Article
Computer Science, Artificial Intelligence
Yong Pang, Maolin Shi, Liyong Zhang, Wei Sun, Xueguan Song
Summary: In this paper, a novel algorithm named SDP-RLR is proposed to segment multivariate TBM time series with outliers caused by harsh operating environments and changeable tunneling statuses. The algorithm combines the 3 sigma rule with linear regression and dynamic programming to identify and remove outliers, and achieves efficient segmentation of TBM time series.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Construction & Building Technology
Shi Jin Feng, Yong Feng, Xiao Lei Zhang, Yi Han Chen
Summary: This paper proposes a deep learning-based approach for automatic, rapid and accurate detection of tunnel lining leakage in metro shield tunnels. The developed models achieve a balance between efficiency and accuracy through the use of multiple evaluation indices. The experimental results demonstrate high accuracy and low time and space complexities compared to existing approaches. The study also provides insights into the mechanisms behind the deep learning-based models.
TUNNELLING AND UNDERGROUND SPACE TECHNOLOGY
(2023)
Article
Computer Science, Artificial Intelligence
Maria Grammatikopoulou, Evangello Flouty, Abdolrahim Kadkhodamohammadi, Gwenole Quellec, Andre Chow, Jean Nehme, Imanol Luengo, Danail Stoyanov
Summary: Video feedback is crucial for surgical procedures and scene understanding in computer assisted interventions. Semantic segmentation is essential for identifying and localizing surgical instruments and anatomical structures, with deep learning advancing techniques in recent years. This paper introduces a dataset for semantic segmentation of cataract surgery videos and benchmarks the performance of deep learning models.
MEDICAL IMAGE ANALYSIS
(2021)
Article
Computer Science, Artificial Intelligence
Jiaxu Miao, Yunchao Wei, Xiaohan Wang, Yi Yang
Summary: This study introduces a large-scale video scene parsing dataset, VSPW, which contains 251,633 frames from 3,536 videos with pixel-level annotations. It covers 231 scenes and 124 object categories. The study also proposes a Temporal Attention Blending (TAB) network to improve pixel-level semantic understanding of videos. Experimental results demonstrate the superiority of the TAB approach on the VSPW dataset.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)
Article
Robotics
Jens Behley, Martin Garbade, Andres Milioto, Jan Quenzel, Sven Behnke, Juergen Gall, Cyrill Stachniss
Summary: Researchers introduced the SemanticKITTI dataset for holistic semantic scene understanding in self-driving, with point-wise semantic annotations of Velodyne HDL-64E point clouds from the KITTI Odometry Benchmark. The dataset includes three benchmark tasks: semantic segmentation, semantic scene completion, and panoptic segmentation, covering different aspects of semantic scene understanding.
INTERNATIONAL JOURNAL OF ROBOTICS RESEARCH
(2021)
Article
Construction & Building Technology
Arsalan Mahmoodzadeh, Mokhtar Mohammadi, Hawkar Hashim Ibrahim, Sazan Nariman Abdulhamid, Hunar Farid Hama Ali, Ahmed Mohammed Hasan, Mohammad Khishe, Hoger Mahmud
Summary: This study proposed four Machine Learning methods to predict TBM disc cutter's life, with Gaussian process regression model being the most accurate and K-nearest neighbors model having the lowest accuracy. Backward selection method revealed the significant contributions of different parameters to TBM disc cutter's life.
AUTOMATION IN CONSTRUCTION
(2021)
Article
Chemistry, Multidisciplinary
Yang Liu, Shuaiwen Huang, Di Wang, Guoli Zhu, Dailin Zhang
Summary: This paper proposes a new predictive model for determining the need to replace disc cutters in tunnel boring machines (TBM) based on operational parameters and geological conditions. The model achieves high accuracy and F-1 scores in predicting cutter replacement using specific parameters and established features.
APPLIED SCIENCES-BASEL
(2022)
Article
Computer Science, Artificial Intelligence
John Lambert, Zhuang Liu, Ozan Sener, James Hays, Vladlen Koltun
Summary: We propose MSeg, a composite dataset that merges semantic segmentation datasets from different domains. The taxonomies and annotations are reconciled to create a unified dataset. MSeg training results in more robust models compared to training on individual datasets or naive mixing of datasets. The models trained on MSeg achieve top rankings in benchmark challenges and show competitive performance on extreme generalization experiments. Comprehensive evaluation and sharing of models and code are crucial for progress.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)
Article
Biology
Guangqi Liu, Qinghai Ding, Haibo Luo, Min Sha, Xiang Li, Moran Ju
Summary: In this study, a new dataset of cervical cytology images called Cx22 is developed, and baseline methods for deep learning-based segmentation tasks are proposed. The suitability of the dataset is validated through experiments, revealing the impact of false-negative objects on the performance of segmentation methods.
COMPUTERS IN BIOLOGY AND MEDICINE
(2022)
Article
Geography, Physical
Weixiao Gao, Liangliang Nan, Bas Boom, Hugo Ledoux
Summary: Recent advancements in data acquisition technology have enabled rapid collection of 3D texture meshes, aiding in urban environment analysis and planning. Semantic segmentation through deep learning enhances understanding but demands a significant amount of labelled data. This research introduces a new benchmark dataset, semi-automatic annotation framework, and annotation tool for 3D meshes, offering potential time savings and comparative analysis for semantic segmentation methods.
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
(2021)
Article
Chemistry, Analytical
Hirra Anwar, Saad Ullah Khan, Muhammad Mohsin Ghaffar, Muhammad Fayyaz, Muhammad Jawad Khan, Christian Weis, Norbert Wehn, Faisal Shafait
Summary: Wheat stripe rust disease is harmful to wheat crop health and crop yield, but artificial intelligence and deep learning provide efficient solutions. AI algorithms can detect patterns that are difficult for humans to identify, enabling early disease detection. The scarcity of data related to specific crop diseases is a major challenge, so a new dataset called NUST Wheat Rust Disease (NWRD) dataset is introduced for wheat stripe rust disease segmentation.
Article
Chemistry, Physical
Ahmed Alnuaim, Ahmed M. Al-Mahbashi, Muawia Dafalla
Summary: This study assesses the use of tunnel boring machine (TBM) materials and granular materials in construction, and examines the impact of blending ratios on pavement design parameters. The study recommends on-site sieving and blending 3/8-inch aggregate with crushed limestone fine-powder material to optimize material quality. Stability and durability tests confirm the suitability of TBM-crushed powder material for subgrade and sub-base layers in pavement construction.
Article
Engineering, Geological
Qi Geng, Fei He, Maoxun Ma, Xiaohui Liu, Xuebin Wang, Zeyu Zhang, Min Ye
Summary: This study introduces a rarely reported full-scale experimental cutterhead system that combines in situ penetration tests and laboratory rock-breaking tests to investigate TBM penetration performance. The study provides insight into the cutting force, chipping performance, and boreability of the cutterhead system, and proposes models to predict cutter normal force and boreability index.
ROCK MECHANICS AND ROCK ENGINEERING
(2022)
Article
Computer Science, Artificial Intelligence
Lixiang Ru, Bo Du, Yibing Zhan, Chen Wu
Summary: We propose a new weakly-supervised semantic segmentation method that addresses the limitations of current methods by introducing a visual words learning module and hybrid pooling approach. Experimental results on two datasets demonstrate significant improvements in segmentation accuracy compared to state-of-the-art methods.
INTERNATIONAL JOURNAL OF COMPUTER VISION
(2022)
Article
Engineering, Mechanical
Jie Li, Bin Zhang, Dong Lyu, Jingbo Guo, Kang Su, Bo Hu
Summary: Based on a fatigue load spectrum, this paper establishes a model to calculate the fatigue propagation life of the cutterhead with different reliability, and analyzes the main factors affecting its reliability. The results show the dangerous failure positions and emphasize the importance of load stress amplitude and initial crack size in determining the crack propagation life and reliability of the cutterhead. The research provides a scientific basis for crack detection, life prediction, and reliability evaluation of the cutterhead.
ENGINEERING FAILURE ANALYSIS
(2022)
Article
Automation & Control Systems
Manna Li, Weijie Mao
Summary: This paper presents a dynamic output feedback control strategy for finite-time local piecewise control of parabolic partial differential equations, with sufficient conditions developed for stabilization of the closed-loop system through matrix inequalities feasibility, supported by simulation studies.
IET CONTROL THEORY AND APPLICATIONS
(2022)
Article
Multidisciplinary Sciences
Liang Chen, Zhitao Liu, Hongye Su, Fulong Lin, Weijie Mao
Summary: In this study, a self-convolution based attention fusion network (SAFN) is proposed for rock mass condition assessment. The network is designed to discover and fuse object attention maps within a deep neural network, enabling fine-grain classification of rock mass. Experimental results show that the proposed method outperforms state-of-the-art models in rock mass assessment and the field test demonstrates its accuracy and efficiency in automated classification of rock mass.
SCIENTIFIC REPORTS
(2022)
Article
Computer Science, Artificial Intelligence
Zhenfeng Xue, Weijie Mao, Yong Liu
Summary: This paper proposes a method to simulate synthetic datasets from a fine-grained perspective, enabling the system to be trained in an end-to-end manner. By converting it into an image-level data synthesis problem and training a generative model to approximate the rendering process, the whole framework becomes fully differentiable, allowing efficient optimization of attributes.
IET IMAGE PROCESSING
(2022)
Article
Computer Science, Artificial Intelligence
Zhenfeng Xue, Liang Chen, Zhitao Liu, Yong Liu, Weijie Mao
Summary: In this paper, a virtual-realistic fused dataset is proposed for assisting rock size analysis in TBM construction process. By using a virtual engine and a learning-based method, the method reduces the challenges in dataset acquisition and improves the test accuracy for segmentation in real applications.
NEURAL COMPUTING & APPLICATIONS
(2022)
Article
Automation & Control Systems
Liang Chen, Zhitao Liu, Weijie Mao, Hongye Su, Fulong Lin
Summary: This article proposes a TransTP network for real-time prediction of tunnel boring machine (TBM) driving parameters. By collecting a hybrid in situ prediction dataset, the network can learn multiperiod feature representation and extract multivariate features, achieving accurate prediction of TBM driving parameters.
IEEE-ASME TRANSACTIONS ON MECHATRONICS
(2022)
Article
Engineering, Civil
Yanfu Li, Xiaopeng Yang, Liang Chen, Yingjian Zhi, Hongli Liu
Summary: This paper focuses on solving the challenges of rail wear measurement in complex field environments. It proposes a solution that includes preprocessing to locate the rail waist, a hybrid model called R-H-ICP for profile registration, and a method for accurately detecting outliers. Extensive experiments validate the efficiency and superiority of the proposed methods.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Article
Construction & Building Technology
Liang Chen, Kimihiro Hashiba, Zhitao Liu, Fulong Lin, Weijie Mao
Summary: The maximum ground surface settlement prediction is a complex problem with many influencing factors. A hybrid prediction dataset is constructed to estimate the settlement, which includes geological and construction parameters based on spatial and temporal series. A spatial-temporal fusion network (STF-Network) is proposed to handle the multi-modal and multi-variate series prediction task.
AUTOMATION IN CONSTRUCTION
(2023)
Article
Computer Science, Artificial Intelligence
Yunliang Jiang, Chenyang Gu, Zhenfeng Xue, Xiongtao Zhang, Yong Liu
Summary: This paper proposes a novel idea to tackle the problems of image person removal by data synthesis. Two dataset production methods are proposed to automatically generate images, masks, and ground truths. A learning framework similar to local image degradation is used to guide the feature extraction process and gather more texture information for final prediction. Experimental results demonstrate the effectiveness of the method and the trained network has good generalization ability.
IET IMAGE PROCESSING
(2023)
Article
Chemistry, Analytical
Tao Huang, Zhe Chen, Wang Gao, Zhenfeng Xue, Yong Liu
Summary: Efficient trajectory generation in complex dynamic environments for unmanned surface vehicles (USVs) remains a challenge due to interference from hull swing and ambient weather. This paper proposes a cooperative trajectory planning algorithm for a coupled USV-UAV system, utilizing a lightweight semantic segmentation network and 3D projection transformation. The algorithm generates an initial obstacle avoidance trajectory using a graph-based search method and introduces a numerical optimization method based on hull dynamic constraints for easier trajectory tracking. A motion control method based on NMPC with energy consumption constraint is also proposed.
Article
Computer Science, Information Systems
Zhenfeng Xue, Weijie Mao, Liang Zheng
Summary: This paper explores content adaptation in the context of semantic segmentation using data simulation engines. The authors propose a scalable discretization-and-relaxation approach to optimize attribute values and generate training sets similar to real-world data. The experiment shows promising results in terms of real-world segmentation accuracy.
IEEE TRANSACTIONS ON MULTIMEDIA
(2022)
Article
Automation & Control Systems
Liang Chen, Zhitao Liu, Weijie Mao, Hongye Su, Fulong Lin
Summary: This article proposes a TransTP network for real-time prediction of tunnel boring machine (TBM) driving parameters, achieving superior performance.
IEEE-ASME TRANSACTIONS ON MECHATRONICS
(2022)
