A5007918444- Ultrasonics and Acoustic Wave Propagation
- Non-Destructive Testing Techniques
- Video Surveillance and Tracking Methods
- Geophysical Methods and Applications
- Metallurgy and Material Forming
- Machine Fault Diagnosis Techniques
- Advanced Vision and Imaging
- High Temperature Alloys and Creep
- Flow Measurement and Analysis
- Aluminum Alloy Microstructure Properties
- Face and Expression Recognition
- Additive Manufacturing Materials and Processes
- Welding Techniques and Residual Stresses
- Structural Health Monitoring Techniques
- Photoacoustic and Ultrasonic Imaging
- Gear and Bearing Dynamics Analysis
- Microstructure and mechanical properties
- Optical measurement and interference techniques
- Face recognition and analysis
- Advanced Welding Techniques Analysis
- Blind Source Separation Techniques
- Advanced Measurement and Detection Methods
- Industrial Vision Systems and Defect Detection
- Advanced Image and Video Retrieval Techniques
- Metal Forming Simulation Techniques
Changsha University of Science and Technology
2016-2025
Xiamen University
2025
Henan University of Urban Construction
2008-2024
Hefei University of Technology
2021-2024
Chongqing Technology and Business University
2024
Tiangong University
2022-2023
Microbiology Institute of Shaanxi
2021-2023
Hunan University
2022
Beijing University of Chemical Technology
2018-2021
National University of Defense Technology
2021
High chromium superalloys have good overall mechanical properties and are widely used in critical hot-end components. The current manufacturing process still faces the problem of low forming efficiency. Extreme high-speed laser metal deposition (EHLMD) is a new efficient additive technology that can quickly directly shape direct difficult-to-process materials. However, EHLMD possesses some inter-layer defects alloy due to its excessively fast scanning speed, implying there potential for...
Sparse coding methods have achieved great success in visual tracking, and we present a strong classifier structural local sparse descriptors for robust tracking. Since the summary features considering codes are sensitive to occlusion other interfering factors, extract from fraction of all patches by performing pooling operation. The collection is combined into boosting-based tracking using discriminative appearance model. Furthermore, reconstruction error based weight computation method...
Abstract Alpha-fetoprotein (AFP) is a liver cancer associated protein and has long been utilized as serum tumor biomarker of disease progression. AFP usually detected in HCC patients by an antibody based system. Recently, however, aptamers generated from systematic evolution ligands exponential enrichment (SELEX) were reported to have alternative potential targeted imaging, diagnosis therapy. In this study, AFP-bound ssDNA screened identified using capillary electrophoresis (CE) SELEX...
Liquid metal embrittlement is a phenomenon in which the mechanical properties of metallic material are significantly reduced after contact with liquid metal, and microscopic mechanism this still controversial. The grain boundary penetration has recently been widely recognized, but theory deficient. To refine penetration, paper, aluminum by gallium obtained situ EBSD, combining it fracture morphology features comparing differences feature changes crack evolution process during tensile...
Feature pooling in a majority of sparse coding-based tracking algorithms computes final feature vectors only by low-order statistics or extreme responses codes. The high-order and the correlations between to different dictionary items are neglected. We present more generalized method for visual utilizing probabilistic function model statistical distribution Since immediate matching two distributions usually requires high computational costs, we introduce Fisher vector derive compact...
In visual tracking, usually only a small number of samples are labeled, and most existing deep learning based trackers ignore abundant unlabeled that could provide additional information for to boost their tracking performance. An intuitive way explain data is incorporate manifold regularization into the common classification loss functions, but high computational cost may prohibit those from practical applications. To overcome this issue, we propose two-stage approach tracker takes account...
The isothermal compression experiments with the strain rates of 0.01–10 s −1 and deformation temperature range 300–420 °C are performed to investigate hot behavior A356 aluminum alloy. Also, complex mechanisms analyzed. It is found that, as gradually increased, flow stress first rises, then stable appears without a tangible peak. microstructures exhibit large elongated grains, only few small new grains appear under most conditions. because dynamic recovery (DRV) dominant softening mechanism....
In this paper, we propose a manifold regularized correlation tracking method with augmented samples. To make better use of the unlabeled data and structure sample space, regularization-based filter is introduced, which aims to assign similar labels neighbor Meanwhile, regression model learned by exploiting block-circulant matrices resulting from translated samples over multiple base cropped both target nontarget regions. Thus, final classifier in our trained positive, negative, samples,...
The hot deformation behavior of AZ42 alloy was observed using thermal compression tests at a temperature scope 250-400 oC and strain rate 0.001-1 s-1. True stress-strain curves exhibited combination work hardening dynamic softening features. A Northern Goshawk algorithm (NGO)-optimized artificial neural network (ANN) model proposed. established NGO-ANN demonstrated impressive prediction accuracy, achieving high determination coefficient 0.991, mean absolute percentage error 3.51%, root...