A5102999616- Advanced Image and Video Retrieval Techniques
- Industrial Vision Systems and Defect Detection
- Robotics and Sensor-Based Localization
- Advanced Neural Network Applications
- Medical Image Segmentation Techniques
- Gas Sensing Nanomaterials and Sensors
- Conducting polymers and applications
- Access Control and Trust
- Digital Media Forensic Detection
- Privacy-Preserving Technologies in Data
- Domain Adaptation and Few-Shot Learning
- Video Surveillance and Tracking Methods
- Advanced Vision and Imaging
- Fire Detection and Safety Systems
- Chaos-based Image/Signal Encryption
- Transition Metal Oxide Nanomaterials
- Image Retrieval and Classification Techniques
- Advanced Computational Techniques and Applications
- Image and Object Detection Techniques
- Advanced Measurement and Detection Methods
- Power Line Inspection Robots
Tianjin University of Science and Technology
2013-2024
In the domain of outdoor construction within power industry, working at significant heights is common, requiring stringent safety measures. Workers are mandated to wear hard hats and secure themselves with seat belts prevent potential falls, ensuring their reducing risk injuries. Detecting belt usage holds immense significance in inspections industry. This study introduces detection method for workers height based on UAV Image YOLO Algorithm. The YOLOv5 approach involves integrating CSPNet...
Outdoor electric power workers usually work at heights. In addition to wearing safety belts, operators wear a secondary protective rope prevent construction from falling. The plays an essential role in protecting the personal of and reducing injury. Therefore, is significant for inspection industry. detection based on UAV images YOLO algorithm effective method. Yolov5 adopts unique data enhancement method integrate CSP structure into backbone network while using improved SPPF Neck section...
This paper presents a feature extraction network called PD-Net (Points and Descriptors-Net) that extracts points generates descriptors from images. The model operates on full-size images using fully convolutional computes the coordinates of point in single forward pass. By building self-coding decoding deep neural learning dataset containing SIFT features, is able to quickly robustly acquire features image, loss function calculated sparse approach during training achieve results similar...