Article
Computer Science, Information Systems
Jan Maria Juszczyk, Agata Wijata, Joanna Czajkowska, Michal Krecichwost, Marcin Rudzki, Marta Biesok, Bartlomiej Pycinski, Jakub Majewski, Jacek Kostecki, Ewa Pietka
Summary: This article presents an image acquisition system and wound surface reconstruction method using multiple imaging modalities, specifically for wounds located on the limbs. The method was validated on 29 data sets and compared with expert delineations and other contemporary methods. The experimental results show that the proposed method is statistically concordant with expert delineations in 3D.
Article
Astronomy & Astrophysics
Weisheng Hou, Hengguang Liu, Tiancheng Zheng, Hui Chang, Fan Xiao
Summary: This study presents a step-wise algorithm that combines sequential process and global optimization for constructing a three-dimensional training datum set and pattern database from two-dimensional geological cross-sections. The iterative optimization process effectively preserves and reproduces the stratigraphic sequence, while also verifying the randomness and effectiveness of the microstructure model.
EARTH AND SPACE SCIENCE
(2022)
Article
Computer Science, Artificial Intelligence
Linghao Chen, Jiaming Sun, Yiming Xie, Siyu Zhang, Qing Shuai, Qinhong Jiang, Guofeng Zhang, Hujun Bao, Xiaowei Zhou
Summary: In this paper, a novel system named Disp R-CNN is proposed for 3D object detection from stereo images. Unlike previous methods, this system predicts disparity only for pixels on objects of interest and leverages category-specific prior for more accurate disparity estimation. By using a statistical shape model to generate pseudo-ground-truth, the system can be trained without requiring LiDAR point clouds, making it more widely applicable. Experimental results show that Disp R-CNN outperforms previous state-of-the-art methods based on stereo input by 20 percent in terms of average precision for all categories.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2022)
Article
Computer Science, Interdisciplinary Applications
Chao Liu, Deli Wang, Han Zhang, Wei Wu, Wenzhi Sun, Ting Zhao, Nenggan Zheng
Summary: Reconstructing neuron morphologies from fluorescence microscope images is crucial for neuroscience studies. This study proposes a strategy of using two-stage generative models to simulate training data with voxel-level labels, resulting in realistic 3D images with underlying voxel labels. The results show that networks trained on synthetic data outperform those trained on manually labeled data in segmentation performance.
IEEE TRANSACTIONS ON MEDICAL IMAGING
(2022)
Article
Environmental Sciences
Chunhui Zhao, Chi Zhang, Yiming Yan, Nan Su
Summary: In this paper, a novel framework for 3D reconstruction of buildings based on a single off-nadir satellite image is proposed. The framework consists of two convolutional neural networks, Scale-ONet and Optim-Net, which can generate water-tight mesh models with exact shape and rough scale of buildings, and reduce scale errors for these models, respectively. Experimental results show that the framework has good robustness for different input images and can achieve ideal reconstruction accuracy on both model shape and scale of buildings.
Article
Engineering, Electrical & Electronic
Jia-Xiang Wang, Zhan-Li Sun, Zhi-Gang Zeng, Kin-Man Lam
Summary: This paper proposes an enhanced sparse representation approach for estimating the 3D shapes of objects in 2D image sequences, utilizing a two-stage scheme and a reweighted sparse representation model to extract shape bases. Experimental results on CMU image sequences demonstrate the effectiveness and feasibility of the proposed approach.
IEEE SIGNAL PROCESSING LETTERS
(2021)
Article
Biology
Liyue Shen, Wei Zhao, Dante Capaldi, John Pauly, Lei Xing
Summary: This study establishes a geometry-informed deep learning framework for ultra-sparse 3D tomographic image reconstruction. The seamless inclusion of geometric priors is shown to be essential for enhancing imaging performance and provides new avenues for data-driven biomedical imaging.
COMPUTERS IN BIOLOGY AND MEDICINE
(2022)
Article
Construction & Building Technology
Chen Wu, Hongwei Huang, Le Zhang, Jiayao Chen, Yue Tong, Mingliang Zhou
Summary: This paper presents a novel method for automated 3D evaluation of a tunnel leakage area using an improved GAN and Swin Transformer model. The Swin Transformer model outperforms other models in segmentation and evaluation, and a new 3D leakage area location model is proposed.
TUNNELLING AND UNDERGROUND SPACE TECHNOLOGY
(2023)
Article
Biochemical Research Methods
Qiufu Li, Linlin Shen
Summary: The paper introduces a 3D wavelet and deep learning-based method for neuron segmentation, utilizing 3D WaveUNet to process neuronal cubes and improve performance in noisy neuronal images. The integrated 3D wavelets efficiently assist in 3D neuron segmentation and reconstruction.
Article
Environmental Sciences
Chong Yang, Fan Zhang, Yunlong Gao, Zhu Mao, Liang Li, Xianfeng Huang
Summary: This paper proposes a method to recognize and eliminate the influence of moving cars in 3D modeling applications. Experimental results demonstrate that the proposed method is effective in recognizing and removing moving cars, repairing the geometric deformation and incorrect texture mapping problems caused by moving cars.
Article
Engineering, Geological
Kang Wang, Weidong Guo, Shaoshuai Shi, Ruijie Zhao, Xin Wang
Summary: This paper analyzes the requirements and main issues in three-dimensional modeling for tunnel engineering, and proposes a rapid reconstruction method based on geological feature points and line data. The method improves modeling speed and accuracy by integrating advanced geological prediction data. The three-dimensional geological model allows for analysis of geological conditions ahead of the tunnel face, and the information is validated through drilling verification. The tunnel construction scheme is optimized based on the information obtained from the model, achieving tunnel safety construction by reducing geological risk.
GEOTECHNICAL AND GEOLOGICAL ENGINEERING
(2023)
Article
Computer Science, Interdisciplinary Applications
Bo Yang, Min Liu, Yaonan Wang, Kang Zhang, Erik Meijering
Summary: This paper presents a 3D neuron segmentation network called SGSNet that enhances weak neuronal structures and removes background noises. The network utilizes two decoding paths, one for acquiring segmentation maps and the other for detecting neuronal structures. A structure attention module is designed to integrate features and provide contextual guidance to improve segmentation performance.
IEEE TRANSACTIONS ON MEDICAL IMAGING
(2022)
Article
Optics
Yulong Zong, Jin Liang, Huan Wang, Maodong Ren, Mingkai Zhang, Wenpan Li, Wang Lu, Meitu Ye
Summary: This study introduces a 3D surface defect detection system based on stereo vision, utilizing binocular stereo cameras and a speckle projector to capture 3D and color texture information of the object surface. The system employs a point-image mapping relationship and a Convolutional Neural Network for defect type identification, achieving accurate evaluation of defects.
OPTICS AND LASERS IN ENGINEERING
(2021)
Article
Engineering, Geological
Xin-Dong Wei, Zhi-Qiang Deng, Qin Li, Yan Huang, Gao-Feng Zhao
Summary: In this study, a surface image-based 3D grain-based model (GBM) reconstruction method is proposed using digital image processing (DIP), periodic random packing, and the simulated annealing algorithm. The mineral compositions are extracted using the K-means clustering algorithm and quantified using the two-point probability function (TPPF). The 3D GBM is generated by a simulated annealing algorithm, and the computational efficiency is improved using periodic boundary conditions. The GBM reconstruction method is successfully tested for reproducing the mechanical behaviors of granite and interpreting the failure mechanism at the mesoscale.
INTERNATIONAL JOURNAL FOR NUMERICAL AND ANALYTICAL METHODS IN GEOMECHANICS
(2023)
Article
Engineering, Multidisciplinary
Xinbin Wu, Jisong Wang, Junjie Li, Xuewu Zhang
Summary: Previous studies mainly focused on cracks and leaks in tunnel damage measurement methods, with little research on siltation at the bottom of water conveyance tunnels. This paper proposes a 3D measurement framework based on image 3D reconstruction technology for accurate measurement and intuitive understanding of underwater siltation. The method utilizes structure from motion (SfM) and multi-view stereo (MVS) algorithms to generate 3D point clouds of siltation, which are further processed to obtain accurate measurements of siltation thickness and area.
Article
Engineering, Multidisciplinary
Yongtao Liu, Jie Qiao, Yanting Hu, Tengyuan Fang, Ting Xu, Yusheng Xiang, Yi Han
MATHEMATICAL PROBLEMS IN ENGINEERING
(2020)
Article
Multidisciplinary Sciences
Yongtao Liu, Jie Qiao, Haibo Xu, Jiahui Liu, Yisong Chen
Article
Chemistry, Analytical
Wencai Zhou, Zhaowen Qiu, Shun Tian, Yongtao Liu, Lang Wei, Reza Langari
Summary: The paper proposes a hybrid scheme based on cost-based FMEA, fuzzy analytic hierarchy process (FAHP), and extended fuzzy multi-objective optimization to evaluate vehicle failure modes efficiently. The effectiveness of the method is validated through a case study and numerical analysis, indicating its potential to help enterprises and researchers in risk evaluation and identification of critical vehicle failure modes.
Article
Automation & Control Systems
Taiqi Wang, Yongtao Liu, Xinfeng Zhang
Summary: This work studies the fixed-time trajectory tracking control of autonomous surface vessels (ASVs) with unmeasured speed. A novel extended state observer (ESO) is used to estimate the unknown system states, including the unmeasured speed and disturbances. The fixed-time tracking task is completed using an output-constrained power integrator method. The practical fixed-time stability (FTS) of the closed-loop system is analyzed and the superiority of the designed scheme is demonstrated.
Article
Engineering, Civil
Qinyu Sun, Yingshi Guo, Yongtao Liu, Chang Wang, Menglu Gu, Yanqi Su
Summary: Driving simulators are widely used for analyzing driving behavior and developing intelligent driving algorithms. However, the validity of driving behavior data obtained from simulators is still uncertain. This study compared the effects of high speed and visual distraction on driving performance in both on-road and simulator experiments. The results showed that visual distractions impaired drivers' lane-keeping ability, and driving speed had similar effects on lane deviation and steering wheel angle in both environments. However, the study also revealed that even a high-fidelity driving simulator cannot achieve perfect absolute validity.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Article
Computer Science, Information Systems
Jie Qiao, Fengxiang Xu, Kunying Wu, Suo Zhang, Yongtao Liu
