A5024595700- Industrial Vision Systems and Defect Detection
- Infrastructure Maintenance and Monitoring
- Chromosomal and Genetic Variations
- Genomics and Phylogenetic Studies
- RNA and protein synthesis mechanisms
- Geophysical Methods and Applications
- Structural Health Monitoring Techniques
- Concrete Corrosion and Durability
- Advanced Neural Network Applications
- Solar Radiation and Photovoltaics
- Effects of Environmental Stressors on Livestock
- Ultrasonics and Acoustic Wave Propagation
- Image Processing Techniques and Applications
- Synthesis and characterization of novel inorganic/organometallic compounds
- Optical measurement and interference techniques
- Non-Destructive Testing Techniques
- MicroRNA in disease regulation
- Advanced Algorithms and Applications
- Tribology and Lubrication Engineering
- Wind and Air Flow Studies
- Machine Learning in Bioinformatics
- Photovoltaic System Optimization Techniques
- Infant Nutrition and Health
- Organometallic Complex Synthesis and Catalysis
- Machine Fault Diagnosis Techniques
Sichuan University
2023-2024
Shandong Jianzhu University
2021-2022
Deakin University
2013-2019
ZheJiang Academy of Agricultural Sciences
2012
Shanghai Institute of Organic Chemistry
1994
Chinese Academy of Sciences
1994
Automatic inspection of concrete surface defects based on visual elements is crucial for the timely detection security risks in infrastructure. Moreover, accurate determination geographical location detected critical subsequent maintenance and reinforcement tasks. This study employed convolutional neural network (CNN) training methods localization. approach employs bounding boxes to confine damaged pixels utilizes projection loss foster similarity learning between pixels. In addition,...
Crack is a common disease of pavement, which will lead to more serious problems if it not found and maintained in time. This means that very important accurately extract measure the damage information pavement cracks. Compared with traditional methods, automatic detection segmentation cracks using visual elements are effective has become focused area. Although extensive researches used deep learning methods crack detection, these only involve single task or segmentation, few research...
The availability of large amounts protein-protein interaction (PPI) data makes it feasible to use computational approaches predict protein functions. base existing is exploit the known function information annotated proteins in PPI functions un-annotated proteins. However, these consider prediction domain (i.e. set from which are predicted) as unchangeable during procedure. This may lead valuable being overwhelmed by unavoidable noise when predicting functions, and turn, results will be...
The defects are usually generated during the structural materials subjected to external loads. Elucidating position distribution of using acoustic emission (AE) technique provides basis for investigating failure mechanism and prevention estimating location potentially dangerous sources. However, accuracy is heavily affected by both limitation localization area reliance on premeasured wave velocity. Here, we propose a novel AE source approach based generalized areal coordinates machine...
Eukaryotic genomes contain high volumes of intronic and intergenic regions in which repetitive sequences are abundant. These represent challenges genomic assignment short read generated through next generation sequencing often excluded analysis losing invaluable information. Here we present a method, known as tandem repeat assembler (TRA), for the assembly by constructing contigs directly from paired-end reads. Using an experimentally acquired data set human chromosome 14, repeats >200 bp...
Eukaryotic genomes contain high volumes of intronic and intergenic regions in which repetitive sequences are abundant. These represent challenges genomic assignment short read generated through next generation sequencing often excluded analysis losing invaluable information. Here we present a method, known as TRA (Tandem Repeat Assembler), for the assembly by constructing contigs directly from paired-end reads. Using an experimentally acquired data set human chromosome 14, tandem repeats...
Eukaryotic genomes contain high volumes of intronic and intergenic regions in which repetitive sequences are abundant. These represent challenges genomic assignment short read generated through next generation sequencing often excluded analysis losing invaluable information. Here we present a method, known as tandem repeat assembler (TRA), for the assembly by constructing contigs directly from paired-end reads. Using an experimentally acquired data set human chromosome 14, repeats >200 bp...