The accurate semantic segmentation of point cloud data is the basis for their application in the inspection of extra high-voltage transmission lines (EHVTL). As deep learning evolves, point-wise-based deep neural networks have shown great potential for the semantic segmentation of EHVTL point clouds. However, EHVTL point cloud data are characterized by a large data volume and significant class imbalance. Therefore, the down-sampling method and point cloud feature extraction method used in current point-wise-based deep neural networks hardly meet the needs of computational accuracy and efficiency. In this paper, we proposed a two-step down-sampling method and a point cloud feature extraction method based on local feature aggregation of the point clouds after down-sampling in each layer of the model (LFAPAD). We then established a deep neural network named PowerLine-Net for the semantic segmentation of the EHVTL point clouds. Furthermore, in order to test and analyze the performance of PowerLine-Net, we constructed a point cloud dataset for the EHVTL scenes. Using this dataset and the Semantic3D dataset, we implemented network parameter testing, semantic segmentation, and an accuracy comparison of different networks based on PowerLine-Net. The results illustrate that the semantic segmentation model proposed in this paper has a high computational efficiency and accuracy in the semantic segmentation of EHVTL point clouds. Compared with conventional deep neural networks, including PointCNN, KPConv, SPG, PointNet++, and RandLA-Net, PowerLine-Net also achieves a higher accuracy in the semantic segmentation of EHVTL point clouds. Moreover, based on the results predicted by PowerLine-Net, the risk point detection for EHVTL point clouds has been achieved, which demonstrates the important value of this network in practical applications. In addition, as shown by the results of Semantic3D, PowerLine-Net also achieves a high segmentation accuracy, which proves its powerful capability and wide applicability in semantic segmentation for the point clouds of large-scale scenes.

Based on the seafloor reflection coefficient obtained from autonomous underwater vehicle (AUV) sub-bottom profile survey data of the northern slope of the South China Sea, combined with the sample test data of seafloor surface sediments, we use the Biot–Stoll model to establish the equations relating the seafloor reflection coefficient to the porosity, density, and mean grain size of the sediments at the dominant frequency of 5 kHz (the dominant frequency of the AUV sub-bottom profiler). The physical property parameters such as the porosity, density, and mean grain size of seafloor surface sediments are further inverted. Comparison of inversion results with measured results shows that the overall deviation ratios of the inverted mean grain size, porosity, and density of the surface sediments are in the ranges of − 13.56 to 14.44%, − 6.15 to 8.06%, and − 10.85 to 0.46%, respectively. Among them, the mean grain size directly reflects the size of seafloor sediment particles, and the particles are finer in deeper water. Overall, the inversion results are basically consistent with the measured values and thus can well reflect the variation characteristics of the physical properties of seafloor surface sediments.

This paper proposed an MSC-Transformer model based on the Transformer’s neural network, which was applied to seabed sediment classification. The data came from about 2900 km² of seabed area on the northern slope of the South China Sea. Using the submarine backscattering intensity and depth data obtained by the sub-bottom profiler, combined with latitude and longitude information, a seabed dataset of the slope area of the South China Sea was constructed. Moreover, using the MSC-Transformer, the accurate identification and judgment of sediment types such as calcareous bio-silt, calcareous bio-clay silt, silty sand, medium sand and gravel sand were realized. Compared with the conventional deep neural network CNN, RNN, etc., the model shows advantages when applied to the sediment dataset of the shallow sea slope region of the South China Sea. This confirms the feasibility and validity of the model and provides a reliable and accurate tool for seabed sediment classification in the field of marine science. The completeness and accuracy of the dataset and the good performance of the model provide a solid foundation for the scientificalness and practicability of the study.

This study focuses on the inversion of sub-bottom profile (SBP) data in the northern South China Sea using an empirical relationship derived from sediment acoustic data. The sub-bottom profile is primarily utilized for various marine applications, such as geological mapping and resource exploration. In this research, we present a study conducted in the northern slope canyon of the South China Sea. Firstly, we obtained the seabed reflection coefficient from sub-bottom profiles obtained by the autonomous underwater vehicle (AUV) detection system. Secondly, we utilized the acoustic empirical relationship in the northern South China Sea to establish relationship equations between the seabed reflection coefficient and the porosity, density, and average particle size of the sediment at a main frequency of 4 kHz (the AUV shallow profile main frequency). Then, using these equations, we were able to invert the physical parameters such as porosity, density, and average particle size of the seabed surface sediments. Finally, the inverted results are compared and analyzed by using the sediment samples test data. The overall deviation rate of the inverted physical parameters is within the range of ±10% when compared. The inverted results closely match the measured values, accurately reflecting the dynamic changes in the physical properties of seabed surface sediments. Notably, the average grain size is a direct indicator of the sediment particles size with smaller particles found in deeper water. The variation characteristics of sediment physical parameters align well with the variation of sediment types in the canyon, which is consistent with changes in the water depth, topography, and hydrodynamic conditions of the area. This further demonstrates the reliability of the inversion results.