Peng Zhang

Automatic Target Recognition (ATR) in single-view sonar imagery is severely hampered by geometric distortions, acoustic shadows, and incomplete target information due to occlusions and the slant-range imaging geometry, which frequently give rise to misclassification and hinder practical underwater detection applications. To address these critical limitations, this paper proposes a Multi-View Deep Convolutional Neural Network (MVDCNN) based on feature-level fusion for robust sonar image target recognition. The MVDCNN adopts a highly modular and extensible architecture consisting of four interconnected modules: an input reshaping module that adapts multi-view images to match the input format of pre-trained backbone networks via dimension merging and channel replication; a shared-weight feature extraction module that leverages Convolutional Neural Network (CNN) or Transformer backbones (e.g., ResNet, Swin Transformer, Vision Transformer) to extract discriminative features from each view, ensuring parameter efficiency and cross-view feature consistency; a feature fusion module that aggregates complementary features (e.g., target texture and shape) across views using max-pooling to retain the most salient characteristics and suppress noisy or occluded view interference; and a lightweight classification module that maps the fused feature representations to target categories. Additionally, to mitigate the data scarcity bottleneck in sonar ATR, we design a multi-view sample augmentation method based on sonar imaging geometric principles: this method systematically combines single-view samples of the same target via the combination formula and screens valid samples within a predefined azimuth range, constructing high-quality multi-view training datasets without relying on complex generative models or massive initial labeled data. Comprehensive evaluations on the Custom Side-Scan Sonar Image Dataset (CSSID) and Nankai Sonar Image Dataset (NKSID) demonstrate the superiority of our framework over single-view baselines. Specifically, the two-view MVDCNN achieves average classification accuracies of 94.72% (CSSID) and 97.24% (NKSID), with relative improvements of 7.93% and 5.05%, respectively; the three-view MVDCNN further boosts the average accuracies to 96.60% and 98.28%. Moreover, MVDCNN substantially elevates the precision and recall of small-sample categories (e.g., Fishing net and Small propeller in NKSID), effectively alleviating the class imbalance challenge. Mechanism validation via t-Distributed Stochastic Neighbor Embedding (t-SNE) feature visualization and prediction confidence distribution analysis confirms that MVDCNN yields more separable feature representations and more confident category predictions, with stronger intra-class compactness and inter-class discrimination in the feature space. The proposed MVDCNN framework provides a robust and interpretable solution for advancing sonar ATR and offers a technical paradigm for multi-view acoustic image understanding in complex underwater environments.

Imaging sonar is a primary means of underwater detection, but it faces challenges of high false alarm rates in sonar image target detection due to factors such as reverberation, noise, and resolution. This paper proposes a method to improve the false alarm rate by self-training a deep learning detector on sonar images. Self-training automatically generates proxy classification tasks based on the sonar image target detection dataset, and pre-trains the deep learning detector through these proxy classification tasks to enhance its learning effectiveness of target and background features. This, in turn, improves the detector’s ability to distinguish between targets and backgrounds, thereby reducing the false alarm rate. For the first time, this paper conducts target detection experiments based on deep learning using high-resolution synthetic aperture sonar images at two frequencies. The results show that, under the conditions of equal or higher recall rates, this method can reduce the false alarm rate by 3.91% and 18.50% on 240 kHz and 450 kHz sonar images, respectively, compared to traditional transfer learning methods.

Underwater target detection exhibits extensive applications in marine target exploration and marine environmental monitoring. However, conventional images of underwater targets present challenges including blurred contour information, complex environmental conditions, and pronounced scattering effects. In this work, an underwater target detection method based on YOLOv10 is designed, and the detection performance is compared with the YOLOv5 model. Experimental results demonstrate that the YOLOv10 model has a mAP50 of 85.6% on the URPC 2020 dataset, improving the mAP50 by 1.2% than that of YOLOv5. This model exhibits high detection accuracy and high proceeding speed, which provides a promising support for precise and fast underwater target detection.

Three-dimensional hydro-acoustic imaging is a research hot spot in the underwater acoustic signal processing field, which has a wide range of application prospects in marine environmental resource surveying, seabed topography and geomorphological mapping, and underwater early warning and monitoring. To solve the problem that the resolution of the current imaging sonar reduces rapidly with increase in distance and a scanning gap exists in side-scan sonar, we designed a down-looking 3D-imaging sonar with a linear array structure. The imaging scheme adopts a time-domain spatial beam-forming method with the Back Projection (BP) algorithm as the core, and the formation of a virtual plane array can effectively improve the along-track resolution. To cope with the interference of the carrier motion error on the imaging, we proposed a high-precision sub-wavelength motion compensation method based on a real-time acoustic calibration system. Simulation and real data experiments show that the motion compensation method can effectively eliminate the influence of motion error and make the imaging energy more focused, leading to higher-quality acoustic images. Under equal average energy, the maximum superimposed sound intensity values in the imaging results increased by 20.75 dB and 6.57 dB, respectively, for simulation and real data. After motion compensation, the resolution of this imaging system reached 3 cm × 3 cm × 2.5 cm @ Depth = 17 m, TBP = 30 s · Hz.

When a synthetic aperture sonar (SAS) system operates under low-frequency broadband conditions, the azimuth range coupling of the point target reference spectrum (PTRS) is severe, and the high-resolution imaging range is limited. To solve the above issue, we first convert multi-receivers’ signal into the equivalent monostatic signal and then divide the equivalent monostatic signal into range subblocks and the range frequency subbands within each range subblock in order. The azimuth range coupling terms are converted into linear terms based on piece-wise linear approximation (PLA), and the phase error of the PTRS within each subband is less than π/4. Then, we use the chirp-z transform (CZT) to correct range cell migration (RCM) to obtain low-resolution results for different subbands. After RCM correction, the subbands’ signals are coherently summed in the range frequency domain to obtain a high-resolution image. Finally, different subblocks are concatenated in the range time domain to obtain the final result of the whole swath. The processing of different subblocks and different subbands can be implemented in parallel. Computer simulation experiments and field data have verified the superiority of the proposed method over existing methods.

Representing and rendering dynamic underwater scenes present significant challenges due to the medium’s inherent properties, which result in image blurring and information ambiguity. To overcome these challenges and accomplish real-time rendering of dynamic underwater environments while maintaining efficient training and storage, we propose Underwater Dynamic Scene Reconstruction Gaussian Splatting (UDR-GS), a method based on Gaussian Splatting. By leveraging prior information from a pre-trained depth estimation model and smoothness constraints between adjacent images, our approach uses the estimated depth as a geometric prior to aid in color-based optimization, significantly reducing artifacts and improving geometric accuracy. By integrating depth guidance into the Gaussian Splatting (GS) optimization process, we achieve more precise geometric estimations. To ensure higher stability, smoothness constraints are applied between adjacent images, maintaining consistent depth for neighboring 3D points in the absence of boundary conditions. The symmetry concept is inherently applied in our method by maintaining uniform depth and color information across multiple viewpoints, which enhances the reconstruction quality and visual coherence. Using 4D Gaussian Splatting (4DGS) as a baseline, our strategy demonstrates superior performance in both RGB novel view synthesis and 3D geometric reconstruction. On average, across multiple datasets, our method shows an improvement of approximately 1.41% in PSNR and a 0.75% increase in SSIM compared with the baseline 4DGS method, significantly enhancing the visual quality and geometric fidelity of dynamic underwater scenes.

This work proposed a novel constant false alarm rate detection method based on Gaussian mixture model (GMM). The main contribution contains two aspects. First, the Gabor wavelet is used to determine the GMM component number. Second, the expression of probability of false alarm is derived based on the GMM parameter.

The utilization of polarimetric synthetic aperture radar (PolSAR) enables the preservation of more comprehensive target scattering characteristics compared to conventional SAR. Despite the significant advancements in deep learning (DL) technology for single polarimetric SAR ship detection, there remains a scarcity of DL research specifically focused on PolSAR ship detection. This article proposed a lightweight theory-driven network (LT-Net) for combining domain knowledge and DL techniques, which is suitable to detect small targets and reduce the computational complexity. The LT-Net incorporates convolutional block attention module, and employs differential downsampling during the process of high-dimensional feature extraction. The integration of domain knowledge lends each convolution step a distinct physical significance. Due to the limited availability of public datasets, this article presents a public fully polarized ship detection (FPSD) dataset for the first time, which contains 853 Pauli pseudocolor maps (in JPG format) and multilook complex data (in TIF format), with a total of 1714 ship targets from AIRSAR, UAVSAR, and RadarSAT-2. The FPSD encompasses a range of scenarios, with the majority of ship targets being small targets. Validation on FPSD shows that LT-Net exhibits significantly lower time complexity and space complexity compared to four popular target detection networks, Meanwhile, LT-Net achieves the best detection performance in the Pauli pseudocolor maps.

Two-dimensional phase unwrapping (2-D PU) is the process of converting the measured phase into the real phase in interferometric signal processing. Reliable unwrapping results are critical for digital elevation model (DEM) generation using interferometric synthetic aperture radar (InSAR) and interferometric synthetic aperture sonar (InSAS). The majority of previous research has concentrated on accuracy. Whereas the computational efficiency must be taken into account for the interferometric measurement system that requires real-time processing. This paper proposes a low-time-consuming algorithm that can accomplish high-precision 2-D PU for this application scenario. The neural network and a new path-based 2-D PU algorithm make up this algorithm. First, the incorrect region in the gradient field is predicted and corrected using the neural network. The output channelwise variance is then calculated and used to generate the quality maps. Finally, to achieve phase reconstruction, the path-based algorithm performs path planning and flooding integral according to quality maps and compensated gradient. This paper also provides a recommended data structure implementation to ensure the algorithm's high efficiency. Experimental results using InSAR and InSAS data show that the proposed algorithm is highly efficient and accurate.