A5101400127- Spectroscopy and Chemometric Analyses
- Spectroscopy Techniques in Biomedical and Chemical Research
- Water Quality Monitoring and Analysis
- Remote-Sensing Image Classification
- Advanced Neural Network Applications
- Power Systems Fault Detection
- Face recognition and analysis
- Infrastructure Maintenance and Monitoring
- Face and Expression Recognition
- Optical Imaging and Spectroscopy Techniques
- Brain Tumor Detection and Classification
- Domain Adaptation and Few-Shot Learning
- Advanced Neuroimaging Techniques and Applications
- Electrical Fault Detection and Protection
- Emotion and Mood Recognition
- Medical Image Segmentation Techniques
- Visual Attention and Saliency Detection
- Advanced Image and Video Retrieval Techniques
Horizon Robotics (China)
2023
Heilongjiang University of Science and Technology
2022
Hubei University of Technology
2022
Huazhong University of Science and Technology
2021
Tiangong University
2019-2020
Tianjin University
2019-2020
Harbin Institute of Technology
2020
Abstract Due to the insufficiency of actual fault samples, machine learning based classification models power lines in smart grid are generally trained using simulated samples acquired from software, such as Matlab/Simulink. Yet, features and different, existed methods might not be valid or accurate enough classifying types lines. Thus, a new model for on deep-adversarial-transfer is proposed. Firstly, conditional generative adversarial network (CGAN) applied augmentation so increase data...
Thanks to the excellent global modeling capability of attention mechanisms, Vision Transformer has achieved better results than ConvNet in many computer tasks. However, generating hierarchical feature maps, still adopts aggregation scheme. This leads problem that semantic information grid area image becomes confused after aggregation, making it difficult for accurately model relationships. To address this, we propose Hierarchy Aware Feature Aggregation framework (HAFA). HAFA enhances...
With edge devices playing an increasingly important role in our daily lives, computing and human-computer interaction, especially facial expression recognition, become research central issues academia industry. However, surprisingly, utilizing neural networks for recognition has been neglected many years, very few can be found. To focusing on such topics. In this paper, we improve Visual Geometry Group 19 with the idea of residual learning. specific, each block 19, add its input to output....
Aiming at the problems of missed detection and poor due to variable scale complex background remote sensing images, this paper proposes a network enhance residuals attention based on YOLOv5 model, named ERA-YOLO. The framework focuses strengthening feature representation capability backbone network. First, Res2block module is designed in conjunction with idea res2net structure enable represent multi-scale features finer level granularity. Then, coordinate inserted improved for improving...