Explore the vast range of titles available.

Accurate estimation of the frequency component is an important issue to identify and track marine objects (e.g., surface ship, submarine, etc.). In general, a passive sonar system consists of a sensor array, and each sensor receives data that have common information of the target signal. In this paper, we consider multiple-measurement sparse Bayesian learning (MM-SBL), which reconstructs sparse solutions in a linear system using Bayesian frameworks, to detect the common frequency components received by each sensor. In addition, the direction of arrival estimation was performed on each detected common frequency component using the MM-SBL based on beamforming. The azimuth for each common frequency component was confirmed in the frequency-azimuth plot, through which we identified the target. In addition, we perform target tracking using the target detection results along time, which are derived from the sum of the signal spectrum at the azimuth angle. The performance of the MM-SBL and the conventional target detection method based on energy detection were compared using in-situ data measured near the Korean peninsula, where MM-SBL displays superior detection performance and high-resolution results.

The utilization of multibeam sonar systems has significantly facilitated the acquisition of underwater bathymetric data. However, efficiently processing vast amounts of multibeam point cloud data remains a challenge, particularly in terms of rejecting massive outliers. This paper proposes a novel solution by implementing a cone model filtering method for multibeam bathymetric point cloud data filtering. Initially, statistical analysis is employed to remove large-scale outliers from the raw point cloud data in order to enhance its resistance to variance for subsequent processing. Subsequently, virtual grids and voxel down-sampling are introduced to determine the angles and vertices of the model within each grid. Finally, the point cloud data was inverted, and the custom parameters were redefined to facilitate bi-directional data filtering. Experimental results demonstrate that compared to the commonly used filtering method the proposed method in this paper effectively removes outliers while minimizing excessive filtering, with minimal differences in standard deviations from human-computer interactive filtering. Furthermore, it yields a 3.57% improvement in accuracy compared to the Combined Uncertainty and Bathymetry Estimator method. These findings suggest that the newly proposed method is comparatively more effective and stable, exhibiting great potential for mitigating excessive filtering in areas with complex terrain.

This study explores the integration of point cloud data obtained from unmanned aerial vehicles (UAVs) and unmanned surface vehicles (USVs) to address limitations in photogrammetry and to create comprehensive models of aquatic environments. The UAV platform (AUTEL EVO II) employs structure-from-motion (SfM) photogrammetry using optical imagery, while the USV (equipped with a NORBIT iWBMS multibeam sonar system) collects underwater bathymetric data. UAVs commonly face constraints in battery life and image-processing capacity, making it necessary to merge smaller UAV point clouds into larger, more complete models. The USV-derived bathymetric data are integrated with UAV-derived surface data to construct unified terrain models that include both above-water and underwater features. This study evaluates three coordinate transformation (CT) methods—4-parameter, 6-parameter, and 7-parameter—across three study areas in Taiwan to assess their effectiveness in registering USV–UAV and UAV–UAV point clouds. For USV–UAV integration, all CT methods improved alignment accuracy compared with results without CT, achieving decimeter-level precision. For UAV–UAV integrations, the 7-parameter method provided the best accuracy, especially in areas with low terrain roughness such as rooftops and pavements, while improvements were less pronounced in areas with high roughness such as tree canopies. These findings demonstrate that the 7-parameter CT method offers an effective and straightforward approach for accurate point cloud integration from different platforms and sensors.

Underwater acoustics has been implemented mostly in the field of sound navigation and ranging (SONAR) procedures for submarine communication, the examination of maritime assets and environment surveying, target and object recognition, and measurement and study of acoustic sources in the underwater atmosphere. With the rapid development in science and technology, the advancement in sonar systems has increased, resulting in a decrement in underwater casualties. The sonar signal processing and automatic target recognition using sonar signals or imagery is itself a challenging process. Meanwhile, highly advanced data-driven machine-learning and deep learning-based methods are being implemented for acquiring several types of information from underwater sound data. This paper reviews the recent sonar automatic target recognition, tracking, or detection works using deep learning algorithms. A thorough study of the available works is done, and the operating procedure, results, and other necessary details regarding the data acquisition process, the dataset used, and the information regarding hyper-parameters is presented in this article. This paper will be of great assistance for upcoming scholars to start their work on sonar automatic target recognition.

Forward-looking sonar is a technique widely used for underwater detection. However, most sonar images have underwater noise and low resolution due to their acoustic properties. In recent years, the semantic segmentation model U-Net has shown excellent segmentation performance, and it has great potential in forward-looking sonar image segmentation. However, forward-looking sonar images are affected by noise, which prevents the existing U-Net model from segmenting small objects effectively. Therefore, this study presents a forward-looking sonar semantic segmentation model called Feature Pyramid U-Net with Attention (FPUA). This model uses residual blocks to improve the training depth of the network. To improve the segmentation accuracy of the network for small objects, a feature pyramid module combined with an attention structure is introduced. This improves the model’s ability to learn deep semantic and shallow detail information. First, the proposed model is compared against other deep learning models and on two datasets, of which one was collected in a tank environment and the other was collected in a real marine environment. To further test the validity of the model, a real forward-looking sonar system was devised and employed in the lake trials. The results show that the proposed model performs better than the other models for small-object and few-sample classes and that it is competitive in semantic segmentation of forward-looking sonar images.

Side-scan sonar image (SSI) are often affected by a combination of multiplicative speckle noise and additive noise, which degrades image quality and hinders target recognition and scene interpretation. To address this problem, this paper proposes a denoising algorithm that integrates non-local similar block clustering with Bayesian sparse coding. The proposed method leverages cross-scale structural features and noise statistical properties of image patches, and employs a similarity metric based on the Equivalent Number of Looks (ENL) along with an improved K-means clustering algorithm to achieve accurate classification and enhance intra-class noise consistency. Subsequently, a joint training strategy is used to construct dictionaries for each cluster, and Bayesian Orthogonal Matching Pursuit (BOMP) is applied for sparse representation. This enables effective modeling and suppression of mixed noise while preserving structural details. Experimental results demonstrate that the proposed method outperforms several classical approaches in both objective metrics such as PSNR and SSIM, and in visual quality, particularly in preserving target edges and textures under severe noise conditions.

Synthetic aperture sonar image reconstruction relies on the coherence of overlapping phase centres to provide accurate micronavigation for a sensed scene. It is shown that phase centres lose coherence for near‐range scattering from large synthetic aperture sonar arrays due to the fundamentally bistatic nature of these sensors. This effect is modelled using the van Cittert‐Zernike theorem and a point‐based sonar scattering model. Reduction of the window length used in the delay estimation process can partially mitigate the loss of coherence at the expense of increased variance in the resulting delay estimates. Synthetic aperture sonar image reconstruction relies on the coherence of overlapping phase centres to provide accurate micronavigation for a sensed scene. It is shown that phase centres lose coherence for near‐range scattering from large synthetic aperture sonar arrays due to the fundamentally bistatic nature of these sensors.

Cuvier's beaked whales (Ziphius cavirostris) are known as extreme divers, though behavioral data from this difficult-to-study species have been limited. They are also the species most often stranded in association with Mid-Frequency Active (MFA) sonar use, a relationship that remains poorly understood. We used satellite-linked tags to record the diving behavior and locations of eight Ziphius off the Southern California coast for periods up to three months. The effort resulted in 3732 hr of dive data with associated regional movements--the first dataset of its kind for any beaked whale--and included dives to 2992 m depth and lasting 137.5 min, both new mammalian dive records. Deep dives had a group mean depth of 1401 m (s.d. = 137.8, n = 1142) and duration of 67.4 min (s.d. = 6.9). The group mean time between deep dives was 102.3 min (s.d. = 30.8, n = 783). While the previously described stereotypic pattern of deep and shallow dives was apparent, there was considerable inter- and intra-individual variability in most parameters. There was significant diel behavioral variation, including increased time near the surface and decreased shallow diving at night. However, maximum depth and the proportion of time spent on deep dives (presumed foraging), varied little from day to night. Surprisingly, tagged whales were present within an MFA sonar training range for 38% of days locations were received, and though comprehensive records of sonar use during tag deployments were not available, we discuss the effects frequent acoustic disturbance may have had on the observed behaviors. These data better characterize the true behavioral range of this species, and suggest caution should be exercised when drawing conclusions about behavior using short-term datasets.

Side-scan sonar imaging is essential for underwater target detection in marine exploration and engineering applications, yet small target detection faces significant challenges including limited frequency domain feature utilization, insufficient multi-scale feature fusion, and high computational complexity. This study develops Multi-Scale Spatial-Frequency Collaborative Detection Transformer (MSF-DETR), a novel end-to-end automatic detection algorithm specifically designed for small targets in side-scan sonar images. The method integrates three core innovations: a Multi-domain Adaptive Spatial-frequency Network (MASNet) backbone employing Cascaded dual-domain Mamba-enhanced Spatial-frequency Synergistic Convolution that simultaneously captures spatial geometric and frequency domain texture features; a Hierarchical Multi-scale Adaptive Feature Pyramid Network implementing intelligent weight allocation across different scales; and an Efficient Sparse Attention Transformer Encoder utilizing Window-based Adaptive Sparse Self-Attention mechanism that reduces computational complexity from quadratic to linear. Experimental validation was conducted on the self-built SSST-3K(Side-Scan Sonar Target Detection 3K Dataset) dataset containing approximately 3000 high-quality sonar images and the public KLSG dataset. Results demonstrate that MSF-DETR achieves 78.5% mAP50 and 38.5% mAP50-95 on the SSST-3K dataset, representing improvements of 2.8% and 3.3% respectively compared to baseline RT-DETR, while reducing computational complexity by 12.0% and achieving 71.2 FPS inference speed. The proposed MSF-DETR provides an effective solution for small target detection in complex marine environments, significantly advancing underwater sonar image processing technology.

Multi beam forward looking sonar plays an important role in underwater detection. However, due to the complex underwater environment, unclear features, and susceptibility to noise interference, most forward looking sonar systems have poor recognition performance. The research on MFLS for underwater target detection faces some challenges. Therefore, this study proposes innovative improvements to the YOLOv5 algorithm to address the above issues. On the basis of maintaining the original YOLOv5 architecture, this improved model introduces transfer learning technology to overcome the limitation of scarce sonar image data. At the same time, by incorporating the concept of coordinate convolution, the improved model can extract features with rich positional information, significantly enhancing the model’s detection ability for small underwater targets. Furthermore, in order to solve the problem of feature extraction in forward looking sonar images, this study integrates attention mechanisms. This mechanism expands the receptive field of the model and optimizes the feature learning process by highlighting key details while suppressing irrelevant information. These improvements not only enhance the recognition accuracy of the model for sonar images, but also enhance its applicability and generalization performance in different underwater environments. In response to the common problem of uneven training sample quality in forward looking sonar imaging technology, this study made a key improvement to the classic YOLOv5 algorithm. By adjusting the bounding box loss function of YOLOv5, the model’s over sensitivity to low-quality samples was reduced, thereby reducing the punishment on these samples. After a series of comparative experiments, the newly proposed CCW-YOLOv5 algorithm has achieved detection accuracy in object detection mAP@0.5 Reached 85.3%, and the fastest inference speed tested on the local machine was 54 FPS, showing significant improvement and performance improvement compared to existing advanced algorithms.