A5101857464- Topic Modeling
- Industrial Technology and Control Systems
- Autonomous Vehicle Technology and Safety
- Remote-Sensing Image Classification
- Natural Language Processing Techniques
- Electric and Hybrid Vehicle Technologies
- Traffic Prediction and Management Techniques
- Remote Sensing in Agriculture
- Biosensors and Analytical Detection
- Advanced Technologies in Various Fields
- VLSI and Analog Circuit Testing
- Remote Sensing and Land Use
- Geographic Information Systems Studies
- Geometry and complex manifolds
- Integrated Circuits and Semiconductor Failure Analysis
- Geometric Analysis and Curvature Flows
- Meat and Animal Product Quality
- Automated Road and Building Extraction
- Text and Document Classification Technologies
- Electric Vehicles and Infrastructure
- Text Readability and Simplification
- Power Systems and Technologies
- Molecular Communication and Nanonetworks
- Belt Conveyor Systems Engineering
- Electrowetting and Microfluidic Technologies
East China University of Technology
2017-2025
Ministry of Natural Resources
2022-2025
Guangdong Pharmaceutical University
2024
Ji Hua Laboratory
2023
Wuhan University of Technology
2023
Hunan University of Technology
2023
China Three Gorges University
2023
Ministry of Transport
2022
Beijing Institute of Technology
2021
Dalian University of Technology
2021
Deeplabv3+ currently is the most representative semantic segmentation model. However, tends to ignore targets of small size and usually fails identify precise boundaries in UAV remote sensing image task. To handle these problems, this paper proposes a algorithm images based on edge feature fusing multi-level upsampling integrated with (EMNet). EMNet uses MobileNetV2 as its backbone adds an detection branch encoder provide information for segmentation. In decoder, method designed retain...
Trajectory prediction is one of the core functions autonomous driving. Modeling spatial-aware interactions and temporal motion patterns for observed vehicles are critical accurate trajectory prediction. Most recent works on utilize recurrent neural networks (RNNs) to model usually need convolutional (CNNs) additionally capture spatial interactions. Although Transformer, a multi-head attention-based network, has shown its notable ability in many sequence-modeling tasks (e.g., machine...
The lake-floodplain wetlands are characterized by high biodiversity, difficult access, and significant environmental changes. Traditional remote sensing mapping methods struggle to generate consistent time-series data on wetland vegetation communities. Current research has endeavoured address this issue through the application of deep learning methodologies. However, a limitation these models is their reliance substantial volume training samples, which contradicts difficulty cost obtaining...
Abstract One of the key objectives in tunnel illness detection is identifying lining leakage, and deep learning-based image semantic segmentation approaches can automatically locate leakage. However, order to meet real-time processing needs professional mobile inspection equipment, existing leakage have difficulties real-time, dealing with voids, edge discontinuities leaking zone. To address aforementioned issues, this study introduces PP-LiteSeg-Attn model, which takes model PP-LiteSeg-B as...
Untethered microrobots provide the prospect for performing minimally invasive surgery and targeted delivery of drugs in hard-to-reach areas human body. Recently, inspired by way prokaryotic flagella rotates to drive body forward, numerous studies have been carried out study swimming properties helical swimmers. In this study, resistive force theory (RFT) was applied analyze influence dimensional kinematical parameters on propulsion performance conventional The efficiency index quantitatively...
The system of chemiluminescent magnetic enzyme-linked immunoassay was developed. E. coli O157:H7 sandwiched between rabbits anti-E. polyclonal antibody-coated nanoparticles (immunomagnetic or IMNPs) and mouse monoclonal antibody. Commercial alkaline phosphatase conjugated horse anti-mouse immunoglobulin (ALP-Ab) used to bind with the antibody, finally signals were detected by adding 3-(2'-spiroadamantane)-4-methoxy-4-(3"-phosphoryloxy)phenyl-1,2-dioxetane (AMPPD) which substrate reagent ALP....
Manufacturing Execution System (MES) is developed for information requirement of manufacturing enterprises. Because different MES in enterprises, almost all are customized. Therefor how to reduce modified codes and decrease the coupling between view page data become key system development. MVVM design model separates software structure view, by XAML Data Binding WPF technology. The number dramatically reduced because using pattern. In this paper, pattern development introduced Framework,...
Abstract In this study, a novel quality control strategy was proposed, aiming to establish multivariate specification for the processing step by exploring correlation between colors, chemical components, and hemostatic effects of carbonized Typhae pollen (CTP) using statistical analysis. The CTP samples were stir‐fried at different durations. Afterward, colorimeter LC–MS techniques applied characterize samples, followed determination bleeding time clotting mice evaluate their effect. Then,...
There has been significant development in wireless technology which enabled us to connect devices with conveniences. This paper makes an effort utilize one of the technologies ZigBee programmable logic controller (PLC) so that remote field can be controlled without wiring.
Chinese spelling correction (CSC), which is designed to detect and correct typos in the text, an important challenging task that has attracted increasing attention. Current CSC models based on pre-trained language (PLM) show powerful error capabilities, but cause problem many characters are incorrectly corrected. The of overcorrection particularly serious real-world application scenarios. To alleviate this situation, we propose a novel post-processing model: EDMSpell, utilizes two...
As an important carrier of individual information, the resume is data source for studying spatio-temporal evolutionary characteristics and group behaviors. This study focuses on information extraction geoparsing from resumes to provide basic technical support research based text. Most current studies text are oriented toward recruitment work, such as automated extraction, classification, recommendation resumes. These ignore behaviors implied in Therefore, this takes public teachers key...
With the rapid development of MEMS technology, MEMS-gyroscopes with its perfect characteristics small size, light weight, low cost and environmental adaptability are getting a wild rang application including both civilian military fields. At first, this paper introduces how MEMS-gyroscope works from basic mechanics, internal structure performance analysis characteristics. And then presents analyzes current status gyroscope technology. based three different market conditions working...
This paper designed a fuel cell power system statically to meet the dynamic requirement, with cell's model and vehicle built by GT-Suite. The influence mechanism of different rule-based energy management strategies on economy was studied. results revealed that worked more efficiently fuzzy logic control strategy than power-follower strategy, given an average efficiency increase 13.27%. Compared on/off heat dissipation reduced 79.67% strategy. Considering feasibility real-time implementation,...
ISIS, which abbreviates Intelligent Speech for Information Systems, is a trilingual spoken dialog system (SDS) the financial domain. It handles two dialects of Chinese (Cantonese and Putonghua), as well English predominant languages in our region. The supports language queries regarding stock market information simulated personal portfolios. Real-time retrieved directly from dedicated Reuters satellite feed. ISIS provides test-bed work multilingual speech recognition generation, speaker...
The uprising of rare-earth-doped fiber amplifier technology provides an effective solution to signal power compensation for interstellar optical communication systems, but significant attenuation induced by spatial irradiation has ever been a issue be resolved in practical applications. In order explore the impact radiation on performances high-speed differential phase shift keying (DPSK) satellite systems using EDFA preamplification technique, effect output central wavelength, half width,...
At present, span-based entity recognition methods are mainly used to accurately identify the span (entity) boundary for recognition, in which relative position information of and words region routinely ignored. This can be improve performance. Therefore, a nested model, integrates within span, is proposed. The representation first obtained with triaffine attention. Then, word region, as well previous representation, fused obtain new label-level another Finally, task carried out by...
We prepared a lenticular microlens array (LMA) using polyvinyl chloride (PVC) gel and an interdigitated electrode. By applying DC voltage to the electrode, surface of PVC can be waved with LMA character. When is removed, wavy recover its flat gradually. With aid polarity-inverted voltage, recovering time largely reduced. The present stable dynamic response when it repetitively impacted by pulse voltage. experimental results are given, mechanism reducing explained. Our improved has potential...
In this paper we prove classification results for gradient shrinking Ricci solitons under two invariant conditions, namely nonnegative orthogonal bisectional curvature and weakly PIC1, without any bound. New on ancient solutions the K\"ahler-Ricci flow are also obtained. The main new feature is that no upper bound assumed.