A5101658179- Advanced Image Processing Techniques
- Generative Adversarial Networks and Image Synthesis
- Image and Video Quality Assessment
- Forecasting Techniques and Applications
- Multimodal Machine Learning Applications
- Innovative concrete reinforcement materials
- Digital Imaging for Blood Diseases
- Polymer composites and self-healing
- Cardiovascular Health and Disease Prevention
- Advanced Image Fusion Techniques
- Non-Invasive Vital Sign Monitoring
- AI in cancer detection
- Computer Graphics and Visualization Techniques
- Multi-Criteria Decision Making
- Optical Systems and Laser Technology
- Supply Chain and Inventory Management
- Visual Attention and Saliency Detection
- Video Surveillance and Tracking Methods
- Mechanical Behavior of Composites
- Medical Image Segmentation Techniques
- Hemodynamic Monitoring and Therapy
- Advanced Neural Network Applications
- Advanced Optical Sensing Technologies
- Optical and Acousto-Optic Technologies
- Cardiovascular Health and Risk Factors
University of Electronic Science and Technology of China
2024
Beijing Jiaotong University
2021-2024
Chinese Academy of Sciences
2020-2022
Hefei Institutes of Physical Science
2020-2022
University of Science and Technology of China
2020-2022
Zhejiang University
2022
Institute of Intelligent Machines
2022
Minzu University of China
2019
Ministry of Education of the People's Republic of China
2019
Northwest Minzu University
2018
Pedestrian detection plays a crucial role in autonomous driving by identifying the position, size, orientation, and dynamic features of pedestrians images or videos, assisting vehicles making better decisions controls. It's worth noting that performance pedestrian models largely depends on quality diversity available training data. Current datasets for have limitations terms diversity, scale, quality. In recent years, numerous studies proposed use data augmentation strategies to expand...
Accurate segmentation of subcellular components is crucial for understanding cellular processes, but traditional methods struggle with noise and complex structures. Convolutional neural networks improve accuracy require large, time-consuming, biased manually annotated datasets. Here, we developed SynSeg, a pipeline that generates synthetic training data to train U-net model structure segmentation, eliminating the need manual annotation. SynSeg leverages datasets variations in intensity,...
Summary This paper presents an efficient metric for evaluation the effect of inpainted Thangka images. In contrast to standard image quality metrics, proposed one takes into account some constraints and characteristics related specific goals inpainting techniques. The key is that we a method decompose reference distorted compare intensity fuzzy edge structure component similarity texture component. By comparative analyzing experimental results with other index delivers high consistency...
The prevalence of chronic disease is increasing annually as the development social economy and rapid changes residents' lifestyle in China. Meanwhile, role community health management prevention control diseases increasingly prominent. Therefore, we design implement a expert system for residents view current situation management. proposed can meet needs daily improve quality efficiency services, due to rich knowledge experience field it has. Finally, verify effectiveness by analyzing actual...