Search Results (864)

Side-scan sonar and multi-beam echo sounder (MBES) are the most widely used underwater surveying tools in marine mapping today. The MBES offers high accuracy in depth measurement but is limited by low imaging resolution due to beam density constraints. Conversely, side-scan sonar provides high-resolution backscatter intensity images but lacks precise positional information and often suffers from distortions. Thus, MBES and side-scan images complement each other in depth accuracy and imaging resolution. To obtain high-quality seafloor topography images in practice, matching between MBES and side-scan images is necessary. However, due to the significant differences in content and resolution between MBES depth images and side-scan backscatter images, they represent a typical example of heterogeneous images, making feature matching difficult with traditional image matching methods. To address this issue, this paper proposes a feature matching network based on the LoFTR algorithm, utilizing the intermediate layers of the ResNet-50 network to extract shared features between the two types of images. By leveraging self-attention and cross-attention mechanisms, the features of the MBES and side-scan images are combined, and a similarity matrix of the two modalities is calculated to achieve mutual matching. Experimental results show that, compared to traditional methods, the proposed model exhibits greater robustness to noise interference and effectively reduces noise. It also overcomes challenges, such as large nonlinear differences, significant geometric distortions, and high matching difficulty between the MBES and side-scan images, significantly improving the optimized image matching results. The matching error RMSE has been reduced to within six pixels, enabling the accurate matching of multi-beam and side-scan images. Full article

Power generation is affected and structural instability may occur when biofouling attaches to the rotor of tidal stream turbines (TSTs). Image signals are used to identify biofouling for biofouling recognition, thus achieving on-demand maintenance, optimizing power generation efficiency, and minimizing maintenance costs. However, image signals are sensitive to background interferences, and underwater targets blend with the water background, making it difficult to extract target features. Changes in water turbidity can affect the effectiveness of image signal biofouling recognition, which can lead to reduced recognition accuracy. In order to solve these problems, a multi-view and multi-type feature fusion (MVTFF) method is proposed to recognize rotor biofouling on TSTs for applications in TST operation and maintenance. (1) Key boundary and semantic information are captured to solve the problem of background feature interference by comparing and fusing the extracted multi-view features. (2) The local geometric description and dependency are obtained by integrating contour features into multi-view features to address the issue of the target mixing with water. The mIoU, mPA, Precision, and Recall of the experimental results show that the method achieves superior recognition performance on TST datasets with different turbidity levels. Full article

Marine cobalt-rich crusts, extensively used in industries such as aerospace, automotive, and electronics, are crucial mineral resources located on the ocean floor. To effectively exploit these valuable resources, underwater imaging is essential for real-time detection and distribution mapping in mining areas. However, the presence of suspended particles in the seabed mining environment severely degrades image quality due to light scattering and absorption, hindering the effective identification of the target objects. Traditional image processing techniques—including spatial and frequency domain methods—are ineffective in addressing the interference caused by suspended particles and offer only limited enhancement effects. This paper proposes a novel underwater image restoration method that combines polarization imaging and homomorphic filtering. By exploiting the differences in polarization characteristics between suspended particles and target objects, polarization imaging is used to separate backscattered light from the target signal, enhancing the clarity of the cobalt crust images. Homomorphic filtering is then applied to improve the intensity distribution and contrast of the orthogonal polarization images. To optimize the parameters, a genetic algorithm is used with image quality evaluation indices as the fitness function. The proposed method was compared with traditional image processing techniques and classical polarization imaging methods. Experimental results demonstrate that the proposed approach more effectively suppresses backscattered light, enhancing the clarity of target object features. With significant improvements in image quality confirmed by several no-reference quality metrics, the method shows promise as a solution for high-quality underwater imaging in turbid environments, particularly for deep-sea mining of cobalt-rich crusts. Full article

Marine geophysical and geological investigations are crucial for evaluating the construction suitability of artificial fish reefs (AFRs). Key factors such as seabed topography, geomorphology, sub-bottom structure, and sediment type significantly influence AFR design and site selection. Challenges such as material sinking, sediment instability, and scouring effects should be critically considered and addressed in the construction of AFR, particularly in areas with soft mud or dynamic environments. In this study, detailed investigations were conducted approximately seven months after the deployment of reef materials in the AFR experimental zones around Xiaoguan Island, located in the western South Yellow Sea, China. Based on morphological factors, using data from multibeam echosounders and side-scan sonar, the study area was divided into three geomorphic zones, namely, the tidal flat (TF), underwater erosion-accumulation slope (UEABS), and inclined erosion-accumulation shelf plain (IEASP) zones. The focus of this study was on the UEABS and IEASP experimental zones, where reef materials (concrete or stone blocks) were deployed seven months earlier. The comprehensive interpretation results of multi-source high-resolution acoustic images showed that the average settlement of individual reefs in the UEABS experimental zone was 0.49 m, and their surrounding seabed experienced little to no scouring. This suggested the formation of an effective range and height, making the zone suitable for AFR construction. However, in the IEASP experimental zone, the seabed sediment consisted of soft mud, causing the reef materials to sink into the seabed after deployment, preventing the formation of an effective range and height, and rendering the area unsuitable for AFR construction. These findings provided valuable scientific guidance for AFR construction in the study area and other similar coastal regions. Full article

In marine remote sensing, underwater images play an indispensable role in ocean exploration, owing to their richness in information and intuitiveness. However, underwater images often encounter issues such as color shifts, loss of detail, and reduced clarity, leading to the decline of image quality. Therefore, it is critical to study precise and efficient methods for assessing underwater image quality. A no-reference multi-space feature fusion and entropy-based metrics for underwater image quality assessment (MFEM-UIQA) are proposed in this paper. Considering the color shifts of underwater images, the chrominance difference map is created from the chrominance space and statistical features are extracted. Moreover, considering the information representation capability of entropy, entropy-based multi-channel mutual information features are extracted to further characterize chrominance features. For the luminance space features, contrast features from luminance images based on gamma correction and luminance uniformity features are extracted. In addition, logarithmic Gabor filtering is applied to the luminance space images for subband decomposition and entropy-based mutual information of subbands is captured. Furthermore, underwater image noise features, multi-channel dispersion information, and visibility features are extracted to jointly represent the perceptual features. The experiments demonstrate that the proposed MFEM-UIQA surpasses the state-of-the-art methods. Full article

The paper introduces a novel method for creating a photographic map of the seabed using images captured by the on-board photo and video systems of autonomous underwater vehicles (AUVs) during various missions, while incorporating navigation parameters. Additionally, it presents a new approach for storing this photo map on the on-board device in a mosaic format (tiles), which significantly accelerates operational visual inspection by enabling the automatic search and recognition of underwater objects that may exceed the coverage area of a single photograph. This capability is achieved by organizing the photo map into layers with varying zoom levels. Semi-natural experiments were conducted with data from actual missions using the real underwater vehicle demonstrate the high efficiency of the proposed method and algorithm. Unlike existing methods that form photo maps after the underwater vehicle has taken pictures of the bottom using special high-performance computers, the developed method forms a photo map directly during the movement of the vehicle, using only the computing power of the on-board computer. In addition, in the event of accidents, when it is necessary to detect objects of interest on the seabed as quickly as possible, it is necessary to provide a quick visual inspection of the generated photo map. For this purpose, we have developed an algorithm for saving a photo map in the form of a mosaic, which is widely used in interactive geographic maps, such as Google Maps. This algorithm differs from existing methods in that it selectively saves data to the on-board storage device to reduce the number of read and write operations, thus ensuring the timely operation of the entire process of creating a photo map at a given frequency of photography. After the generated map has been stored as a mosaic and a high-speed connection with the vehicle has appeared, the operator can immediately view the entire generated map using a regular web browser. Full article

This paper theoretically and experimentally studies the effect of underwater waveguide interface scattering on the nonlinear sound field characteristics of parametric array (PA) radiation. Based on the image source method, the components of the sound field in the waveguide are first analyzed. Then, a non-paraxial model is developed to account for the influence of interface scattering. This model enables accurate calculation of the wide-angle sound field. The impact of the sound source depth and the interface reflection coefficient on the distribution of the difference-frequency wave (DFW) sound field in the waveguide is studied. The interface alters the phase distribution of the DFW’s virtual source density function, thereby affecting the sound field accumulation process. Waveguide interfaces with different absorption coefficients influence the amplitude oscillation caused by interface reflection and change the sidelobe size of the DFW beam. The DFW sound field distribution is measured at three typical frequencies. Simulation and experimental results show that the attenuation of the DFW’s axial sound pressure level in the waveguide oscillates, and the DFW’s beamwidth gradually widens as the frequency decreases. The calculated results from the proposed model agree well with the measured data, with average errors along the sound axis and depth being less than 3 dB and 6 dB, respectively. This demonstrates the model’s superior applicability compared to the existing free-field model. Full article

When using enhanced images for underwater object detection, issues such as detail loss and increased noise often arise, leading to decreased detection efficiency. To address this issue, we propose the Feature Fusion Network with Local Information Exchange (FFNLIE) for underwater object detection. We input raw and enhanced images into the Swin Transformer in parallel for feature extraction. Then, we propose a local information exchange module to enhance the feature extraction capability of the Swin Transformer. In order to fully utilize the complementary information of the two images, our feature fusion module consists of two core components: the Discrepancy Information Addition Block (DIAB) and the Common Information Addition Block (CIAB). The DIAB and CIAB are designed by utilizing and modifying cross-attention mechanisms, which can easily extract image discrepancy information and common information. Finally, the fused features are fed into the object detector to perform object detection tasks. The experimental findings demonstrate that the FFNLIE exhibited exceptional performance across four underwater datasets. Full article

This model solves the problems of insufficient global feature attention, colour distortion, low contrast, and blurred details in previous methods by using a symmetric U-Net architecture and two new modules: the partial instance normalisation (PIN) module and the colour detail modulation (CDM) module. PIC-GAN effectively balances texture enhancement and feature retention while restoring colour details, ensuring excellent image quality. Experimental results on several publicly available underwater image datasets demonstrate the effectiveness of PIC-GAN, which improves PSNR by 3.23 dB and SSIM by 0.06 compared to WaterGAN, highlighting the potential of PIC-GAN as a powerful solution for underwater image enhancement tasks. Full article

This study examines the impact of seabed conditions on image segmentation for seabed target images acquired via side-scan sonar during sea experiments. The dataset comprised cylindrical target images overlying on two seabed types, mud and sand, categorized accordingly. The deep learning algorithm (U-NET) was utilized for image segmentation. The analysis focused on two key factors influencing segmentation performance: the weighting method of the cross-entropy loss function and the combination of datasets categorized by seabed type for training, validation, and testing. The results revealed three key findings. First, applying equal weights to the loss function yielded better segmentation performance compared to pixel-frequency-based weighting. This improvement is indicated by Intersection over Union (IoU) for the highlight class in dataset 2 (0.41 compared to 0.37). Second, images from the mud area were easier to segment than those from the sand area. This was due to the clearer intensity contrast between the target highlight and background. This difference is indicated by the IoU for the highlight class (0.63 compared to 0.41). Finally, a network trained on a combined dataset from both seabed types improved segmentation performance. This improvement was observed in challenging conditions, such as sand areas. In comparison, a network trained on a single-seabed dataset showed lower performance. The IoU values for the highlight class in sand area images are as follows: 0.34 for training on mud, 0.41 for training on sand, and 0.45 for training on both. Full article

Underwater fish image segmentation is a crucial technique in marine fish monitoring. However, typical underwater fish images often suffer from issues such as color distortion, low contrast, and blurriness, primarily due to the complex and dynamic nature of the marine environment. To enhance the accuracy of underwater fish image segmentation, this paper introduces an innovative neural network model that combines the attention mechanism with a feature pyramid module. After the backbone network processes the input image through convolution, the data pass through the enhanced feature pyramid module, where it is iteratively processed by multiple weighted branches. Unlike conventional methods, the multi-scale feature extraction module that we designed not only improves the extraction of high-level semantic features but also optimizes the distribution of low-level shape feature weights through the synergistic interactions of the branches, all while preserving the inherent properties of the image. This novel architecture significantly boosts segmentation accuracy, offering a new solution for fish image segmentation tasks. To further enhance the model’s robustness, the Mix-up and CutMix data augmentation techniques were employed. The model was validated using the Fish4Knowledge dataset, and the experimental results demonstrate that the model achieves a Mean Intersection over Union (MIoU) of 95.1%, with improvements of 1.3%, 1.5%, and 1.7% in the MIoU, Mean Pixel Accuracy (PA), and F1 score, respectively, compared to traditional segmentation methods. Additionally, a real fish image dataset captured in deep-sea environments was constructed to verify the practical applicability of the proposed algorithm. Full article

Monitoring marine life through underwater object detection technology serves as a primary means of understanding biodiversity and ecosystem health. However, the complex marine environment, poor resolution, color distortion in underwater optical imaging, and limited computational resources all affect the accuracy and efficiency of underwater object detection. To solve these problems, the YOLOv9s-SD underwater target detection algorithm is proposed to improve the detection performance in underwater environments. We combine the inverted residual structure of MobileNetV2 with Simple Attention Module (SimAM) and Squeeze-and-Excitation Attention (SE) to form the Simple Enhancement attention Module (SME) and optimize AConv, improving the sensitivity of the model to object details. Furthermore, we introduce the lightweight DySample operator to optimize feature recovery, enabling better adaptation to the complex characteristics of underwater targets. Finally, we employ Wise-IoU version 3 (WIoU v3) as the loss function to balance the loss weights for targets of different sizes. In comparison with the YOLOv9s model, according to the experiments conducted on the UPRC and Brackish underwater datasets, YOLOv9s-SD achieves an improvement of 1.3% and 1.2% in the mean Average Precision (mAP), reaching 83.0% and 94.3% on the respective datasets and demonstrating better adaptability to intricate underwater environments. Full article

The field of underwater image processing has gained significant attention recently, offering great potential for enhanced exploration of underwater environments, including applications such as underwater terrain scanning and autonomous underwater vehicles. However, underwater images frequently face challenges such as light attenuation, color distortion, and noise introduced by artificial light sources. These degradations not only affect image quality but also hinder the effectiveness of related application tasks. To address these issues, this paper presents a novel deep network model for single under-water image restoration. Our model does not rely on paired training images and incorporates two cycle-consistent generative adversarial network (CycleGAN) structures, forming a dual-CycleGAN architecture. This enables the simultaneous conversion of an underwater image to its in-air (atmospheric) counterpart while learning a light field image to guide the underwater image towards its in-air version. Experimental results indicate that the proposed method provides superior (or at least comparable) image restoration performance, both in terms of quantitative measures and visual quality, when compared to existing state-of-the-art techniques. Our model significantly reduces computational complexity, resulting in a more efficient approach that maintains superior restoration capabilities, ensuring faster processing times and lower memory usage, making it highly suitable for real-world applications. Full article

Traditional underwater polarization imaging methods can only provide clear degree of polarization (DOP) and intensity images of the object but cannot provide images of the polarization state of the object. This paper proposes a method to extract clear object information from turbid water in all four Stokes parameter (I, Q, U, and V) channels by using the full Stokes camera, enabling clear polarization state image reconstruction. The method utilizes multiple images from different angles to construct a low-rank and sparse matrix. Then, by decomposing this matrix into sparse and low-rank components, clear Q, U, and V images (i.e., the full polarization state) can be obtained. Unlike traditional methods that assume the circularly polarized component (V component) to be zero, this method retains V channel information, allowing for circular polarization component measurement. The study successfully reconstructed clear underwater images of samples with inhomogeneous DOP distribution and obtained the clear polarization states of polarizers and fish in the turbid water. The results show that the proposed method can visualize and analyze the object’s polarization state quantitatively with high accuracy in turbid water for the first time, potentially extending the applicability of polarization underwater imaging in ocean exploration. Full article

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. Full article