A5076311162- Industrial Vision Systems and Defect Detection
- Advanced Neural Network Applications
- Surface Roughness and Optical Measurements
- Image and Signal Denoising Methods
- Medical Imaging Techniques and Applications
- Infrastructure Maintenance and Monitoring
- Sparse and Compressive Sensing Techniques
- Robotic Path Planning Algorithms
- Image and Object Detection Techniques
- Medical Image Segmentation Techniques
- Hand Gesture Recognition Systems
- Lung Cancer Diagnosis and Treatment
- Wireless Sensor Networks and IoT
- Advanced Algorithms and Applications
- Advanced Computing and Algorithms
- Advanced Image Processing Techniques
- Robotics and Sensor-Based Localization
- COVID-19 diagnosis using AI
- Robotics and Automated Systems
- Advanced Image Fusion Techniques
- Simulation and Modeling Applications
- Digital Media Forensic Detection
- Metaheuristic Optimization Algorithms Research
- Advanced X-ray and CT Imaging
- Non-Destructive Testing Techniques
Chongqing University of Science and Technology
2023-2024
Hebei University of Technology
2021-2024
Aiming at the problems of low detection efficiency and poor accuracy caused by texture feature interference dramatic changes in scale defect on steel surfaces, an improved YOLOv5s model is proposed. In this study, we propose a novel re-parameterized large kernel C3 module, which enables to obtain larger effective receptive field improve ability extraction under complex interference. Moreover, construct fusion structure with multi-path spatial pyramid pooling module adapt variation surface...
The use of low-dose computed tomography (LDCT) in medical practice can effectively reduce the radiation risk patients, but it may increase noise and artefacts, which compromise diagnostic information. methods based on deep learning improve image quality, most them a training set aligned pairs, are difficult to obtain practice. In order solve this problem, basis Wasserstein generative adversarial network (GAN) framework, we propose combining multi-perceptual loss fidelity loss....
Abstract Aiming at the defect inspection under characteristics of scale change, high reflection, inclined deformation defects lead bars and meeting needs for faster detection, this paper proposes a lighter cross-scale feature aggregation network (FLCNet). In study, we focus on redundancy channel information, design new partial group convolution, based which Faster C3 module lightweight fusion module. addition, slim neck to reduce redundant transfer model. Finally, propose uniform brightness...
Abstract Traditional bulky and complex control devices such as remote ground station cannot meet the requirement of fast flexible unmanned aerial vehicles (UAVs) in environments. Therefore, a data glove based on multi-sensor fusion is designed this paper. In order to achieve goal gesture UAVs, method can accurately recognize various gestures convert them into corresponding UAV commands. First, wireless fuses fiber optic sensors inertial construct dataset. Then, trained neural network model...
Path planning is a key technology to realize the autonomous navigation of mobile robots. The Informed-RRT* algorithm current path that solves high sampling efficiency in complex environments, but it also suffers from long times and redundant turns environments. For this reason, multi-strategy optimization proposed, first introduce WOA during operation re-selecting parent node so new can find selection optimal search radius, improve search, second select suitable Bessel curve interpolate...
Aiming at the problems of low detection efficiency and poor accuracy caused by different shapes sizes pill defects in existing market, an improved YOLOv5s model is proposed. In this study, deformable convolution introduced into CBS structure combined with C3 module to enable adaptively adjust receptive field more effectively approach shape size defect target. addition, Efficient Decoupled Head (EDH) replace head, separating positioning task classification task, improving target coverage. The...