Side-scan Sonar Research Articles

Unveiling Underwater Structures: MobileNet vs. EfficientNet in Sonar Image Detection

Deep learning techniques have led to an increased use of Convolutional Neural Networks (CNN) in recognizing images for marine surveys and classifying underwater objects. Applying CNN for the automatic identification of targets in side-scan sonar (SSS) images can boost accuracy and efficiency. Using transfer learning and deep learning as lenses, this work explores the field of underwater item detection. Comparing performance of MobileNet and EfficientNet in performing the task of underwater item classification from sonar images is the central research objective of this study. These two well-known convolutional neural networks were the subject of a thorough investigation to help explain this. In order to fine-tune pre-trained models on a dataset of sonar images, the study approach used the transfer learning technique. We conducted tests on MobileNet and EfficientNet and thoroughly assessed each of their accuracy levels. This study has two main conclusions. First, showing promise as a strong model for underwater object identification, EfficientNet performed exceptionally well in the classification challenge.

RPFNet: Recurrent Pyramid Frequency Feature Fusion Network for Instance Segmentation in Side-Scan Sonar Images

Side-scan sonar (SSS) is an essential acoustic sensor device for obtaining underwater information. The instance segmentation of sonar images can effectively locate and detect underwater objects. Although various CNN-based frameworks have achieved promising results in natural image instance segmentation, the noise interference, highlight shadow, and blurred edge in sonar images bring more significant challenges for sonar image instance segmentation. To solve these problems, we propose a novel recurrent pyramid frequency feature fusion network (RPFNet), which mainly consists of the pyramid frequency feature fusion network (PFN), recurrent residual attention mechanism (RRAM), mask prediction module, and semantic segmentation module. By enhancing and fusing different frequency features of SSS images, PFN can effectively extract fine-grained features and reduce background information interference. The RRAM uses residual structure and attention mechanism to enhance the correlation of different frequency features and improve nonlinear feature representation ability. The mask prediction and semantic segmentation modules are used to discriminate object instance categories and generate corresponding semantic segmentation results. We extensively evaluate the proposed framework on the real-scenes sonar image dataset. The evaluation results demonstrate that the proposed method outperforms the state-of-the-art methods in various evaluation metrics. Code is available at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/darkseid-arch/SonarRPFNet</uri> .

Side-Scan Sonar Image Classification With Zero-Shot and Style Transfer

Side-Scan Sonar Image Classification With Zero-Shot and Style Transfer

A Level Set Method With Heterogeneity Filter for Side-Scan Sonar Image Segmentation

A Level Set Method With Heterogeneity Filter for Side-Scan Sonar Image Segmentation

Unsupervised Terrain Reconstruction From Side-Scan Sonar Constrained to the Imaging Mechanism

Unsupervised Terrain Reconstruction From Side-Scan Sonar Constrained to the Imaging Mechanism

Bedforms and sediment dynamics under reduced sediment input and estuarine engineering: the case of the Bou Regreg estuary, Morocco

The Bou Regreg riverine estuary constitutes a microtidal environment located within an urban region and characterized by limited sediment inflow from its catch basin hinterlands. The construction of the Sidi Mohamed Ben Abdellah (SMBA) dam has notably diminished sediment supply, especially coarser fractions. In light of these conditions, this study aimed to classify sediment types according to the Folk and Ward classification, stratify morphometric parameters and temporal fluctuations based on dune types, and ultimately evaluate the sediment dynamics driving observed bedform variations. Using an integrated approach, data collection involved various techniques, including CTD profiling, bottom videography, side-scan sonar imaging, sub-bottom profiling, and Van Veen bottom grab sampling. The bed morphology includes segments of smooth bed, erosional and dredged bottom, localized scour features, and submarine dune fields, displaying notable heterogeneity. Acoustic seabed reflections indicated two bedform clusters with maximum lengths of 42.6 m and heights of 1 m. Flow velocities were linked to the presence of 2D and 3D dune forms. Morphological variations encompassed straight, sinuous, and barchan dune configurations, mainly asymmetric with NNW–SSE seaward orientation. Sediment mean size fluctuations spanned from muddy to medium sand deposition. The findings revealed the intricate interplay of dynamics shaping dune field configurations and the heterogeneous dispersion of bedforms throughout the estuary likely due to a sharp decline of sediment discharge caused by the dam upstream and coastal engineering. The study emphasizes the dynamic nature of dunes, highlighting their sensitivity to shifting hydrodynamic conditions and sediment availability, leading to morphological fluctuations over time.

RepDNet: A re-parameterization despeckling network for autonomous underwater side-scan sonar imaging with prior-knowledge customized convolution

Side-scan sonar (SSS) is now a prevalent instrument for large-scale seafloor topography measurements, deployable on an autonomous underwater vehicle (AUV) to execute fully automated underwater acoustic scanning imaging along a predetermined trajectory. However, SSS images often suffer from speckle noise caused by mutual interference between echoes, and limited AUV computational resources further hinder noise suppression. Existing approaches for SSS image processing and speckle noise reduction rely heavily on complex network structures and fail to combine the benefits of deep learning and domain knowledge. To address the problem, RepDNet, a novel and effective despeckling convolutional neural network is proposed. RepDNet introduces two re-parameterized blocks: the Pixel Smoothing Block (PSB) and Edge Enhancement Block (EEB), preserving edge information while attenuating speckle noise. During training, PSB and EEB manifest as double-layered multi-branch structures, integrating first-order and second-order derivatives and smoothing functions. During inference, the branches are re-parameterized into a 3 × 3 convolution, enabling efficient inference without sacrificing accuracy. RepDNet comprises three computational operations: 3 × 3 convolution, element-wise summation and Rectified Linear Unit activation. Evaluations on benchmark datasets, a real SSS dataset and Data collected at Lake Mulan aestablish RepDNet as a well-balanced network, meeting the AUV computational constraints in terms of performance and latency.

Interpreting and processing side-scan sonar data with the objective of further automation

One of the most effective tools of remote sensing and visualization of underwater surfaces and objects are acoustic devices, in particular side-scan sonars (SSS). Recently, largely due to the emergence of affordable devices, the geography and scope of application of this device has been significantly expanded. Meanwhile, despite certain progress achieved in terms of improving and minimizing the SSS hardware, the software used remains, in general, at a basic level, providing the operator mainly with a simple tool for visualizing benthic environments and data recording for further post-processing. Existing experience in SSS exploitation reveals that the key issue of interpreting acoustic images lies in the physical essence of their acquisition. Arguably, attempts to implement methods of automated interpretation of optical images have no perspective. Hence, the objective of this paper is to provide a theoretical and practical background of SSS data interpretation and processing. A layout for the potential automation of this process is provided with the objective of eliminating human operation in the process of conducting survey and search operations. The authors consider the conditions of SSS exploitation including special attention to such issues as the vastness of search areas, which, as argued is the key problem of data and data pattern recognition. SSS image recognition is an issue relevant for a wide range of academic topics such as geometric distortion, image recognition, and navigation satellite system target localization.

Posidonia oceanica Cartography and Evolution of the Balearic Sea (Western Mediterranean)

The Balearic coastline presents an environmental and biological heterogeneity, which confers great complexity on the marine environment and treasures important biodiversity, both at the level of species and marine communities. The endemic phanerogam of the Mediterranean Sea, Posidonia oceanica, holds a pivotal role in maintaining high biodiversity, warranting protection as stipulated in the Posidonia Decree 25/2018. The purpose of this study is to provide quantitative criteria that will allow the delimitation of areas with Posidonia oceanica for conservation and to aid planning and management of this species, contributing to the reduction of biodiversity loss caused by anthropogenic impacts and global change. Utilizing a comprehensive approach, the study employs photo interpretation of aerial photographs taken at depths between 0 and 5 m, data from Side-Scan Sonar (SSS) campaigns, reprocessing information from the LIFE Posidonia project at depths between 20 and 30 m, and targeted sampling using Remote Operated Vehicles (ROV) and ocular recognition at strategic points. The research not only seeks to assess the present state of the phanerogam but also analyzes its evolution, establishing a technological database for consultation and integrated analysis. This database facilitates effective management by tracking habitat changes, representing a significant contribution to the understanding of the impact of global change on ecosystems through Geographical Information Technologies (TIGs).

Using side-scan sonar and sediment sampling to identify concentrated areas of coal ash within a recreational lake

Backstrom JT, Burdette KM, Connell PM. 2024. Using side-scan sonar and sediment sampling to identify concentrated areas of coal ash within a recreational lake. Lake Reserv Manage. 40:36–42. Coal ash is a globally recognized legacy pollutant associated with coal-powered electricity generating stations, containing numerous heavy metals. These include arsenic, mercury, selenium, and cadmium, which are known to cause significant impacts on aqueous and benthic environments. In this pilot study, side-scan sonar surveys, coupled with sediment sampling and chemical analysis, were used to map and identify areas of contaminated coal ash in and around Sutton Lake, a recreational waterbody located adjacent to the Sutton coal-fired power plant, near Wilmington, North Carolina. Preliminary results are promising with this innovative and cost-effective survey method of mapping and locating legacy coal pollution, which may ultimately assist with monitoring, management, and remediation of coal ash waste locally and nationally.

The Impact of Side-Scan Sonar Resolution and Acoustic Shadow Phenomenon on the Quality of Sonar Imagery and Data Interpretation Capabilities

Side-scan sonar is designed and used for a variety of survey work, in both military and civilian fields. These systems provide acoustic imageries that play a significant role in a variety of marine and inland applications. For this reason, it is extremely important that the recorded sonar image is characterized by high resolution, detail and sharpness. This article is mainly aimed at the demonstration of the impact of side-scan sonar resolution on the imaging quality. The article also presents the importance of acoustic shadow in the process of analyzing sonar data and identifying underwater objects. The real measurements were carried out using two independent survey systems: hull-mounted sonar and towed side-scan sonar. Six different shipwrecks lying in the Baltic Sea were selected as the objects of research. The results presented in the article also constitute evidence of how the sonar technology has changed over time. The survey findings show that by maintaining the appropriate operational conditions and meeting several requirements, it is possible to obtain photographic-quality sonar images, which may be crucial in the process of data interpretation and shipwreck identification.

A Partially Buried Shipwreck Site: Implication from the Laoniu Reef Shipwreck No. 2

Underwater cultural relics are considered to be a scarce and valuable cultural resource. The significance of partially buried underwater cultural relics, which represents a distinct category of underwater cultural relics, has been largely overlooked in previous research. The Laoniu reef shipwreck No. 2, as an example of a partially buried shipwreck, was discovered through the utilization of side-scan sonar, multibeam echosounder and diver verification. In the investigation, multiangle surveys provided evidence to effectively discover this unique site. Beam-focusing technology and high-frequency side-scan sonar with proper distance and angular parameters can help to verify this site further. Given the potential significance of this particular site in relation to environmental transformations and human interventions, it is imperative to consider the insights derived from this case when reevaluating the methodology employed in underwater cultural relic surveys conducted in prospective regions. Consequently, this investigation holds the potential to unveil novel underwater cultural relic sites, thereby presenting a promising avenue for future discoveries.

Improving Yolo5 for Real-Time Detection of Small Targets in Side Scan Sonar Images

Improving Yolo5 for Real-Time Detection of Small Targets in Side Scan Sonar Images

Assessment and Characterization of Woji Creek Riverbed in Port Harcourt, Rivers State, Nigeria, Utilizing Side Scan Sonar Technology

Woji Creek plays a crucial role in vessel navigation, supports diverse flora and fauna habitats, and sustains local livelihoods. However, human activities and natural events have led to changes in its riverbed over time. Given its significance for coastal stability and geomorphology, there is a pressing need to classify, evaluate, and ensure its environmental sustainability. The primary objective of this study is to employ side scan sonar technology to classify and assess the riverbed of Woji Creek in Port Harcourt, Nigeria. The study aims to classify the riverbed, analyze water depth variations, evaluate navigational suitability, and determine turbidity levels within the creek. To achieve these objectives, the methodology involved acquiring Side-Scan Sonar (SSS) and Sub-bottom profile data. These datasets underwent backscatter processing to create geocoded backscatter images. Feature points were extracted and matched from these images to derive riverbed classification, depth categories, water volume distribution, and river turbidity analysis. The riverbed classification revealed the presence of three primary sediment types: Clayey Silty Sand, Silty Clay, and Silty Sand, each with distinct implications for navigation. Shallow, Moderate, Deep, and Very Deep areas were identified within Woji Creek, each influencing navigational conditions. Additionally, the water volume distribution analysis provided essential insights into depth limitations and route planning.&#x0D; Moreover, the assessment of river turbidity identified low, moderate, and high turbidity zones, reflecting water clarity and suspended particle levels. These findings serve as invaluable decision support tools for navigation planning and management in Woji Creek, offering comprehensive insights into the riverbed, depth suitability, volume distribution, and water quality. Leveraging this data can enhance strategic decision-making processes and contribute to the sustainable management of this vital waterway.

Long-term morphological and sedimentological changes caused by bottom trawling on the northern Catalan continental shelf (NW Mediterranean)

Bottom trawling is a fishing method that involves towing of nets along the seafloor to catch demersal species. The dragging of trawling gears along the seafloor results in scraping and ploughing of the surficial sediments, leading to the formation of turbid plumes of resuspended sediments and causing measurable changes in the seabed morphology. High-resolution multibeam data, side scan sonar, sediment grain size and vessel tracking data have been used to investigate the impact of bottom trawling on the seafloor morphology and surficial sediments of the northern Catalan continental shelf (NW Mediterranean), providing new insights into the impact of this anthropogenic activity on the seafloor. Multibeam data evidenced the occurrence of large-scale erosive features as a consequence of repeated scouring by fishing gears in localized areas. They are characterized by elongated (70-300 m wide and up to 8 km long) channelized areas of high backscatter with variable incision (from 0.2 m to 1.2 m). The spatial distribution of these morphologies shows a similar pattern to that observed in the fishing intensity, with maximum values in the areas of increased trawling intensity, corresponding to the main fishing grounds. Side scan sonar data also shows higher densities of trawl marks in these areas than in the surroundings. Sediment cores collected on these features show an upward-coarsening trend in the first 4-5 cm of the core, suggesting that part of the finer fraction resuspended by trawling is winnowed, increasing the sand content of the surface sediment. The identification of such large erosive morphologies in the main fishing grounds evidences that repeated trawling over the same fishing ground during decades can result in deep excavation of the seafloor, leading to permanent large-scale morphological changes. Furthermore, the evolution of these erosive features over a 13-year interval points towards long recovery periods after the cessation of trawling activities.

Detection of Small Objects in Side-Scan Sonar Images Using an Enhanced YOLOv7-Based Approach

Deep learning-based object detection methods have demonstrated remarkable effectiveness across various domains. Recently, there has been growing interest in applying these techniques to underwater environments. Conventional optical imaging methods face severe limitations when operating in underwater conditions, restricting their ability to identify objects with good visibility and at close distances. Consequently, side-scan sonar (SSS) has emerged as a common equipment choice for underwater detection due to its compatibility with the characteristics of sound waves in water. This paper introduces a novel method, termed the Enhanced YOLOv7-Based Approach, for detecting small objects in SSS images. Building upon the widely-adopted YOLOv7 method, the proposed approach incorporates several enhancements aimed at improving detection accuracy. First, a dedicated detection layer tailored for small objects is added to the original network architecture. Additionally, two attention mechanisms are integrated within the backbone and neck components of the network, respectively, to strengthen the network’s focus on object features. Finally, the network features are recombined based on the BiFPN structure. Experimental results demonstrate that the proposed method outperforms mainstream object detection algorithms. In comparison to the original YOLOv7 network, it achieves a Precision of 95.5%, indicating a significant improvement of 4.8%. Moreover, its Recall reaches 87.0%, representing an enhancement of 5.1%, while the mean Average Precision (mAP) at an IoU threshold of 0.5 (mAP@.5) reaches 86.9%, reflecting a 6.7% improvement. Furthermore, the mAP@.5:.95 reaches 55.1%, a 4.8% enhancement. Therefore, the method presented in this paper enhances the performance of YOLOv7 for object detection in SSS images, providing a fresh perspective on small object detection based on SSS images and contributing to the advancement of underwater detection techniques.

Technical Design and Data Analysis of Autonomous Underwater Vehicle-Based Side-Scan Sonar Operations

In response to the absence of standardized work practices, work safety measures, efficient work procedures, and suitable line planning methods for exploring seabed topography using autonomous underwater vehicles (AUVs) equipped with side-scan sonar systems, this paper proposes a design framework for side-scan sonar measurement technology based on AUVs. Detailed research has been conducted on line planning methods that differ significantly from those used in shipborne operations. The new line planning method was developed based on the performance characteristics of AUV equipment, the principles of motion modes, and shipborne side-scan sonar operations. The validity of the line planning method was verified through actual sea experiments using different types of AUVs in various terrains. The experimental results demonstrate that the new survey line planning method is distinct from existing AUV path planning methods and is highly applicable for conducting seabed topography exploration operations. Furthermore, the technical design scheme studied in this article is applicable in different sea areas and ensures the safety, operational efficiency, and data quality of AUVs. It holds significant importance for marine science engineering and practical applications. The article also features an analysis of successful data collected from AUVs equipped with side-scan sonar for conducting seabed topography exploration operations.

Large gravitational collapse structure on a rocky coast (Kvarner, NE Adriatic Sea)

The studied rock collapse structure is located on the Liburnian coast (Rijeka Bay, channel zone of the NE Adriatic). The relief of the southern part of this coast, with a length of 6.5 km, is a large escarpment with very steep to vertical slopes reaching heights of 100 m above sea level, as a result of tectonic movements along the Kvarner fault zone. These events probably led to a sudden relaxation of the highly fractured rock mass. The progressive expansion occurred at locations where previously favourably oriented faults and fissures had formed a polygonal rock collapse resembling a rock-slide which is the focus of this study. Another aim of this study is to reconstruct and explain the complex morphological evolution of the studied landslide, from the pre-failure deformations, through the failure itself, to post-failure displacements, as well as possible future instabilities. Recent techniques to survey the instability, location and to analyse the evolution of the rupture surface and its dimensions were combined (Unmanned Aerial Vehicle, Side Scan Sonar and Remotely Operated Vehicles). The estimated total volume of displaced rock mass is 950,000 m3. The lower part of the instability phenomenon was submerged during the Holocene sea level rise. Since then, a large part of the displaced rock mass has been in a stable position, with sporadic rock falls. However, given unfavourable orientation and discontinuity characteristics, as well as unfavourable environmental influences, possible instabilities might also be expected in the future.

Seabed characterization of the Trindade island shelf, Brazil

This paper presents the first mapping of bedform morphological features on the Trindade Island shelf, with the goal of identifying and defining them.By using a side scan sonar, collecting seabed samples, and taking submarine photographs, five different morphological types were identified on this shelf: an erosive type: (i) characterized by the dispersion of rocks and rock blocks; and four depositional types: (ii) one formed by calcareous algal reefs, (iii) one composed by rhodoliths, (iv) one formed by medium sands with large symmetrical wave ripples, (v) and the last one formed by fine sands with asymmetrical wave ripples. The main result is a morphological map of the Trindade Island shelf, which shows the occurrence of these five seabed units.To further understand the sedimentological factors that affect the island shelf, we also examined the sediments that were collected for this study. Our results show that the amount of carbonate-formed grains on the collected sand sediments is higher than the amount of grains formed by the volcanoclastic ones, generated by the disaggregation of the main building on Trindade Island. The sediments formed by bioclastic grains are more frequently found in the coarser fractions than in the finer fractions when compared to lithic ones.

An Image Quality Improvement Method in Side-Scan Sonar Based on Deconvolution

Side-scan sonar (SSS) is an important underwater imaging method that has high resolutions and is convenient to use. However, due to the restriction of conventional pulse compression technology, the side-scan sonar beam sidelobe in the range direction is relatively high, which affects the definition and contrast of images. When working in a shallow-water environment, image quality is especially influenced by strong bottom reverberation or other targets on the seabed. To solve this problem, a method for image-quality improvement based on deconvolution is proposed herein. In this method, to increase the range resolution and lower the sidelobe, a deconvolution algorithm is employed to improve the conventional pulse compression. In our simulation, the tolerance of the algorithm to different signal-to-noise ratios (SNRs) and the resolution ability of multi-target conditions were analyzed. Furthermore, the proposed method was applied to actual underwater data. The experimental results showed that the quality of underwater acoustic imaging could be effectively improved. The ratios of improvement for the SNR and contrast ratio (CR) were 32 and 12.5%, respectively. The target segmentation results based on this method are also shown. The accuracy of segmentation was effectively improved.