This is our latest selection of worldwide publications and patents in english on 3D SLAM, between many scientific online journals, classified and focused on SLAM, LiDAR, point cloud, scan matching, feature extraction, loop closure, simultaneous localization, localization and mapping, odometry, Kalman filter, object detection, iterative closest point, voxel grid and 3D reconstruction.
A Symbol Conditional Entropy-Based Method for Incipient Cavitation Prediction in Hydraulic Turbines
Published on 2025-03-11 by Mengge Lv, Feng Li, Yi Wang, Tianzhen Wang, Demba Diallo, Xiaohang Wang @MDPI
Abstract: The accurate prediction of incipient cavitation is of great significance for ensuring the stable operation of hydraulic turbines. Hydroacoustic signals contain essential information about the turbine’s operating state. Considering that traditional entropy methods are easily affected by environmental noise when the state pattern is chaotic, leading to the extracted cavitation features not being obvious, a Symbol Conditional Entropy (SCE) feature extraction method is proposed to clas[...]
Our summary: Symbol Conditional Entropy method proposed for incipient cavitation prediction in hydraulic turbines using hydroacoustic signals. Improved fault information extraction and trend prediction with reduced uncertainty. Validation with RMSE, MAE, and MAPE showing superior performance compared to traditional methods.
Conditional Entropy, Incipient Cavitation, Hydraulic Turbines, Feature Extraction
Publication
A Transformer-Based Model for Encrypted Traffic Classification
Published on 2025-03-10 by Ziao Liu, Yuanyuan Xie, Yanyan Luo, Yuxin Wang, Xiangmin Ji @MDPI
Abstract: Encrypted network traffic classification remains a critical component in network security monitoring. However, existing approaches face two fundamental limitations: (1) conventional methods rely on manual feature engineering and are inadequate in handling high-dimensional features; and (2) they lack the capability to capture dynamic temporal patterns. This paper introduces TransECA-Net, a novel hybrid deep learning architecture that addresses these limitations through two key innovations. First,[...]
Our summary: A novel hybrid deep learning architecture, TransECA-Net, addresses limitations in encrypted network traffic classification by integrating ECA-Net modules with CNN architecture and incorporating a Transformer encoder to model global temporal dependencies. Extensive experiments demonstrate superior performance in classifying encrypted traffic types and faster convergence speed during training. This framework enables fine-grained service identification of encrypted traffic and real-time responsiveness, supporting adaptive network security monitoring systems.
Transformer-Based Model, Encrypted Traffic Classification, Deep Learning, Network Security
Publication
Loss Function Optimization Method and Unsupervised Extraction Approach D-DBSCAN for Improving the Moving Target Perception of 3D Imaging Sonar
Published on 2025-03-10 by Jingfeng Yu, Aigen Huang, Zhongju Sun, Rui Huang, Gao Huang, Qianchuan Zhao @MDPI
Abstract: Imaging sonar is a crucial tool for underwater visual perception. Compared to 2D sonar images, 3D sonar images offer superior spatial positioning capabilities, although the data acquisition cost is higher and lacks open source references for data annotation, target detection, and semantic segmentation. This paper utilizes 3D imaging sonar to collect underwater data from three types of targets with 1534 effective frames, including a tire, mannequin, and table, in Liquan Lake, Shanxi Province, Chi[...]
Our summary: This paper introduces a novel approach to improve the moving target perception of 3D imaging sonar by utilizing a density-based loss function, unsupervised extraction method D-DBSCAN, and rapid data annotation for underwater visual perception.
Loss Function Optimization, Unsupervised Extraction, D-DBSCAN, 3D Imaging Sonar
Publication
A Lightweight Defect Detection Network Aimed at Elevator Guide Rail Pressure Plates
Published on 2025-03-10 by Ruizhen Gao, Meng Chen, Yue Pan, Jiaxin Zhang, Haipeng Zhang, Ziyue Zhao @MDPI
Abstract: In elevator systems, pressure plates secure guide rails and limit displacement, but defects compromise their performance under stress. Current detection algorithms face challenges in achieving high localization accuracy and computational efficiency when detecting small defects in guide rail pressure plates. To overcome these limitations, this paper proposes a lightweight defect detection network (LGR-Net) for guide rail pressure plates based on the YOLOv8n algorithm. To solve the problem of exce[...]
Our summary: Lightweight defect detection network for elevator guide rail pressure plates based on YOLOv8n algorithm, achieves high localization accuracy and computational efficiency, outperforms other YOLO-series models.
Defect Detection, Lightweight Network, Elevator Guide Rail, YOLOv8n
Publication
A Multi-Scale Feature Fusion Model for Lost Circulation Monitoring Using Wavelet Transform and TimeGAN
Published on 2025-03-10 by Yuan Sun, Jiangtao Wang, Ziyue Zhang, Fei Fan, Zhaopeng Zhu @MDPI
Abstract: Lost circulation is a major challenge in the drilling process, which seriously restricts the safety and efficiency of drilling. The traditional monitoring model is hindered by the presence of noise and the complexity of temporal fluctuations in lost circulation data, resulting in a suboptimal performance with regard to accuracy and generalization ability, and it is not easy to adapt to the needs of different working conditions. To address these limitations, this study proposes a multi-scale feat[...]
Our summary: A model is proposed to monitor lost circulation in drilling process by fusing features at multiple scales using wavelet transform and TimeGAN, improving accuracy and generalization ability.
wavelet transform, TimeGAN, multi-scale feature fusion, drilling process
Publication
Research on Multi-Scale Point Cloud Completion Method Based on Local Neighborhood Dynamic Fusion
Published on 2025-03-10 by Yalun Liu, Jiantao Sun, Ling Zhao @MDPI
Abstract: Point cloud completion reconstructs incomplete, sparse inputs into complete 3D shapes. However, in the current 3D completion task, it is difficult to effectively extract the local details of an incomplete one, resulting in poor restoration of local details and low accuracy of the completed point clouds. To address this problem, this paper proposes a multi-scale point cloud completion method based on local neighborhood dynamic fusion (LNDF: adaptive aggregation of multi-scale local features throu[...]
Our summary: Research proposes LNDF method for multi-scale point cloud completion with dynamic feature aggregation and Transformer integration. Experimental results show significant improvement in completion accuracy and local detail preservation, outperforming SOTA on PCN and ShapeNet datasets. Enhanced generalization capability and completion fidelity demonstrated on real-world 3D scenarios from KITTI dataset.
Point Cloud Completion, Local Neighborhood Dynamic Fusion, Multi-Scale, Feature Extraction
Publication
Point cloud data transmission device, point cloud data transmission method, point cloud data reception device, and point cloud data reception method
Patent published on the 2025-03-06 in WO under Ref WO2025048473 by LG ELECTRONICS INC [KR] (Lee Jinwon [kr], Suh Jongyeul [kr], Park Hanje [kr])
Abstract: A mesh data decoding method, according to embodiments, may comprise the steps of: receiving a bitstream including mesh data; and decoding the mesh data. A mesh data encoding method, according to embodiments, may comprise the steps of: encoding mesh data; and transmitting a bitstream including the mesh data.[...]
Our summary: Methods for transmitting and receiving point cloud data, as well as encoding and decoding mesh data, are described in embodiments.
point cloud data transmission, point cloud data reception, mesh data decoding, mesh data encoding
Patent
On-chip lidar switching failure monitoring
Patent published on the 2025-03-06 in WO under Ref WO2025049882 by VOYANT PHOTONICS INC [US] (Tzuang Lawrence [us], Miller Steven [us])
Abstract: Disclosed are systems for monitoring on-chip LiDAR switching failure. A reduced number of power monitors are implemented between the switch fabric and transmit / receive pixels to capture the scan pattern thereby determining an operational status of the switches of the switch fabric.[...]
Our summary: Systems for monitoring on-chip LiDAR switching failure, reduced number of power monitors implemented, capture scan pattern to determine operational status.
on-chip lidar, switching failure monitoring, power monitors, switch fabric
Patent
EN 
FR 
DE 
ES 
IT