Underwater Target Research Articles

A planning method for chemical plume tracking and source localization with autonomous underwater vehicle

Underwater target detection technology is an important mean for the development and utilization of marine environment, which has important value in commercial and military fields, and is valued by more and more experts, scholars, and production technicians. A set of behavior-based autonomous underwater vehicle online task planning algorithm is proposed in this article, including the searching behavior based on changing Z-shape, tracking behavior based on an improved artificial potential field theory with likelihood map, rediscovering behavior, and source declare behavior. The proposed algorithm can improve the accuracy and success rate of autonomous underwater vehicle to the source positionficient. The physical experiment results verify the effectiveness of the proposed method.

Study of the acoustic scattering characteristics of a rigid sphere in a vortex acoustic field

This paper investigates the scattering sound field of a rigid sphere positioned in a vortex beam generated by a phase-modulated circular transducer array. The effects of the dimensionless parameter ka, the number of array elements, the topological charge, the radius of the transducer array, the distance between the array and the sphere, and its offset position on the scattering sound field distribution of a sphere are studied. Simulation results show that as the ka increases, the incident wave intensity and the scattering sound field of a rigid sphere increase, and the angle between the two main lobes decreases. Once ka is determined, the number of the array elements has no effect on the scattering pattern. However, the topological charge of the vortex beam has greater effect on the scattering sound field. As the topological charge increases, the incident wave intensity decreases, but the angle between two main lobes of the scattering field increases. The radius of the array has an effect on the intensity of the scattering field and the larger the radius, the larger the area of low intensity radiation between the two main lobes. In addition, the distance between the sphere and the array has a weak effect on the scattering pattern when it exceeds a certain value. For an off-axis sphere, the scattering pattern becomes more complex as the topological charge changes. These results can be used to adjust the constituent factors of the transducer array and to investigate the advantages of the vortex beam for underwater target detection.

IFE-net: improved feature enhancement network for weak feature target recognition in autonomous underwater vehicles

AbstractThe recognizing underwater targets is a crucial component of autonomous underwater vehicle patrols and detection efforts. In the process of visual image recognition in real underwater environment, the spatial and semantic features of the target often appear to different degrees of loss, and the scarcity of specific types of underwater samples leads to unbalanced data on categories. This kind of problem makes the target features appear weak and seriously affects the accuracy of underwater target recognition. Traditional deep learning methods based on data and feature enhancement cannot achieve ideal recognition effect. Based on the above difficulties, this paper proposes an improved feature enhancement network for weak feature target recognition. Firstly, a multi-scale spatial and semantic feature enhancement module is constructed to extract the feature information of the extraction target accurately. Secondly, this paper solves the influence of target feature distortion on classification through multi-scale feature comparison of positive and negative samples. Finally, the Rank & Sort Loss function was used to train the depth target detection to solve the problem of recognition accuracy under highly unbalanced sample data. Experimental results show that the recognition accuracy of the proposed method is 2.28% and 3.84% higher than that of the existing algorithms in the recognition of underwater fuzzy and distorted target images, which demonstrates the effectiveness and superiority of the proposed method.

Unraveling complex data diversity in underwater acoustic target recognition through convolution-based mixture of experts

Underwater acoustic target recognition is a difficult task owing to the intricate nature of underwater acoustic signals. The complex underwater environments, unpredictable transmission channels, and dynamic motion states greatly impact the real-world underwater acoustic signals, and may even obscure the intrinsic characteristics related to targets. Consequently, the data distribution of underwater acoustic signals exhibits high intra-class diversity, thereby compromising the accuracy and robustness of recognition systems. To address these issues, this work proposes a convolution-based mixture of experts (CMoE) that recognizes underwater targets in a fine-grained manner. The proposed technique introduces multiple expert layers as independent learners, along with a routing layer that determines the assignment of experts according to the characteristics of inputs. This design allows the model to utilize independent parameter spaces, facilitating the learning of complex underwater signals with high intra-class diversity. Furthermore, this work optimizes the CMoE structure by balancing regularization and an optional residual module. To validate the efficacy of our proposed techniques, we conducted detailed experiments and visualization analyses on three underwater acoustic databases across several acoustic features. The experimental results demonstrate that our CMoE consistently achieves significant performance improvements, delivering superior recognition accuracy when compared to existing advanced methods.

Research on a Feature Enhancement Extraction Method for Underwater Targets Based on Deep Autoencoder Networks

The low-frequency line spectrum of the radiated noise signals of hydroacoustic targets contains features describing the intrinsic properties of the target that make the target susceptible to exposure. In order to extract the line spectral features of underwater acoustic targets, a method combining image processing and a deep autoencoder network (DAE) is proposed in this paper to enhance the low-frequency weak line spectrum of underwater targets in an extremely low signal-to-noise ratio environment based on the measured data of large underwater vehicles. A Gauss–Bernoulli restricted Boltzmann machine (G–BRBM) for real-value signal processing was designed and programmed by introducing a greedy algorithm. On this basis, the encoding and decoding mechanism of the DAE network was used to eliminate interference from environmental noise. The weak line spectrum features were effectively enhanced and extracted under an extremely low signal-to-noise ratio of 10–300 Hz, after which the reconstruction results of the line spectrum features were obtained. Data from large underwater vehicles detected by far-field sonar arrays were processed and the results show that the method proposed in this paper was able to adaptively enhance the line spectrum in a data-driven manner. The DAE method was able to achieve more than double the extractable line spectral density in the frequency band of 10–300 Hz. Compared with the traditional feature enhancement extraction method, the DAE method has certain advantages for the extraction of weak line spectra.

Acoustic Scattering Characteristics and Geometric Parameter Prediction for Underwater Multiple Targets Arranged in a Linear Pattern

The construction of wind farm pilings, submarine pipelines, and underwater submarines involves multiple cylinders. However, there is currently a lack of economic research on predicting the mechanism and characteristics of mutual coupling of acoustic scattering from multiple cylindrical targets. This study investigates the mechanism and prediction method of acoustic scattering for the structural distribution characteristics of underwater multi-cylindrical targets. A model of a multi-cylindrical target’s two-dimensional acoustic field was established using the finite element method. Numerical calculations were then carried out to elucidate the scattering characteristics of the frequency–angle spectrum in far-field omnidirectional scattering. The simulation of echoes in the time domain explains how echoes propagate and interact with each other, and provides formulas for calculating interference and resonance frequencies. The frequency calculation formula extracts key features from the spectrum, providing a basis for predicting the characteristics of multi-cylindrical targets in terms of scale and spatial position. Measurement experiments were conducted on a double-cylindrical target in a water tank, and the theoretical calculations and experimental data were used to estimate the target’s radius and distance. The actual layout confirms the accuracy of the interference and resonance frequency prediction formulas. This study offers a valuable solution for refined feature extraction and spatial estimation of underwater targets.

Research on multibeam line layout based on geometric mathematical model and regression analysis

A multibeam sounding system is a system that emits acoustic pulse waves through a mechanical device to map the seafloor topography in real time. It can accurately and rapidly measure the size, shape and elevation changes of underwater targets. In this study, the mathematical model of the multibeam sounding system is proposed, and the optimized design of the survey line is carried out according to different seabed geomorphology. Firstly, relevant formulas are deduced by mathematical geometry and trigonometric functions in plane space, then a spatial right-angled coordinate system is established and the direction of the survey line is projected to the x-axis, which transforms the survey line problem in three-dimensional space into a planar type, and finally, according to the mathematical definitions, the shape of the multibeam irradiation on the sloping surface is judged to be an ellipse and the projection model is constructed. The bivariate model is constructed using regression analysis, and the line measurement that meets the requirements is determined by solving the optimal solution. The conclusion shows that the optimal solution for the angle between the direction of the survey line and the projection of the normal direction of the seabed slope on the horizontal plane is 45°, and the survey line that meets the requirements of the shortest measurement length, covering the entire sea area to be surveyed, and the overlap rate of 10-20% can be obtained.

Detecting Weak Underwater Targets Using Block Updating of Sparse and Structured Channel Impulse Responses

In this paper, we considered the real-time modeling of an underwater channel impulse response (CIR), exploiting the inherent structure and sparsity of such channels. Building on the recent development in the modeling of acoustic channels using a Kronecker structure, we approximated the CIR using a structured and sparse model, allowing for a computationally efficient sparse block-updating algorithm, which can track the time-varying CIR even in low signal-to-noise ratio (SNR) scenarios. The algorithm employs a conjugate gradient formulation, which enables a gradual refinement if the SNR is sufficiently high to allow for this. This was performed by gradually relaxing the assumed Kronecker structure, as well as the sparsity assumptions, if possible. The estimated CIR was further used to form a residual signal containing (primarily) information of the time-varying signal responses, thereby allowing for the detection of weak target signals. The proposed method was evaluated using both simulated and measured underwater signals, clearly illustrating the better performance of the proposed method.

Zero-Shot Learning-Based Recognition of Highlight Images of Echoes of Active Sonar

Reducing the impact of underwater disturbance targets and improving the ability to recognize real moving targets underwater are important directions of active sonar research. In this paper, the highlight model of underwater targets was improved and a method was proposed to acquire highlight images of the echoes of these targets. A classification convolutional neural network called HasNet-5 was designed to extract the global features and local highlight features of the echo highlight images of underwater targets, which achieved the true/false recognition of targets via multi-classification. Five types of target highlight models were used to generate simulation data to complete the training, validation and testing of the network. Tests were performed using experimental data. The results indicate that the proposed method achieves 92% accuracy in real target recognition and 94% accuracy in two-dimensional disturbance target recognition. This study provides a new approach for underwater target recognition using active sonar.

Underwater target correlation detection and location method based on normalized magnetic moment

For the dynamic magnetic detection platform represented by airborne magnetic anomaly detection, the magnetic field measured by the vector magnetometer is easily affected by the rapid attitude change. Therefore, traditional anti-submarine aircraft are equipped with scalar magnetometers that are less affected by changes in the carrier attitude. However, the scalar magnetometer can only measure the total magnetic field signal, and its measurement information is single. It contains a time-varying environmental magnetic field signal, including the geomagnetic field, which is difficult to effectively reject. This brings great difficulties to the single real-time location of a magnetic anomaly target. This paper proposes a target correlation detection method based on the normalized magnetic moment that can effectively solve the above problems, considering the rotational invariance of magnetic anomaly measurements for dynamic magnetic exploration platforms and the efficient detection methods required for the single real-time positioning of targets. Correlation detection based on the normalized magnetic moment characteristic function and characteristic time search based on the correlation detection are included. Simulation and experimental results verify the single-point real-time positioning ability of the improved detection method within an effective operating distance. This method has strong usability for actual magnetic anomaly detection of underwater targets.

Station layout optimization for underwater acoustic positioning system based on combined cone configuration

The theoretical minimum of the geometric dilution of precision (GDOP) for target positioning cannot be achieved because of the coplanar constraint for underwater station configurations, and the conventional single-cone configuration leads to an infinite GDOP. Therefore, this study extends the single-cone configuration to the combined cone configuration. The proposed configuration is theoretically proven to be superimposable, rotatable, and GDOP-rotation-invariant. A criterion based on the global GDOP for underwater targets is constructed to address the problem of a single criterion for station layout optimization. However, traditional optimization algorithms are prone to local optimality and low convergence efficiency. Therefore, an improved adaptive particle swarm optimization (APSO) algorithm is proposed to optimize the configuration of the underwater acoustic positioning systems. Finally, it is verified that the proposed APSO can enhance the global optimality and improve the convergence efficiency of the algorithm. Combined with the measured data of the long baseline system, it is shown that the optimal combined cone configuration significantly reduces the global GDOP and improves the positioning accuracy compared with traditional configurations.

Robust Positioning Estimation for Underwater Acoustics Targets with Use of Multi-Particle Swarm Optimization

With the extensive application of sensor technology in scientific ocean research, ocean resource exploration, underwater engineering construction, and other fields, underwater target positioning technology has become an important support for the ocean field. This paper proposes a robust positioning algorithm that combines the disadvantages of distributed estimation and particle swarm optimization, which can solve the large localization error problem caused by uncertainties in underwater acoustic communication and sampling processes. Considering the presence of ranging anomalies and sampling packet loss in underwater acoustic measurements, a weighted coupling filling method is used to correct the outliers in an underwater acoustic ranging signal. Based on the mapping model from the element array to the underwater acoustic responder, an unconstrained optimization algorithm for one-time localization estimation of underwater acoustic targets was established. Based on the one-time localization estimation results of underwater acoustic targets, an improved multi-particle swarm optimization estimation based on interactive search is proposed, which improves the accuracy of underwater target localization. The numerical results show that the positioning accuracy of the proposed algorithm can be effectively enhanced in cases of distance measurement errors and azimuth measurement errors. Compared with the positioning error before optimization, the positioning error can be reduced after optimization. Additionally, the experiment was carried out in the underwater environment of Hangzhou Qiandao Lake, which verified the positioning performance of the proposed algorithm.

Measurement of backscattering strength of artificial bubbles in the Southern Sea of the Korean Peninsula

Bubbles greatly affect the propagation of sound waves as the acoustic characteristics of a bubble cluster in water are distinct from those of undisturbed water. Therefore, bubble clusters affect the ability of sensors to detect underwater targets by blocking or reflecting sound waves. Additionally, the bubble wakes produced by ships in movement can be detected by wake-homing torpedoes, thus greatly threatening the safety of the ship. Thus, research on bubble dynamics in water is crucial for the development of military technology. Recently, we conducted a quantitative estimation of the acoustic characteristics of artificial bubbles, including their backscattering strength, existence time, population density spectrum level and void fraction in an ideal water-tank environment. Based on our previous findings, the present study sought to measure the acoustic characteristics of artificial bubbles using an acoustic Doppler current profiler (ADCP) in the southern sea of the Korean Peninsula, which is a real marine environment. Additionally, we validated the ADCP measurements by comparing them to those obtained using a scattering strength measurement system (SSMS) developed by our team. Collectively, our findings provide a basis for the development of military technology, as well as for the study of bubble in water.

PE-Transformer: Path enhanced transformer for improving underwater object detection

Traditional object detection methods cannot effectively identify underwater objects with complex backgrounds, and it is difficult to fully obtain the details of small-scale underwater targets, resulting in poor detection performance. We propose a path-augmented Transformer detection framework to address these limitations to explore the semantic details of small-scale underwater targets in complex environments. On one hand, an embedded local path detection information scheme is devised to facilitate the interaction between high-level and low-level features, thereby enhancing the semantic representation of distinctive features of small-scale underwater targets. Rich dependency relationships are established between the acquired high-level and low-level features within the CSWin-Transformer framework, thus fortifying the semantic representation during the encoding phase. Furthermore, a individualized loss function is employed to optimize and fine-tune features at various hierarchical levels. On the other hand, a detection module with flexible and adaptive point representation that is different from conventional square detection methods is designed. This module covers the underwater target from any direction, and the salient point samples in classification localization and feature selection between points realize the feature selection improves the detection accuracy of underwater objects simultaneously. We designed a new weighted loss function to encourage the network to converge better. Experimental results on open-source underwater and remote sensing images of UTDAC, RUOD, and ADios show that the proposed method outperforms other underwater object detection methods in terms of precision(P), recall(R), comprehensive evaluation index of F1-score and FPS(frames per second).

Underwater acoustic target recognition based on knowledge distillation under working conditions mismatching

Underwater acoustic target recognition based on knowledge distillation under working conditions mismatching

Deep learning-driven underwater polarimetric target detection based on the dispersion of polarization characteristics

Underwater target detection especially in turbidity environments is a crucial and challenging area of study due to its extensive applications. In this work, we propose an underwater target detection method that combines the dispersion of polarization characteristics with a neural network framework. The polarization dispersion distribution features are effective at reducing the influence of the scattering effect of particulate matter in a turbidity environment, and highlighting the polarization state variations, especially in the edge contour features between the target and background. According to the physical detection model of dispersion of polarization characteristics, the neural network is constructed to extract dispersion values of the angle of polarization (AOP) and degree of linear polarization (DOLP) from the underwater polarization image datasets with different turbidity levels, which we then used to train the network to map physical characteristic parameters of the target for the detection. When compared to target detection methods based on light intensity images and polarization characteristic images (AOP and DOLP), the experimental results indicate that the proposed network model with polarization dispersion images has a significant improvement in locating and identifying targets of different materials, particularly in high turbidity underwater environments. These results provide deeper understanding of polarization polarimetric target detection, and further improve the functionality of a polarimetric target detection optical system.

Tracking-by-Detection Algorithm for Underwater Target Based on Improved Multi-Kernel Correlation Filter

Tracking-by-Detection Algorithm for Underwater Target Based on Improved Multi-Kernel Correlation Filter

Semi-analytical solution for sound propagation from a moving directional source in a shallow-water waveguide

Understanding the waveguide fields produced by moving directional sources is of crucial importance in underwater acoustic detection. In practice, moving underwater targets often exhibit directionality, especially at high frequencies. Herein, we present a semi-analytical method to calculate the Doppler-shifted field (DSF) from a directional source moving horizontally in a range-independent, shallow-water waveguide. We develop a novel normal-mode model using a modal-projection approach with a perfectly matched layer. This comprehensively accounts for changes in modal shape, eigenvalue shifts, and the branch-integral contribution. We derive an analytical expression for the DSF based on Huygens’ principle, and the solution is presented as a sum over propagating eigenmodes, with the modal excitation determined by the source directivity depending upon the grazing angle of each pair of modal plane waves. This expression is valid for any type of radiator moving in a range-independent environment with a smoothly varying sound-speed profile. We then present a theoretical analysis of the Doppler beam shift produced by a piston-like radiator. This reveals that the observed angular offset of modal plane waves can be attributed to the Doppler beam shift and confirms that this is a new mechanism causing considerable differences between the fields generated by a directional moving source and by the same source at rest. This has potential value in various applications, particularly for underwater acoustic detection of moving targets.

A Lightweight Network Based on Multi-Scale Asymmetric Convolutional Neural Networks with Attention Mechanism for Ship-Radiated Noise Classification

Ship-radiated noise classification is critical in ocean acoustics. Recently, the feature extraction method combined with time–frequency spectrograms and convolutional neural networks (CNNs) has effectively described the differences between various underwater targets. However, many existing CNNs are challenging to apply to embedded devices because of their high computational costs. This paper introduces a lightweight network based on multi-scale asymmetric CNNs with an attention mechanism (MA-CNN-A) for ship-radiated noise classification. Specifically, according to the multi-resolution analysis relying on the relationship between multi-scale convolution kernels and feature maps, MA-CNN-A can autonomously extract more fine-grained multi-scale features from the time–frequency domain. Meanwhile, the MA-CNN-A maintains its light weight by employing asymmetric convolutions to balance accuracy and efficiency. The number of parameters introduced by the attention mechanism only accounts for 0.02‰ of the model parameters. Experiments on the DeepShip dataset demonstrate that the MA-CNN-A outperforms some state-of-the-art networks with a recognition accuracy of 98.2% and significantly decreases the parameters. Compared with the CNN based on three-scale square convolutions, our method has a 68.1% reduction in parameters with improved recognition accuracy. The results of ablation explorations prove that the improvements benefit from asymmetric convolution, multi-scale block, and attention mechanism. Additionally, MA-CNN-A shows a robust performance against various interferences.

Underwater acoustic target recognition method based on WA-DS decision fusion

Deep learning methods can recognize the various underwater acoustic targets in complex marine environments. In this study, we apply deep learning methods in decision recognition of underwater acoustic targets. However, decision recognition methods rely on a complex evidence theory that combines the probabilities of various scenarios as evidence sources, and in the process of combination produces conflicting evidence. For this reason, we propose an underwater acoustic target recognition method based on weighted average Dempster -Shafer (WA-DS) decision fusion, which replaces anomalous evidence with weighted average evidence, and solves the problem of synthesizing anomalous evidence. Firstly, the underwater acoustic dataset is preprocessed with downsampling, denoising, and audio segmentation. Then, the Mel-frequency cepstral coefficients (MFCC), Gammatone frequency cepstral coefficients (GFCC), and time–frequency (TF) features are extracted, and the category probabilities are output using the ResNet18 network to learn and train the multi-feature and then output the category probabilities. Finally, the category probabilities of multi-feature are computed as evidence using the Basic Probability Assignment (BPA) function, and the various types of evidence are fused using our decision fusion algorithm. Experimental results show that our decision fusion method is more capable of synthesizing anomalous evidence than traditional decision fusion methods. Our work achieves recognition accuracy of up to 98.34% on the ShipsEar dataset, which is a significant improvement in recognition rate compared to single features.