A5010881869- Computer Graphics and Visualization Techniques
- Brain Tumor Detection and Classification
- Generative Adversarial Networks and Image Synthesis
- Advanced Vision and Imaging
- Industrial Vision Systems and Defect Detection
- Asphalt Pavement Performance Evaluation
- Text and Document Classification Technologies
- Medical Image Segmentation Techniques
- Anomaly Detection Techniques and Applications
- Infrastructure Maintenance and Monitoring
- Single-cell and spatial transcriptomics
- Advanced Fluorescence Microscopy Techniques
- Advanced Neural Network Applications
- Privacy-Preserving Technologies in Data
- Cell Image Analysis Techniques
- Adversarial Robustness in Machine Learning
- Neural Networks and Applications
- Concrete Corrosion and Durability
South China University of Technology
2024
University of Copenhagen
2021
Shandong Jianzhu University
2019
East China Normal University
2019
Techniques involving three-dimensional (3D) tissue structure reconstruction and analysis provide a better understanding of changes in molecules function. We have developed AutoCUTS-LM, an automated system that allows the latest advances 3D cellular developments using light microscopy on various tissues, including archived tissue. The workflow this paper involved advanced sampling methods human cerebral cortex, serial section collection system, digital library, cell detection convolution...
Histopathological whole-slide image (WSI) segmentation is essential for precise tissue characterization in medical diagnostics. However, traditional approaches require labor-intensive pixel-level annotations. To this end, we study weakly supervised semantic (WSSS) which uses patch-level classification labels, reducing annotation efforts significantly. the complexity of WSIs and challenge sparse labels hinder effective dense pixel predictions. Moreover, due to multi-label nature WSI,...
Orthotropic steel bridge decks and box girders are key structures of long-span bridges. Fatigue cracks often occur in these due to coupled factors initial material flaws dynamic vehicle loads, which drives the need for automating crack identification condition monitoring. With use unmanned aerial (UAV), acquirement surface pictures is convenient, facilitates development vision-based In this study, a combination convolutional neural network (CNN) with fully (FCN) designed Firstly, 120 images...
Distributed machine learning (DML) is widely used in resource-constrained networks where the intelligent decision needed, since no single node can work out accurate results from massive dataset within an acceptable time complexity. However, this will inevitably expose more potential targets to adversaries compared with non-distributed environment, especially when periodically sent distributed nodes for updating model, which should be adapted changing environments and requirements. In order...