Pub Date : 2025-03-09DOI: 10.1109/JOE.2025.3551018
Iain McLeod;Brendan Walsh;Thomas Kelly;John V. Ringwood
There is a common perception when monitoring ocean waves, that data buoys, measuring parameters, such as significant wave height and dominant period, must have the characteristics of true wave followers, where the movement of the device is assumed to follow the free surface of the ocean. This presents an obstacle to using wave energy to power data buoys, as wave energy converters necessarily interact with passing waves to harness their energy. This study proposes a Kalman filter-based unknown input estimator to be used as a soft sensor to process readings from an existing motion sensor mounted a data buoy, taking into account the effects of an internal moonpool acting as an oscillating water column (OWC), including tests with an orifice plate to simulate a turbine power take-off (PTO). The estimator described in this article is tested against wave tank data in both regular and irregular waves, for a fully sealed moonpool, acting as a linear system. This article also describes how the Kalman filter can be extended to handle the nonlinearities introduced by fitting an orifice plate simulating an OWC turbine PTO, and tests this against regular wave data. The proposed sensor is found to accurately return values for significant wave height and zero-crossing period, as well as time series estimates of the free surface elevation, at 0.1 s time steps, for both linear and nonlinear system representations.
{"title":"Free Surface Elevation Estimator as a Sensor for Wave-Powered Data Buoys","authors":"Iain McLeod;Brendan Walsh;Thomas Kelly;John V. Ringwood","doi":"10.1109/JOE.2025.3551018","DOIUrl":"https://doi.org/10.1109/JOE.2025.3551018","url":null,"abstract":"There is a common perception when monitoring ocean waves, that data buoys, measuring parameters, such as significant wave height and dominant period, must have the characteristics of true wave followers, where the movement of the device is assumed to follow the free surface of the ocean. This presents an obstacle to using wave energy to power data buoys, as wave energy converters necessarily interact with passing waves to harness their energy. This study proposes a Kalman filter-based unknown input estimator to be used as a soft sensor to process readings from an existing motion sensor mounted a data buoy, taking into account the effects of an internal moonpool acting as an oscillating water column (OWC), including tests with an orifice plate to simulate a turbine power take-off (PTO). The estimator described in this article is tested against wave tank data in both regular and irregular waves, for a fully sealed moonpool, acting as a linear system. This article also describes how the Kalman filter can be extended to handle the nonlinearities introduced by fitting an orifice plate simulating an OWC turbine PTO, and tests this against regular wave data. The proposed sensor is found to accurately return values for significant wave height and zero-crossing period, as well as time series estimates of the free surface elevation, at 0.1 s time steps, for both linear and nonlinear system representations.","PeriodicalId":13191,"journal":{"name":"IEEE Journal of Oceanic Engineering","volume":"50 3","pages":"2232-2247"},"PeriodicalIF":3.8,"publicationDate":"2025-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10994677","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144646055","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-09DOI: 10.1109/JOE.2025.3556153
Weilin Luo;Chengyu Lin;Huan Zhou
The performance of underwater target detection techniques is limited by various factors. First, in underwater image data sets, there exists degradation such as color bias, low contrast, and blurring, which affect the accuracy of the detection algorithms. Second, the underwater image data set is difficult to obtain and the cost of making the labeled data sets is high, which also prevents underwater object detection algorithms from fully leveraging their potential. In this article, we propose a self-supervised learning network for underwater target detection. Considering the deficiency that the conventional contrastive learning network pays more attention to the global information and ignores the local information, an auxiliary branch inspired by masked autoencoders is added to the baseline SimSiam network, which collaborates with the main branch to optimize the target network and help the target network learn the local information of the target feature map. A residual spatial cooperative attention module is proposed to be embedded within the proposed self-supervised learning network to obtain remote information through residual structure and construct spatial context features. The method of cooperative attention is used to enhance feature learning ability. Experiments are carried out on a reconstructed underwater target data set. Results show that compared with the baseline network, the method proposed in this article is more suitable for underwater environments and has better mean average precision.
{"title":"Underwater Target Detection by Residual Spatial Cooperative Attention Module–Based Self-Supervised Learning","authors":"Weilin Luo;Chengyu Lin;Huan Zhou","doi":"10.1109/JOE.2025.3556153","DOIUrl":"https://doi.org/10.1109/JOE.2025.3556153","url":null,"abstract":"The performance of underwater target detection techniques is limited by various factors. First, in underwater image data sets, there exists degradation such as color bias, low contrast, and blurring, which affect the accuracy of the detection algorithms. Second, the underwater image data set is difficult to obtain and the cost of making the labeled data sets is high, which also prevents underwater object detection algorithms from fully leveraging their potential. In this article, we propose a self-supervised learning network for underwater target detection. Considering the deficiency that the conventional contrastive learning network pays more attention to the global information and ignores the local information, an auxiliary branch inspired by masked autoencoders is added to the baseline SimSiam network, which collaborates with the main branch to optimize the target network and help the target network learn the local information of the target feature map. A residual spatial cooperative attention module is proposed to be embedded within the proposed self-supervised learning network to obtain remote information through residual structure and construct spatial context features. The method of cooperative attention is used to enhance feature learning ability. Experiments are carried out on a reconstructed underwater target data set. Results show that compared with the baseline network, the method proposed in this article is more suitable for underwater environments and has better mean average precision.","PeriodicalId":13191,"journal":{"name":"IEEE Journal of Oceanic Engineering","volume":"50 3","pages":"1930-1943"},"PeriodicalIF":3.8,"publicationDate":"2025-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144646676","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-09DOI: 10.1109/JOE.2025.3536578
Blake Cole;Peter Traykovski
In this article, we describe a deterministic optimization framework for the conceptual-stage design of aerodynamically-actuated wingsails. The primary objective of high-performance sailing is well understood: maximize the conversion of unsteady aerodynamic forces into forward thrust, without inducing excessive overturning moments. However, designing a sail to meet this goal is by no means straightforward due to the existence of multiple recursive, nonlinear design relationships. Consequently, most wingsails are designed in an iterative fashion, using some combination of linear heuristics and engineering intuition. This approach can produce viable designs, but it does so at the expense of time and capital, and provides little physical insight into the underlying design space. By formulating the wingsail design problem as a geometric program, it is possible to quickly generate hundreds of optimal candidate designs, assess their sensitivity to specific constraints and parameters, and determine the shape of Pareto frontiers. Unlike general nonlinear optimization methods, geometric programming optimization is computationally efficient, and requires no initial guesses or hyperparameter tuning. Perhaps most importantly, all the decision variables in a geometric program are determined simultaneously, eliminating the need for iterative piecewise optimization of subsystems. These benefits come at a price: all objective and constraint functions must be described as posynomials. Nevertheless, we demonstrate that this restricted set of functional forms can adequately capture the key physical relationships between wingsail parameters, and provide quantifiable physics-based design guidance.
{"title":"Geometric Programming for Aerodynamically-Actuated Wingsail Design Optimization","authors":"Blake Cole;Peter Traykovski","doi":"10.1109/JOE.2025.3536578","DOIUrl":"https://doi.org/10.1109/JOE.2025.3536578","url":null,"abstract":"In this article, we describe a deterministic optimization framework for the conceptual-stage design of aerodynamically-actuated wingsails. The primary objective of high-performance sailing is well understood: maximize the conversion of unsteady aerodynamic forces into forward thrust, without inducing excessive overturning moments. However, designing a sail to meet this goal is by no means straightforward due to the existence of multiple recursive, nonlinear design relationships. Consequently, most wingsails are designed in an iterative fashion, using some combination of linear heuristics and engineering intuition. This approach can produce viable designs, but it does so at the expense of time and capital, and provides little physical insight into the underlying design space. By formulating the wingsail design problem as a geometric program, it is possible to quickly generate hundreds of optimal candidate designs, assess their sensitivity to specific constraints and parameters, and determine the shape of Pareto frontiers. Unlike general nonlinear optimization methods, geometric programming optimization is computationally efficient, and requires no initial guesses or hyperparameter tuning. Perhaps most importantly, all the decision variables in a geometric program are determined simultaneously, eliminating the need for iterative piecewise optimization of subsystems. These benefits come at a price: all objective and constraint functions must be described as posynomials. Nevertheless, we demonstrate that this restricted set of functional forms can adequately capture the key physical relationships between wingsail parameters, and provide quantifiable physics-based design guidance.","PeriodicalId":13191,"journal":{"name":"IEEE Journal of Oceanic Engineering","volume":"50 3","pages":"2063-2089"},"PeriodicalIF":3.8,"publicationDate":"2025-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144646419","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-09DOI: 10.1109/JOE.2025.3556152
Joanna Rzempołuch;Kevin Goddard;Sunny Chaudhary;George Callender;Robert G. Olsen;Justin Dix;Paul Lewin;David Renew
With the raised investment in interconnectors and offshore wind farms located further from the shore, there has been an increased need for the licensing and environmental assessment of high voltage direct current (HVDC) submarine cables. The motionally induced electric fields due to HVDC submarine cables have not been previously modeled in the literature. In this article, a methodology for modeling this phenomenon is outlined. Factors, such as geographical location of the cable system, electrical conductivities of the media, reference frames, and water velocity profile, are included in the analysis. The results are compared with an approximation adopted in other publications. The resulting electric fields and their spatial distributions are presented and discussed for different cable systems, cable orientations and reference frames. The dependency of the local electric fields on the geographical location is shown. The investigation demonstrates the sensitivity of electric fields to variations in seabed and water conductivities. Specifically, the most extreme combination yielded an 85$%$ increase compared to the case with equal conductivities, which emphasises the substantial impact of media conductivities on local electric field prediction. The methodology outlined in this paper can provide a basis for future empirical validation, and inform biological experiments and licensing processes.
{"title":"Electric Fields Induced by Water Movement in Proximity to HVDC Submarine Cables","authors":"Joanna Rzempołuch;Kevin Goddard;Sunny Chaudhary;George Callender;Robert G. Olsen;Justin Dix;Paul Lewin;David Renew","doi":"10.1109/JOE.2025.3556152","DOIUrl":"https://doi.org/10.1109/JOE.2025.3556152","url":null,"abstract":"With the raised investment in interconnectors and offshore wind farms located further from the shore, there has been an increased need for the licensing and environmental assessment of high voltage direct current (HVDC) submarine cables. The motionally induced electric fields due to HVDC submarine cables have not been previously modeled in the literature. In this article, a methodology for modeling this phenomenon is outlined. Factors, such as geographical location of the cable system, electrical conductivities of the media, reference frames, and water velocity profile, are included in the analysis. The results are compared with an approximation adopted in other publications. The resulting electric fields and their spatial distributions are presented and discussed for different cable systems, cable orientations and reference frames. The dependency of the local electric fields on the geographical location is shown. The investigation demonstrates the sensitivity of electric fields to variations in seabed and water conductivities. Specifically, the most extreme combination yielded an 85<inline-formula><tex-math>$%$</tex-math></inline-formula> increase compared to the case with equal conductivities, which emphasises the substantial impact of media conductivities on local electric field prediction. The methodology outlined in this paper can provide a basis for future empirical validation, and inform biological experiments and licensing processes.","PeriodicalId":13191,"journal":{"name":"IEEE Journal of Oceanic Engineering","volume":"50 3","pages":"2369-2380"},"PeriodicalIF":3.8,"publicationDate":"2025-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144646067","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-08DOI: 10.1109/JOE.2025.3550984
Shaobo Li;Tie Li;Aiguo Sun;Shiqi Wang;Yunlong Wu
The 3-D sub-bottom profiler (SBP) is widely used for observing sub-bottom structures due to its high resolution and spatial coverage. However, traditional picking methods are limited by scattering noise and imbalanced intensity, resulting in poor picking of reflections with weak intensities and overpicking due to scattering noise. This limitation has hindered the development of 3-D SBP processing technologies relative to the widespread applications. In this article, we present a 3-D SBP horizon-picking method that takes into account both amplitude and phase information. First, we apply an amplitude data enhancement filtering algorithm considering the plate-like structure and the nonvertical characteristic to avoid scattering noise interference and highlight plate-like horizon features. Subsequently, a threshold-based algorithm is applied to pick the amplitude horizons, reflecting the main structures of layer interfaces based on the processed amplitude data. Then, the local phase information of the SBP data is derived through the monogenic analysis, and phase horizons are picked, which can describe detailed reflections with weak intensities, although overpicking may occur. To leverage both amplitude and phase horizons, a combination method is proposed to ensure continuous and fine horizon picking containing both main and detailed 3-D SBP horizon structures. The method was validated by comparing automated horizon results with manual results using experimental data. A total of 87% and 86% F-measures were achieved on two data sets, respectively.
{"title":"Automated Amplitude and Phase Attribute-Based Horizon Picking Applied to 3-D Sub-bottom Data","authors":"Shaobo Li;Tie Li;Aiguo Sun;Shiqi Wang;Yunlong Wu","doi":"10.1109/JOE.2025.3550984","DOIUrl":"https://doi.org/10.1109/JOE.2025.3550984","url":null,"abstract":"The 3-D sub-bottom profiler (SBP) is widely used for observing sub-bottom structures due to its high resolution and spatial coverage. However, traditional picking methods are limited by scattering noise and imbalanced intensity, resulting in poor picking of reflections with weak intensities and overpicking due to scattering noise. This limitation has hindered the development of 3-D SBP processing technologies relative to the widespread applications. In this article, we present a 3-D SBP horizon-picking method that takes into account both amplitude and phase information. First, we apply an amplitude data enhancement filtering algorithm considering the plate-like structure and the nonvertical characteristic to avoid scattering noise interference and highlight plate-like horizon features. Subsequently, a threshold-based algorithm is applied to pick the amplitude horizons, reflecting the main structures of layer interfaces based on the processed amplitude data. Then, the local phase information of the SBP data is derived through the monogenic analysis, and phase horizons are picked, which can describe detailed reflections with weak intensities, although overpicking may occur. To leverage both amplitude and phase horizons, a combination method is proposed to ensure continuous and fine horizon picking containing both main and detailed 3-D SBP horizon structures. The method was validated by comparing automated horizon results with manual results using experimental data. A total of 87% and 86% F-measures were achieved on two data sets, respectively.","PeriodicalId":13191,"journal":{"name":"IEEE Journal of Oceanic Engineering","volume":"50 3","pages":"2355-2368"},"PeriodicalIF":3.8,"publicationDate":"2025-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144646510","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-08DOI: 10.1109/JOE.2025.3553982
Jongmin Ahn;Geun-Ho Park;Wanjin Kim;Hyung-Moon Kim;Dong-Hun Lee
Underwater biomimetic communication (UBC) technology has been studied to overcome the restricted physical covertness of artificial acoustical communication signals. Recent advancements have demonstrated data transmission at 300 b/s with imperceptible mimicry levels and communication distances of up to 60 km. While these results have raised expectations for practical applications, the actual performance and maturity of this technology have remained unclear. This article investigates and evaluates existing biomimetic communication methods to clarify the current level of UBC technology to plan a development strategy. To analyze their performance, we classified them based on sound types, such as whistles and clicks, and modulation techniques. Simultaneously, the technological maturity of these methods is also assessed using the technical readiness level framework. The results show that biomimetic communication could be a more promising solution for military underwater communication requiring covertness. Finally, we suggest research directions to further develop this technology into a fully operational system.
{"title":"Analysis of Research Trends and Technological Maturity of Biomimetic Underwater Communication","authors":"Jongmin Ahn;Geun-Ho Park;Wanjin Kim;Hyung-Moon Kim;Dong-Hun Lee","doi":"10.1109/JOE.2025.3553982","DOIUrl":"https://doi.org/10.1109/JOE.2025.3553982","url":null,"abstract":"Underwater biomimetic communication (UBC) technology has been studied to overcome the restricted physical covertness of artificial acoustical communication signals. Recent advancements have demonstrated data transmission at 300 b/s with imperceptible mimicry levels and communication distances of up to 60 km. While these results have raised expectations for practical applications, the actual performance and maturity of this technology have remained unclear. This article investigates and evaluates existing biomimetic communication methods to clarify the current level of UBC technology to plan a development strategy. To analyze their performance, we classified them based on sound types, such as whistles and clicks, and modulation techniques. Simultaneously, the technological maturity of these methods is also assessed using the technical readiness level framework. The results show that biomimetic communication could be a more promising solution for military underwater communication requiring covertness. Finally, we suggest research directions to further develop this technology into a fully operational system.","PeriodicalId":13191,"journal":{"name":"IEEE Journal of Oceanic Engineering","volume":"50 3","pages":"1676-1702"},"PeriodicalIF":3.8,"publicationDate":"2025-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10993376","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144646578","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-08DOI: 10.1109/JOE.2025.3545239
Yichen Duan;Xiaohong Shen;Haiyan Wang;Yongsheng Yan
The recognition of ship-radiated noise signals is currently the crucial means of perceiving ships. Numerous methods for the recognition of ship-radiated noise signals have been developed based on deep learning techniques. Prior studies on ship-radiated noise signal recognition have assumed a single-target scenario. In this article, we establish a multilabel recognition scenario for ship-radiated noise signals. We propose a multilabel recognition framework for ship-radiated noise signals with multidomain information fusion. Initially, we adopt two fundamental backbone network structures to extract preliminary features from both time-domain and time–frequency domain signal data. Subsequently, we construct a feature enhancement and fusion module based on the theory of deep balanced networks. This module enables information interaction from low-level to high-level between the time-domain and time-frequency domain information of ship-radiated noise signals. We introduce a transformer-based feature enhancement approach and a gated fusion feature update network structure. We also design a fusion strategy for secondary fusion updates and interdomain representations to obtain stable enhanced fusion feature representations. Finally, a linear classifier is employed to determine the categories of the mixed signals. We simulate multilabel data for ship-radiated noise signals using the publicly available Deepship data set. Experimental results demonstrate satisfactory recognition performance.
{"title":"Multilabel Recognition Method for Ship-Radiated Noise Signals Based on Multidomain Information Fusion With Deep Equilibrium Models","authors":"Yichen Duan;Xiaohong Shen;Haiyan Wang;Yongsheng Yan","doi":"10.1109/JOE.2025.3545239","DOIUrl":"https://doi.org/10.1109/JOE.2025.3545239","url":null,"abstract":"The recognition of ship-radiated noise signals is currently the crucial means of perceiving ships. Numerous methods for the recognition of ship-radiated noise signals have been developed based on deep learning techniques. Prior studies on ship-radiated noise signal recognition have assumed a single-target scenario. In this article, we establish a multilabel recognition scenario for ship-radiated noise signals. We propose a multilabel recognition framework for ship-radiated noise signals with multidomain information fusion. Initially, we adopt two fundamental backbone network structures to extract preliminary features from both time-domain and time–frequency domain signal data. Subsequently, we construct a feature enhancement and fusion module based on the theory of deep balanced networks. This module enables information interaction from low-level to high-level between the time-domain and time-frequency domain information of ship-radiated noise signals. We introduce a transformer-based feature enhancement approach and a gated fusion feature update network structure. We also design a fusion strategy for secondary fusion updates and interdomain representations to obtain stable enhanced fusion feature representations. Finally, a linear classifier is employed to determine the categories of the mixed signals. We simulate multilabel data for ship-radiated noise signals using the publicly available Deepship data set. Experimental results demonstrate satisfactory recognition performance.","PeriodicalId":13191,"journal":{"name":"IEEE Journal of Oceanic Engineering","volume":"50 3","pages":"1760-1771"},"PeriodicalIF":3.8,"publicationDate":"2025-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144646516","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-06DOI: 10.1109/JOE.2025.3529132
Haotian Zheng;Yushan Sun;Hao Xu;Liwen Zhang;Yatong Han;Shuguang Cui;Zhen Li
Forward-looking sonar is a critical tool for underwater target detection, and segmentation is an essential component of forward-looking sonar image processing. Accurate segmentation of sonar images is vital, but the complexity of the underwater environment introduces challenges, such as low resolution, significant noise, and blurred target edges. These factors make real-time, precise segmentation particularly difficult. To address these challenges, we propose a novel real-time segmentation network, the multilevel mixed feature fusion network (MLMFFNet), specifically designed for forward-looking sonar images. Our approach leverages a unique MFF module and a multiscale MFF module to extract both local and contextual information using deep convolutional networks, dilated convolutions, and partial convolution combinations for effective information integration. Additionally, we incorporate a context connection module to enhance feature fusion by utilizing high-level contextual information. To further improve accuracy, we introduce three weighted loss functions designed to address imbalanced sample distributions and blurred boundaries. Experimental evaluations on two distinct forward-looking sonar data sets demonstrate that MLMFFNet significantly outperforms many state-of-the-art general and sonar-specific semantic segmentation networks, delivering superior segmentation accuracy while maintaining real-time performance.
{"title":"MLMFFNet: Multilevel Mixed Feature Fusion Network for Real-Time Forward-Looking Sonar Image Segmentation","authors":"Haotian Zheng;Yushan Sun;Hao Xu;Liwen Zhang;Yatong Han;Shuguang Cui;Zhen Li","doi":"10.1109/JOE.2025.3529132","DOIUrl":"https://doi.org/10.1109/JOE.2025.3529132","url":null,"abstract":"Forward-looking sonar is a critical tool for underwater target detection, and segmentation is an essential component of forward-looking sonar image processing. Accurate segmentation of sonar images is vital, but the complexity of the underwater environment introduces challenges, such as low resolution, significant noise, and blurred target edges. These factors make real-time, precise segmentation particularly difficult. To address these challenges, we propose a novel real-time segmentation network, the multilevel mixed feature fusion network (MLMFFNet), specifically designed for forward-looking sonar images. Our approach leverages a unique MFF module and a multiscale MFF module to extract both local and contextual information using deep convolutional networks, dilated convolutions, and partial convolution combinations for effective information integration. Additionally, we incorporate a context connection module to enhance feature fusion by utilizing high-level contextual information. To further improve accuracy, we introduce three weighted loss functions designed to address imbalanced sample distributions and blurred boundaries. Experimental evaluations on two distinct forward-looking sonar data sets demonstrate that MLMFFNet significantly outperforms many state-of-the-art general and sonar-specific semantic segmentation networks, delivering superior segmentation accuracy while maintaining real-time performance.","PeriodicalId":13191,"journal":{"name":"IEEE Journal of Oceanic Engineering","volume":"50 2","pages":"1356-1369"},"PeriodicalIF":3.8,"publicationDate":"2025-03-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143852390","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-06DOI: 10.1109/JOE.2025.3532036
Xiaodong Cui;Zhuofan He;Yangtao Xue;Peican Zhu;Jing Han;Xuelong Li
The complex dynamics of the marine environment pose substantial challenges for underwater acoustic target recognition (UATR) systems, especially when there are limited training samples. However, existing image-based few-shot learning methods might not be applicable, mainly because they fail to capture the temporal and spectral features from acoustic targets and lack the competent domain adaptation ability due to the inefficient usage of base samples. In this article, we develop a novel Domain Adaptation-based Attentional Time–Frequency few-shot recognition method (DAATF) for underwater acoustic targets. The DAATF explicitly utilizes a self-attention-based feature extractor to capture the time–frequency structural dependencies and constructs an autoencoder-based domain adapter to improve the cross-domain knowledge transfer through reusing the base dataset. In addition, a knowledge distillation module is designed to enable the model to reserve the general feature extraction ability of the pretrained network to avoid overfitting. Extensive experiments are conducted to assess prediction accuracy, noise robustness, and cross-domain adaptation. The obtained results validate that the DAATF can achieve outstanding performance, demonstrating its great potential for practical UATR applications. Furthermore, we provide free and open access to the DanShip data set.
{"title":"Few-Shot Underwater Acoustic Target Recognition Using Domain Adaptation and Knowledge Distillation","authors":"Xiaodong Cui;Zhuofan He;Yangtao Xue;Peican Zhu;Jing Han;Xuelong Li","doi":"10.1109/JOE.2025.3532036","DOIUrl":"https://doi.org/10.1109/JOE.2025.3532036","url":null,"abstract":"The complex dynamics of the marine environment pose substantial challenges for underwater acoustic target recognition (UATR) systems, especially when there are limited training samples. However, existing image-based few-shot learning methods might not be applicable, mainly because they fail to capture the temporal and spectral features from acoustic targets and lack the competent domain adaptation ability due to the inefficient usage of base samples. In this article, we develop a novel Domain Adaptation-based Attentional Time–Frequency few-shot recognition method (DAATF) for underwater acoustic targets. The DAATF explicitly utilizes a self-attention-based feature extractor to capture the time–frequency structural dependencies and constructs an autoencoder-based domain adapter to improve the cross-domain knowledge transfer through reusing the base dataset. In addition, a knowledge distillation module is designed to enable the model to reserve the general feature extraction ability of the pretrained network to avoid overfitting. Extensive experiments are conducted to assess prediction accuracy, noise robustness, and cross-domain adaptation. The obtained results validate that the DAATF can achieve outstanding performance, demonstrating its great potential for practical UATR applications. Furthermore, we provide free and open access to the DanShip data set.","PeriodicalId":13191,"journal":{"name":"IEEE Journal of Oceanic Engineering","volume":"50 2","pages":"637-653"},"PeriodicalIF":3.8,"publicationDate":"2025-03-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143852405","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This article describes an affordable and setup-friendly cable-based localization technique for underwater remotely operated vehicles, which exploits the piecewise linear shape of the umbilical being equipped with a sliding ballast. Each stretched part of the cable is instrumented with a waterproof inertial measurement unit (IMU) to measure its orientation. Using the cable's geometry, the vehicle's location can be calculated in relation to the fixed or moving end of the cable. Experiments carried out with a robotic system in a water tank prove the reliability of this localization strategy. The study investigates the influence of measurement uncertainties on cable orientation and length, as well as the impact of the IMU location along the cable on localization precision. The accuracy of the localization method is discussed.
{"title":"ROV Localization Using Ballasted Umbilical Equipped With IMUs","authors":"Juliette Drupt;Christophe Viel;Claire Dune;Vincent Hugel","doi":"10.1109/JOE.2024.3467448","DOIUrl":"https://doi.org/10.1109/JOE.2024.3467448","url":null,"abstract":"This article describes an affordable and setup-friendly cable-based localization technique for underwater remotely operated vehicles, which exploits the piecewise linear shape of the umbilical being equipped with a sliding ballast. Each stretched part of the cable is instrumented with a waterproof inertial measurement unit (IMU) to measure its orientation. Using the cable's geometry, the vehicle's location can be calculated in relation to the fixed or moving end of the cable. Experiments carried out with a robotic system in a water tank prove the reliability of this localization strategy. The study investigates the influence of measurement uncertainties on cable orientation and length, as well as the impact of the IMU location along the cable on localization precision. The accuracy of the localization method is discussed.","PeriodicalId":13191,"journal":{"name":"IEEE Journal of Oceanic Engineering","volume":"50 2","pages":"1045-1064"},"PeriodicalIF":3.8,"publicationDate":"2025-03-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143848853","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：