- Civil and Geotechnical Engineering Research
- Railway Engineering and Dynamics
- Advanced Research in Systems and Signal Processing
- Geotechnical Engineering and Underground Structures
- Drilling and Well Engineering
- Hydraulic Fracturing and Reservoir Analysis
- Information Systems and Technology Applications
- Advanced Scientific Research Methods
- Digital Communication and Language
- Brain Tumor Detection and Classification
- Soil Mechanics and Vehicle Dynamics
- Natural Language Processing Techniques
- COVID-19 diagnosis using AI
- Linguistic Variation and Morphology
- AI in cancer detection
- Advanced Numerical Analysis Techniques
- Translation Studies and Practices
- Particle Detector Development and Performance
- Higher Education and Teaching Methods
- China's Ethnic Minorities and Relations
- Advanced Image Fusion Techniques
- Linguistics and Cultural Studies
- Advanced Neural Network Applications
- Optical measurement and interference techniques
- Seismic Imaging and Inversion Techniques
Chengdu University of Technology
2025
Fudan University
2023
Wannan Medical College
2023
Shenzhen University
2022-2023
Qingdao University
2020-2023
Donghua University
2022
Xiangtan University
2011
Tung Fang Design Institute
2005
Railway ballast performance is dictated by a complex mix of mechanical properties. These effect its at the particle level for example in terms degradation, but also track system settlement and stability. Therefore this paper seeks to develop new understandings behaviour identify opportunities future research directions. First, size distribution curves are discussed, considering improve breakage, drainage characteristics. Next, geometry with focus on form, angularity surface texture. This...
Large-scale hydraulic fracturing is a prevalent technique for exploiting low-porosity and low-permeability shale reservoirs. The propagation morphology of the fracture in laminated formations are significantly influenced by densely developed bedding planes, which can be classified into three categories: continuous, transitional, discontinuous, with each characterized distinct properties. This categorization complicates prediction optimization plans. In this research, comparative study was...
Back-translation (BT) has been widely used and become one of standard techniques for data augmentation in Neural Machine Translation (NMT), BT proven to be beneficial improving the performance translation effectively, especially low-resource scenarios. While most previous works related mainly focus on European languages with high relatedness, few them study less-related other areas around world. In this paper, we choose language pair less relatedness Asia: Chinese Vietnamese, investigate...
As an emerging business model, live broadcast e-commerce is growing rapidly in China recent years and prompts the emergence of network anchors, attracting customers to buy a variety goods their broadcasts. In this paper, we take Li Jiaqi, one most popular streaming anchors as example, collect transcribe his speech data broadcasts build speech-text corpus, then conduct corpus-based analysis on linguistic characteristics from aspects phonetic, usage various language units, rhetoric ways, by...
Aiming at the problem of frequent false and missed detection due to dense scattered distribution cancer cells in pathological images cervical cells, complex image background, rich proportion small objects, an improved method is proposed by combining advantage convolutional neural network transformer on basis YOLOv4 network. Firstly, this introduces CoT module into backbone feature extraction improve global information acquisition ability model visual expression model. Secondly, according...
Optical freeform surface with high geometric complexity requires appropriate sampling strategy to achieve holistic measurement improved efficiency and accuracy. This paper presents an adaptively plans measure the points based on particle swarm optimization algorithm for problem of measurement. In order provide mathematical foundation parametric modeling surfaces, model was then analyzed. The studied, error function between interpolation original constructed, coordinates were converted into...
Multispectral medical images contain many kinds of tissues, and the characteristics tissues in different band are different. How to extract valuable tissue information from multi-spectral is key diagnosis aided by image processing technology, segmentation an important prerequisite realize aided. In this article, algorithm spectral based on ANN improved. According convolutional neural network (CNN), method adding normalization layer between input hidden layer, numerical reduction after output...
Tire-derived aggregate (TDA) has been proposed to apply in the ballast or sub-ballast layer for degradation mitigation and vibration (noise) reduction. The TDA can change stiffness, which affect train-track-subgrade dynamic performance cause travel comfort safety issues. Towards this, this study aims at confirming 1) how much application of train layer; 2) what extent TDA-ballast distribute loadings reduce subgrade surface stress.
Hydraulic fracturing has become common practice for low-porosity and low-permeability shale oil or gas formations. The fracture propagation behavior is strongly affected by heterogeneous characteristics such as bedding planes natural fractures, leading to complex morphology. Considering the limitations inherent in true triaxial experiments conventional numerical simulation methods, a method based on discrete element (DEM) that couples hydro-mechanical deformation microearthquake analysis was...