A5036688125- BIM and Construction Integration
- Infrastructure Maintenance and Monitoring
- Metaheuristic Optimization Algorithms Research
- Advanced Multi-Objective Optimization Algorithms
- Structural Analysis and Optimization
- Structural Health Monitoring Techniques
- Land Use and Ecosystem Services
- Anomaly Detection Techniques and Applications
- 3D Surveying and Cultural Heritage
- Structural Engineering and Vibration Analysis
- Autonomous Vehicle Technology and Safety
- Structural Load-Bearing Analysis
- Topology Optimization in Engineering
- Economic and Environmental Valuation
- Advanced Manufacturing and Logistics Optimization
- Wind and Air Flow Studies
- Advanced Neural Network Applications
- Efficiency Analysis Using DEA
- Vibration Control and Rheological Fluids
- Structural Behavior of Reinforced Concrete
- Vehicle Dynamics and Control Systems
- Adversarial Robustness in Machine Learning
- Construction Project Management and Performance
- Mechanical Behavior of Composites
- Manufacturing Process and Optimization
Tongji University
2015-2025
China University of Mining and Technology
2022-2025
Shanghai Tenth People's Hospital
2020-2025
Puer University
2025
Anhui University of Science and Technology
2025
Yangzhou University
2024
China United Network Communications Group (China)
2023-2024
Ministry of Natural Resources
2024
Xiamen University
2023
Zhejiang University
2023
Deep neural networks have been shown to be very powerful modeling tools for many supervised learning tasks involving complex input patterns. However, they can also easily overfit training set biases and label noises. In addition various regularizers, example reweighting algorithms are popular solutions these problems, but require careful tuning of additional hyperparameters, such as mining schedules regularization hyperparameters. contrast past methods, which typically consist functions the...
In this paper, we propose a neural motion planner for learning to drive autonomously in complex urban scenarios that include traffic-light handling, yielding, and interactions with multiple road-users. Towards goal, design holistic model takes as input raw LIDAR data HD map produces interpretable intermediate representations the form of 3D detections their future trajectories, well cost volume defining goodness each position self-driving car can take within planning horizon. We then sample...
Modern autonomous driving systems rely heavily on deep learning models to process point cloud sensory data; meanwhile, have been shown be susceptible adversarial attacks with visually imperceptible perturbations. Despite the fact that this poses a security concern for self-driving industry, there has very little exploration in terms of 3D perception, as most only applied 2D flat images. In paper, we address issue and present method generate universal objects fool LiDAR detectors. particular,...
Detecting and locating precast components to confirm update components' status information is the key construction progress monitoring. There are several ways collect of elements in sites, such as manual collection, laser scanning, tag-based methods. But each these methods has its own limitations. Considering that effective accurate monitoring fundamental management, this paper proposes a novel framework integrates latest computer vision realize automatically walls, one essential...
The rapid urbanization and industrialization have caused excessive utilization of natural resources, which seriously threatened food security ecological in the world. Improving agricultural eco-efficiency (AEE) to obtain greater output at a lower environmental consume, is an significant part sustainable development. However, existing studies ignore carbon absorption when measuring AEE, leads evaluating results distortive some extent. Additionally, few research focuses on this topic within...
Prostate cancer (PCa) is a common malignant tumor with high morbidity and mortality worldwide. The prostate stem cell (PCSC) model provides novel insights into the pathogenesis of PCa its therapeutic response. However, roles molecular mechanisms specific genes in mediating fate decisions PCSCs carcinogenesis remain to be elusive. In this study, we have explored expression, function, mechanism AZGP1P2, pseudogene AZGP1, regulating stemness apoptosis treatment resistance docetaxel...
The suspension system is an important component of any vehicle as it transmits the force and torque between wheel frame, satisfying requirements ride comfort handling stability. To solve problem active control, a seven-degree-of-freedom model with electrohydraulic actuators established. Through approximate expansion in rolling time domain, robust predictive controller (RMPC) for designed RMPC deduced by defining performance evaluation function. A fractional PID used to control hydraulic...
Limited to the current building information model (BIM) development stage, many engineers are still using CAD 2D drawings for top-down design, and then generate BIM models BIM-related applications. As manual method of creating a semantic-rich is time consuming could cause defects, reluctant apply model. To solve this problem, semiautomatic methodology proposed structural from two-dimensional (2D) computer-aided design (CAD) drawings. First, it uses classified by layers geometric beams,...
In construction sites, safety issue has always been a threatening risk that must be taken seriously. order to improve the management capability of site, this paper establishes site fall hazard system. This system takes computer vision instead sensor systems as an information exchange bridge between physical and virtual model. The main contribution is propose deep learning-based approach achieve bi-directional from sites digital models. To train learning model, builds dataset containing...
Tower cranes can cover most of the area a construction site, which brings significant safety risks, including potential collisions with other entities. To address these issues, it is necessary to obtain accurate and real-time information on orientation location tower hooks. As non-invasive sensing method, computer vision-based (CVB) technology widely applied sites for object detection three-dimensional (3D) localization. However, existing methods mainly localization ground plane or rely...
Recently, dynamic optimization has received much attention from the swarm and evolutionary computation community. However, few studies have investigated data-driven optimization, most algorithms for are based on analytical mathematical functions. In this paper, we investigate optimization. First, develop a surrogate-assisted framework solving problems (DD-DOPs). Second, employ benchmark typical set in order to verify performance of proposed framework. The experimental results demonstrate...
The Fusarium species is an important plant pathogen that can cause diseases in grassland, leading to the degradation of grassland quality. However, morphology greatly affected by environmental factors, which makes it difficult identify its species. In addition, pathogenic ability different plants has not been fully studied. this study, isolates were obtained from herbaceous via tissue separation. Through morphological means and based on ITS, RPB2, TEF-1 gene sequences, we compared...
Background: The study aimed to investigate the association between non-high-density lipoprotein cholesterol (non-HDL-C) and risk of incident hypertension, especially concerning different levels low-density (LDL-C) non-HDL-C with hypertension. Methods: A total 10,623 participants from Rural Chinese Cohort Study were included in analyses. Non-HDL-C is blood minus high-density (HDL-C). odds ratios (ORs) 95% confidence intervals (95% CIs) non-HDL-C, LDL-C, hypertension estimated using logistic...