness is severely hindered by complex acoustic distortions, environmental volatility and the scarcity of fine-annotated data, which limit the success of standard deep learning approaches. To address these challenges, this study proposes a novel quasi-static in situ learning paradigm for underwater acoustic target detection (UATD). The hybrid methodology integrates scene priors into a lightweight deep learning detector by incorporating explicit probability weighting based on echo-intensity statistics and acoustic attenuation compensation. By using these models for pixel-wise image enhancement and fusing statistical descriptors with deep learning predictions at the score level, the framework dynamically adapts to in situ environmental contexts during quasi-static operational tranches. Experimental evaluations on the UATD dataset demonstrate that this in situ adaptation significantly enhances overall detection performance, achieving an F1-score of 0.865 for our approach, an 8.1% improvement over the baseline YOLOv12n, with only a 2.1 G increase in FLOPs, while outperforming YOLOv12x (F1 = 0.844) with 95% fewer FLOPs. Ultimately, this paradigm overcomes the limitations of purely deep learning-based methods, offering a robust and interpretable solution tailored for practical AUV deployment. Full article

Quantifying the detection efficiency of buoy-based sonar and optimizing deployment strategies in complex marine environments remain significant challenges. This study proposes a transceiver depth optimization method based on the Bellhop ray model to enhance underwater remote sensing data quality. For the first time, we validated the applicability of acoustic reciprocity in deep-sea environments exceeding 5000 m, characterized by non-uniform sound speed profiles, horizontal inhomogeneity, and steep seamount terrain, with a maximum relative error of <1.2%. This extends the applicable boundaries of the acoustic reciprocity theorem from idealized simple waveguides to complex, realistic deep-sea environments. Building on this validation, we developed a novel, equivalent, superposition modeling framework for bidirectional transmission loss (TL), which converts the computationally intractable TL from target to receiver into the calculable TL from receiver to target, thus significantly reducing computational complexity. Systematic simulations uncovered a depth-layered dependency mechanism: shallow sources (23.14~69.42 m) and deep sources (≥347.10 m) show robustness to large depth differences exceeding 500 m, whereas mid-layer sources (161.98~231.40 m) exhibit a distinct critical threshold effect. Static simulations identify a performance degradation cliff with an onset at an approximate depth difference of 185 m, leading to a 50% reduction in detection range and fragmented near-field detection coverage. To accommodate environmental temporal variability (e.g., internal waves), a conservative safety margin was incorporated, establishing a robust engineering threshold of 150 m. Accordingly, we define 160~350 m as the optimal detection depth window and propose a layered deployment protocol that fills a critical industry gap in quantitative deployment design for deep-sea acoustic detection. Specifically, transceiver depth differences should be strictly constrained to <150 m for mid-layer operations, while more-flexible depth configurations are permissible for shallow and deep sources. These findings furnish quantitative engineering criteria for the design of reliable underwater remote sensing networks, while balancing long-range detection stability and near-field coverage integrity. Full article

Cross-domain semantic classification of 3D point clouds remains challenging due to strong domain shifts between heterogeneous sensing modalities. Most existing classification frameworks are domain-specific, limiting their use in integrated land–water mapping applications. This study evaluates the transferability of classical geometric machine learning classifiers between terrestrial and underwater point cloud domains without target-domain retraining. Experiments were conducted using terrestrial data acquired with a Leica BLK360 terrestrial laser scanner (TLS) and underwater point clouds collected with a Blueview BV5000 mechanical scanning sonar (MSS). Two dimensionality-based frameworks, CANUPO–Support Vector Machine (SVM) and 3DMASC–Random Forest (RF), were implemented in CloudCompare and assessed under intra-domain and cross-domain configurations. Strong intra-domain performance was achieved, with terrestrial–terrestrial accuracies of 0.99 for CANUPO–SVM and 0.97 for 3DMASC. In underwater evaluation, CANUPO maintained high accuracy (0.97), whereas 3DMASC decreased to 0.86 due to increased variability in the submerged data. Under cross-domain transfer, CANUPO achieved 0.93 accuracy for terrestrial-to-underwater and 0.89 for underwater-to-terrestrial classification, while 3DMASC demonstrated stable generalisation with 0.95 accuracy in both directions. Overall, dimensionality-based geometric descriptors capture stable structural cues across sensing environments, providing an interpretable and efficient pathway for applications such as hydrographic surveying, coastal monitoring, and underwater search-and-rescue detection. Future work will extend validation to larger datasets and explore domain adaptation strategies to further reduce cross-modality domain shift. Full article

High-precision underwater perception and scene reconstruction are critical techniques for marine surveying and resource exploration. Multi-sensor data fusion is currently the dominant method in underwater sensing. In this paper, a new approach for underwater sensing based on an integration of a 3D rotating scanning imaging sonar, an RTK (Real-Time Kinematic), and an IMU (Inertial Measurement Unit) systems onboard an unmanned surface vehicle (USV) is raised. By employing multi-sensor data fusion and image correlation calibration, combined with multi-view acoustic image synthesis, the system achieves accurate reconstruction of both water column and seabed scenes. The new system offers high reconstruction accuracy, and provides a cost-effective solution for scene reconstruction with a low requirement of the precise motion control of the USV platform. High-precision seabed imaging results have been validated through lake bed imaging tests. Full article

Active sonar tracking of multiple underwater targets is frequently challenged by intense reverberation, which leads to sonar returns that are both non-stationary and non-Gaussian. In such scenarios, the generalized labeled multi-Bernoulli (GLMB) filter, which relies on a Gaussian assumption, often experiences a rise in an Optimal Subpattern Assignment (OSPA) distance, along with recurrent label switching. To mitigate this problem, a robust delta-generalized labeled multi-Bernoulli technique (ST-δ-GLMB) is introduced; it characterizes noise using a Student’s t-distribution and employs variational Bayes to estimate the corresponding parameters. More precisely, the Student’s t-distribution is utilized to represent measurement non-stationarity, and an online variational Bayesian estimation of the noise parameters is conducted within a multi-target framework based on the Student’s t-model. Moreover, without altering the GLMB data-association and label-management machinery, we derive closed-form updates and propagation for the Student’s t-parameters, thereby keeping the recursive computational burden and practical implementability under control. Finally, Monte Carlo simulations and lake-trial data demonstrate that, under non-stationary and heavy-clutter conditions, ST-δ-GLMB maintains stable track continuity and accurate target-number (cardinality) estimates in the presence of non-stationary measurements. Full article

This study focuses on how mesoscale eddies at the Kuroshio boundary in the East China Sea modulate underwater acoustic propagation. Using high-resolution reanalysis data from the Hybrid Coordinate Ocean Model (HYCOM) and validated acoustic ray-tracing simulations, the OW + SLA method is employed for eddy identification and classification. Statistical analysis of 120 eddy events from 2015 to 2020 clarifies their seasonal variation characteristics. Warm eddies shift the convergence zone 15–30 km away from the sound source and broaden it by 20–40%, while cold eddies shift it 10–25 km toward the source and narrow it by 15–35%. A linear relationship exists between eddy amplitude and acoustic transmission loss (TL = 72.4 + 0.42 h, R² = 0.61), where TL is the transmission loss in decibels (dB) and h is the eddy amplitude in meters (m), and there are depth-dependent transmission loss modulation effects. These results provide practical guidance not only for sonar system design and acoustic communication optimization but also for error correction in underwater acoustic navigation systems operating in eddy-prone environments. Full article

To scientifically assess the fish resource status and spatial distribution in the Huoyanshan waters of Poyang Lake for the conservation of endangered species like Coilia nasus, an acoustic survey was conducted using a dual-frequency identification sonar (DIDSON) in July 2024. Fish targets were identified and extracted by combining an Echoview-based identification and deep learning models. Catch statistics were integrated to estimate fish density, abundance, biomass, and spatial distribution patterns. A total of 1891 fish targets were detected. The Echoview model achieved an average accuracy of 90.83%, while the YOLO model attained average precision and recall of 0.941 and 0.869, and the DeepSORT model attained precision and recall of 0.887 and 0.911. The total fish abundance was estimated at approximately 223,775 individuals, with a total biomass of about 199,742 kg. Spatially, fish were predominantly distributed in nearshore areas horizontally and concentrated at depths of 5–15 m vertically. The integrated approach combining DIDSON, Echoview and deep learning models proved effective for high-accuracy fish target identification and resource estimation, with deep learning models offering greater objectivity and processing efficiency. This study provides a technical reference for intelligent fish target identification in sonar images and provides baseline data and a technical reference for subsequent fish resource monitoring and management in the Huoyanshan waters of Poyang Lake. Full article

Time-series machine learning models represented by long short-term memory (LSTM) networks provide an effective way to obtain high-precision sound speed profiles (SSPs) quickly and at low cost, which can meet the practical application requirements of underwater sonar systems. However, in sea areas with frequent strong internal solitary waves, the large-amplitude sound speed anomalies caused by them will seriously interfere with model learning in the form of strong outlier features, resulting in a sharp drop in SSP prediction accuracy and significant degradation of the generalization stability and robustness of the model. To address this problem, this paper proposes a time-series SSP prediction method based on a bidirectional long short-term memory (Bi-LSTM) network. First, Empirical Orthogonal Function (EOF) decomposition is used to realize the low-dimensional feature representation of SSPs, and then the bidirectional time-series feature capture capability of Bi-LSTM is used to predict the SSP sequence with large disturbances caused by strong internal solitary waves. Multiple groups of comparative experiments based on the measured temperature chain data in the continental slope area of the South China Sea show that the Bi-LSTM model has a significant improvement in prediction accuracy and robustness compared with the classical LSTM model. Among them, the Bi-LSTM model with EOF decomposition achieves a correlation coefficient of 0.995 and an average Root Mean Square Error (RMSE) as low as 0.387 m/s. Under the condition of internal solitary wave disturbance, the classical LSTM is difficult to effectively capture the large abrupt change in sound speed, while the proposed Bi-LSTM model can still achieve accurate prediction of the SSP in the disturbance section, and has both the feature recognition and evolution prediction capabilities for the strongly nonlinear internal solitary wave process. This method provides effective technical support for the rapid and large-scale reconstruction of the sound speed field under the disturbance of strong internal solitary waves. Full article