A5101940716- Industrial Vision Systems and Defect Detection
- Surface Roughness and Optical Measurements
- Optical measurement and interference techniques
- Advanced Measurement and Detection Methods
- Non-Destructive Testing Techniques
- Image and Object Detection Techniques
- Ultrasonics and Acoustic Wave Propagation
- Advanced Vision and Imaging
- Video Surveillance and Tracking Methods
- Advanced Neural Network Applications
- Magnetic Properties and Applications
- Advanced Manufacturing and Logistics Optimization
- Welding Techniques and Residual Stresses
- Advanced Measurement and Metrology Techniques
- Scheduling and Optimization Algorithms
- Image Processing Techniques and Applications
- Infrastructure Maintenance and Monitoring
- Thermography and Photoacoustic Techniques
- Fault Detection and Control Systems
- Mineral Processing and Grinding
- Biometric Identification and Security
- Advanced Image Fusion Techniques
- Advanced machining processes and optimization
- Iron and Steelmaking Processes
- Metallurgical Processes and Thermodynamics
University of Science and Technology Beijing
2016-2025
Guangxi Normal University
2024
Third Affiliated Hospital of Zhengzhou University
2024
City University of Hong Kong
2024
José Rizal University
2024
Zhejiang University of Technology
2024
Northwest University
2023-2024
Shenyang The Fourth Hospital of People
2021-2024
The Fourth People's Hospital
2021-2024
Hunan University of Technology
2024
The detection of surface defects is very important for the quality improvement steel plates. In actual production, as plate production line runs faster, defect algorithm required to meet requirements real-time (less than 100 ms/image), and accuracy improved (at least 90%). this paper, an multi-block local binary pattern (LBP) proposed. This not only has simplicity efficiency LBP algorithm, but also finds a suitable scale describe features by changing block sizes, thus ensuring high...
In order to better control the quality of Flos Puerariae (FP), qualitative and quantitative analyses were initially performed by using chemical fingerprint chemometrics methods in this study. First, FP was developed HPLC markers screened out similarity analysis (SA), hierarchical clustering (HCA), principal components (PCA), orthogonal partial least squares discriminant (OPLS-DA). Next, constituents profiled identified coupled Fourier transform ion cyclotron resonance mass spectrometry...
High-speed railway in China has undergone rapid development recent years. Technology for structure measurement is recognized as an important aspect of steel rail quality inspection. The shape the welding base mainly gauged using mechanical contact technology. Matching this technology with escalating demands inspection rails a challenging task. In paper, structured light approach proposed and employed which intersects through structural plane projected by inner outer laser sensors. These...
It is difficult to detect roll marks on hot-rolled steel plates as they have a low contrast in the images. A periodical defect detection method based convolutional neural network (CNN) and long short-term memory (LSTM) proposed periodic defects, such marks, according strong time-sequenced characteristics of defects. Firstly, features image are extracted through CNN network, then feature vectors inputted into an LSTM for recognition. The experiment shows that rate this 81.9%, which 10.2%...
Defects on the surface of steel plates are one most important factors affecting quality plates. It is great importance to detect such defects through online inspection systems, whose ability defect identification comes from self-learning training samples. Extreme Learning Machine (ELM) a fast machine learning algorithm with high accuracy identification. ELM implemented by hidden matrix generated random initialization parameters, while different parameters usually result in performances. To...
This paper addresses the distributed filter design issue of nonlinear systems in sensor networks. Furthermore, by considering synthesized gain fluctuations, nonfragile scheme is developed to improve implementation robustness and applicableness. Since system model more practical general describe real‐world dynamical systems, objective measured a network grouped nodes investigated. Moreover, order solve this theoretical filtering problem, control theory are applied with transformation. More...
This paper studies the distributed state estimation issue of nonlinear dynamical systems with parameter uncertainties based on sensor networks under non-fragile control framework. Moreover, all sensors are in a fully framework information exchanges to reduce communication and computation resources. In particular, nonlinearities network gain fluctuations taken into account for more general applicability. With help Lyapunov-Krasovskii approach, sufficient convex optimization criteria can be...
Characteristics of surface defects hot rolled steel plates were analyzed, and an algorithm for detection was developed. Based on undecimated wavelet transform mathematical morphology, the can solve problem false alarms by scales water marks. Positions determined modular maximum inter-scale correlation coefficient at first, then classified with priori knowledge about characteristic plates. Experimental results shows that is not sensitive to disturbance scales, marks uneven illumination,...
The inclusion is a crucial factor affecting the quality of cord steel. formation inclusions closely related to abnormal production process in continuous casting process. Automatic anomaly detection algorithms are proposed replace manual visual screening according smart manufacturing paradigm, and then relationship between product mined through data-driven methods this paper. Convolutional neural networks autoencoder models employed detect various types anomalies time-dependent parameters. A...