- Extraction and Separation Processes
- Medical Image Segmentation Techniques
- Icing and De-icing Technologies
- Face recognition and analysis
- Poxvirus research and outbreaks
- Video Surveillance and Tracking Methods
- Medical Imaging and Analysis
- Target Tracking and Data Fusion in Sensor Networks
- Software Testing and Debugging Techniques
- COVID-19 diagnosis using AI
- Image Processing Techniques and Applications
- Advanced Neural Network Applications
- Sustainable Supply Chain Management
- Air Quality Monitoring and Forecasting
- Machine Learning and Data Classification
- Gait Recognition and Analysis
- Imbalanced Data Classification Techniques
- Recycling and Waste Management Techniques
Zhengzhou University of Light Industry
2022-2025
University College London
2023-2025
Shandong University
2023
Monkeypox has been recognized as the next global pandemic after COVID-19 and its potential damage cannot be neglected. Computer vision-based diagnosis detection method with deep learning models have proven effective during period. However, limited samples, are difficult to full trained. In this paper, twelve CNN-based models, including VGG16, VGG19, ResNet152, DenseNet121, DenseNet201, EfficientNetB7, EfficientNetV2B3, EfficientNetV2M InceptionV3, used for monkeypox skin pictures. Numerical...
Brain medical image registration is a fundamental premise for the computer-assisted treatment of brain diseases. The one most important and complex organs human body, it very challenging to perform accurate fast on it. Aiming at problem voxel folding in deformation field low accuracy when facing fine objects, this paper proposed fully convolutional multi-constraint cascaded attention network (MCANet). composed two sub-network cascades performs coarse-to-fine input pairs an iterative manner....
In recent years, deep learning has made successful applications and remarkable achievements in the field of medical image registration, method registration based on become current research hotspot. However, performance convolutional neural networks may not be fully exploited due to neglect spatial relationships between distant locations incomplete updates network parameters. To avoid this phenomenon, MHNet, a multiscale hierarchical deformable for 3D brain MR images, was proposed paper. This...
Person re-identification can quickly locate and find all the specified targets in complex scenes with multiple cameras, which has been widely applied intelligent video surveillance security system. As a state-of-art method proposed recently, ResNet exhibits promising performances on person re-identification. However, without intermediate fully connected layer, fails to grasp global information detection process. To overcome above problem, this paper proposes named RG-BoTNet by fusing...