A5101448230- Photonic and Optical Devices
- Semiconductor Lasers and Optical Devices
- Semiconductor Quantum Structures and Devices
- Advanced Computing and Algorithms
- Vehicle License Plate Recognition
- Robotic Path Planning Algorithms
- Service and Product Innovation
- Advanced Data and IoT Technologies
- Iterative Learning Control Systems
- Robotics and Automated Systems
- AI in cancer detection
- Medical Imaging and Analysis
- Service-Oriented Architecture and Web Services
- Advanced Technology in Applications
- Radiomics and Machine Learning in Medical Imaging
- Advanced Text Analysis Techniques
- Robotics and Sensor-Based Localization
- Optical Network Technologies
- Face and Expression Recognition
- Traffic Prediction and Management Techniques
Southwest Jiaotong University
2023-2024
Beijing University of Posts and Telecommunications
2005-2007
Institute of Optics and Electronics, Chinese Academy of Sciences
2005
Based on their balance of movement flexibility, obstacle surmounting ability and load capacity, quadruped robots are increasingly employed in industrial inspections. Quadruped usually utilize laser simultaneous localization mapping (SLAM) for autonomous navigation. However, SLAM coupled with legged odometry is susceptible to inaccuracies under the rapid rotation robots. To address this challenge, paper proposes a hybrid-dimensional framework indoor that operates using only 3D hybrid...
Aiming at the slow speed and low accuracy of traditional facial expression recognition, a new method combining attention mechanism is proposed. Firstly, group convolution used to reduce network parameters. The channels are grouped cut off redundant connections so that number parameters decreases significantly. Secondly, ERFNet model was improved by asymmetric residual module weak bottleneck improve running loss accuracy. Finally, added into feature extraction recognition precision....
This study aims to leverage a deep learning approach, specifically deformable convolutional layer, for staging cervical cancer using multi-sequence MRI images. is in response the challenges doctors face simultaneously identifying multiple sequences, task that computer-aided diagnosis systems can potentially improve due their vast information storage capabilities.
Due to the lack of domain and interface knowledge, it is difficult for users create suitable service processes according their needs. Thus, paper puts forward a new composition recommendation method. The method composed two steps: first step component based on recurrent neural network (RNN). When user selects component, RNN algorithm exploited recommend other matched services user, aiding completion composition. second Naive Bayes. completes composition, considering diversity interests,...
A two-dimensional spatially independent rate equation model of vertical-cavity surface-emitting lasers (VCSELs) is derived and then used to analysis the multi-mode behaviour VCSELs. The transverse mode characteristics VCSELs, carrier distribution in both radial azimuthal directions, effects non-uniformity injection current on behaviours are investigated detail. By using Bessel Fourier expansion density, 2D equations for formulated, which take into account diffusion direction as well gain...
This paper studies the distributed model free adaptive iterative learning control (MFAILC) of multiple high-speed trains (MHSTs) under malicious denial-of-service (DoS) attacks. By using equivalent linearization technique, discrete-time dynamic MHSTs is firstly converted into a linear data-based one. Then, strategy DoS attacker introduced, which represented by Bernoulli variable with unknown mathematical expectation. Next, MFAILC scheme conducted, belongs to scope data-driven approach....
In the field of high-speed train fault diagnosis, great achievements have been made in recent years. Although it has achieved considerable accuracy, most them require a large amount labeled training data, which is difficult to collect and obtain actual industrial environment. To solve this problem, paper proposes diagnosis framework based on federated transfer learning. Under premise protecting data privacy security, create model with excellent generalization capabilities can diagnose client...
A new comprehensive two-dimensional model of vertical-cavity surface-emitting lasers (VCSELs) is built up based on multimode rate equations. Combining the spatial properties, solved numerically in a self-consistent way with fourth-order Rugge-Kutta method. The modal characteristics, carrier distribution, relative intensity noise spectra and modulation features according to eye diagram are presented. numerical results useful design VCSELs, especially for applications fiber optical links.