Article
Environmental Sciences
Chaofeng Yuan, Yuelei Xu, Qing Zhou
Summary: This study proposes a point-line SLAM system based on dynamic environments. It obtains dynamic region features through detection and segmentation of dynamic regions. The separation of dynamic and static objects is achieved through a geometric constraint method for matching line segments and a dynamic feature tracking method based on Bayesian theory, improving the robustness and accuracy of the SLAM system.
Article
Chemistry, Analytical
Shih-Hong Chio
Summary: A plane-based dynamic calibration method was proposed for the GeoSLAM ZEB Horizon handheld LiDAR scanner, and its efficiency was investigated through multiple datasets collected at different times and dates. The calibration results showed that the proposed method significantly reduced average residuals, improved the RMSE of check planes, and demonstrated consistent results across different calibration data.
Article
Engineering, Electrical & Electronic
Jichao Jiao, Chenxu Wang, Ning Li, Zhongliang Deng, Wei Xu
Summary: This article proposes a novel SLAM framework for dynamic environments, which combines neural network and motion information of dynamic objects to make the system more adaptable to dynamic scenes. By tightly coupling the results of object detection with geometric information in the SLAM system, and associating feature points in frames with dynamic probabilities, the proposed method greatly improves the localization accuracy in dynamic environments.
IEEE SENSORS JOURNAL
(2022)
Article
Environmental Sciences
Xuan He, Wang Gao, Chuanzhen Sheng, Ziteng Zhang, Shuguo Pan, Lijin Duan, Hui Zhang, Xinyu Lu
Summary: This study presents a LiDAR-Visual-Inertial Odometry (LVIO) system based on optimized visual point-line features, which effectively compensates for the limitations of a single sensor in real-time localization and mapping. The proposed algorithm extracts line features using a scale space and constraint matching strategy, and optimizes LiDAR matching accuracy using initial estimation results of Visual-Inertial Odometry. A factor graph based on Bayesian network is used for LVIO fusion, and evaluations show that the algorithm outperforms other state-of-the-art algorithms in real-time efficiency, positioning accuracy, and mapping effect.
Article
Computer Science, Interdisciplinary Applications
Ji Hee Kim, Naeun Choi, Seongmin Heo
Summary: This work proposes a novel iterative least squares method to approximate nonlinear functions using constrained least squares to ensure continuity. The method improves upon the existing continuous piecewise linear (CPWL) method by modifying the main steps and employing partitioned least squares and constrained least squares to reduce computational complexity. An iterative procedure with gradient descent using momentum is used for breakpoint updates to improve convergence characteristics.
COMPUTERS & CHEMICAL ENGINEERING
(2022)
Article
Computer Science, Artificial Intelligence
Nicolas Nadisic, Jeremy E. Cohen, Arnaud Vandaele, Nicolas Gillis
Summary: This paper introduces a new form of sparse MNNLS problem and a two-step algorithm to solve it. By dividing the problem into subproblems and selecting Pareto front solutions, a matrix that satisfies the sparsity constraint is constructed. Experimental results show that this method is more accurate than existing heuristic algorithms.
Article
Environmental Sciences
Pengju Tian, Xianghong Hua, Wuyong Tao, Miao Zhang
Summary: This paper proposes a novel method for extracting 3D line segment features from unorganized building point clouds, which has been demonstrated to work well in both high-quality and low-quality point clouds and exhibit robustness in the presence of noise. Compared to state-of-the-art techniques, the proposed method is considerably faster and more scalable.
Article
Chemistry, Analytical
Fei Zhou, Limin Zhang, Chaolong Deng, Xinyue Fan
Summary: This study introduces a monocular visual SLAM system that combines point features and line segment features to improve robustness in complex environments. Optimized algorithms and weight allocation methods ensure effective performance improvement of the system.
Article
Engineering, Multidisciplinary
Fei Xie, Wallace W. L. Lai, Xavier Derobert
Summary: The development of ground penetrating radar (GPR) as a near-surface geophysical detection method has led to an accurate survey equipment requiring an understanding of measurement errors and uncertainties. By using a constrained least squares algorithm, error sources affecting depth measurement of buried objects were modeled and uncertainty analysis was performed. Experimental validation showed that with a 95% confidence level, a centimetre-order of uncertainty can be achieved for depth estimation of objects at several meters deep, with errors in GPR center frequencies dominating the evaluation of uncertainty.
Article
Chemistry, Analytical
Zhangzhen Zhao, Tao Song, Bin Xing, Yu Lei, Ziqin Wang
Summary: This paper proposes a visual inertial SLAM algorithm based on point-line feature fusion, which improves the efficiency of the algorithm by improving line segment extraction and line feature matching, and achieves high-accuracy pose estimation by fusing point, line, and inertial data in a sliding window. Experiments show that the proposed algorithm performs better than traditional methods in handling SLAM tasks in indoor low-texture environments.
Article
Remote Sensing
Yonghua Jiang, Zhen Li, Meilin Tan, Shaodong Wei, Guo Zhang, Zhichao Guan, Bin Han
Summary: In this study, we propose a method for block adjustment without ground control points, based on the Rational Function Model (RFM). By using the image space affine model of RFM as the systematic error compensation model, we solve the abnormal convergence problem and achieve rapid and correct convergence. Comparative tests using multiple regional imagery validate the effectiveness and practicality of the proposed method.
INTERNATIONAL JOURNAL OF REMOTE SENSING
(2022)
Article
Computer Science, Software Engineering
Qingyu Li, Xin Wang, Tian Wu, Huijun Yang
Summary: This paper proposes an RGB-D SLAM method based on point-line feature fusion for real-time field 3D reconstruction. By optimizing the joint poses of point-line features, a 3D scene map of the field is constructed, and a joint point cloud filtering method based on keyframe optimization is designed.
COMPUTERS & GRAPHICS-UK
(2022)
Article
Computer Science, Software Engineering
Rusong Wu, Jing Bai, Wenjing Li, Jinzhe Jiang
Summary: In this study, a learning framework called dynamic confusion network (DCNet) is proposed, which captures subtle differences between samples from different sub-categories more robustly in fine-grained 3D point cloud classification. The experiments show that DCNet outperforms state-of-the-art methods and achieves the best performance in three fine-grained categories.
Article
Chemistry, Analytical
Hyeong Geun Jo, Beom Hoon Park, Do Yeong Joung, Jung Ki Jo, Jeong-Kyu Hoh, Won Young Choi, Kwan Kyu Park
Summary: The study introduces a wearable bladder scanner system that can continuously measure bladder volume in daily life. The system has shown similar measurement accuracy compared to commercial bladder imaging systems.
Article
Engineering, Multidisciplinary
Guangmin Li, Yu Gan, Guodong Liu, Fengdong Chen
Summary: This paper proposes a high-accuracy registration method based on double constrained intersurface mutual projections for 3D shape measurement. It effectively reduces the dislocation of discrete point clouds in the overlapping area, improving registration accuracy and reliability.
