A5101736283- Video Surveillance and Tracking Methods
- Industrial Vision Systems and Defect Detection
- Gait Recognition and Analysis
- Advanced Manufacturing and Logistics Optimization
- Advanced Algorithms and Applications
- Computer Graphics and Visualization Techniques
- Simulation and Modeling Applications
- Face recognition and analysis
- Remote-Sensing Image Classification
- 3D Shape Modeling and Analysis
- Manufacturing Process and Optimization
- Advanced Image Fusion Techniques
- Anomaly Detection Techniques and Applications
- Remote Sensing and Land Use
- Geoscience and Mining Technology
- Fluid Dynamics Simulations and Interactions
- Image Processing Techniques and Applications
- Generative Adversarial Networks and Image Synthesis
- Remote Sensing and LiDAR Applications
- Color Science and Applications
- Advanced Numerical Analysis Techniques
- Infrared Target Detection Methodologies
- Advanced Sensor and Control Systems
- Semiconductor materials and devices
- Advanced Vision and Imaging
Anhui University of Technology
2020-2024
New York University
2021-2024
Hengyang Normal University
2024
China University of Geosciences
2024
Nantong University
2024
Changchun University
2024
Anhui University
2023
Anhui Science and Technology University
2023
Anhui University of Science and Technology
2023
Jiaxing University
2021-2023
In recent years, more and scholars devoted themselves to the research of target detection algorithm due continuous development deep learning. Among them, recognition small complex targets are still a problem be solved. The authors this article have understood shortcomings learning in detecting defect would like share new improved steel surface detection. defects will affect quality seriously. We find that most current algorithms for NEU‐DET dataset accuracy low, so we choose verify based on...
In the rapidly evolving electric vehicle industry, reliability of electronic systems is critical to ensuring safety and performance. Printed circuit boards (PCBs), serving as a cornerstone in these systems, necessitate efficient accurate surface defect detection. Traditional PCB detection methods, like basic image processing manual inspection, are inefficient error-prone, especially for complex, minute, or irregular defects. Addressing this issue, study introduces technology based on YOLOv5...
This paper considers the impact of setup time in production scheduling and proposes energy-aware distributed hybrid flow shop problem with sequence-dependent (EADHFSP-ST) that simultaneously optimizes makespan energy consumption. We develop a mixed integer linear programming model to describe this present two-stage adaptive memetic algorithm (TAMA) surprisingly popular mechanism. First, initialization strategy is designed based on two optimization objectives ensure convergence diversity...
Vehicle and license plate detection plays an important role in intelligent transportation systems is still a challenging task real applications, such as on-road scenarios. Recently, Convolutional Neural Network (CNN)-based detectors achieve the state-of-the-art performance. However, it difficult to efficiently detect vehicle simultaneously most cases. With single network, can affect of due inclusion relation. In this paper, we propose end-to-end deep neural network for detecting given image,...
Wear state is an important indicator of machinery operation condition that reveals whether faults have developed and maintenance should be scheduled. Among the available techniques, vision-based on-line monitoring wear particles in lubricant circuit preferred, where three-dimensional particle characterizations can obtained for mode analysis. This article presents application imaging system captures flow development image processing procedures multiview feature extraction. In particular, a...
To meet the practical needs of detecting various defects on pointer surface and solve difficulty some surface, this paper proposes a transfer learning improved Cascade-RCNN deep neural network (TICNET) algorithm for defects. Firstly, convolutional layers ResNet-50 are reconstructed by deformable convolution, which enhances feature extraction network. Furthermore, problems missing detection caused internal differences weak features effectively solved. Secondly, idea online hard example mining...
Abstract The fluid catalytic cracking (FCC) naphtha critical component‐oriented separation process is an efficient method to produce ultra‐low‐sulfur (<10 μg/g) gasoline with minimal loss of octane number (<1 RON). However, the product quality highly dependent on structure components FCC naphtha. Aromatics and thiophene sulfides without a methyl side chain favor olefin. major impulse olefin solvent‐induced dipole aromatics or sulfides, leading “Plane‐to‐Plane” combination between...
In Internet of things, lots instruments and sensors will connect into can be controlled through to achieve Smart Planet. One the key challenges is integrate business process. SOA an ideal infrastructure for process management definition languages are now available orchestration. Instruments functions encapsulate web services organized with other equally. But these device-oriented different from common because devices can't by more than one client at same time. This feature supported in...
With a high incidence rate and low diagnosis rate, vestibular migraine (VM) can seriously affect the quality of life patients, but it remains difficult to manage by current treatment options. Acupuncture may be potential option for VM prophylaxis, currently available evidence is still uncertain. Therefore, this trial aims evaluate efficacy safety acupuncture prophylaxis.
Research on transferring the learned person re-identification (ReID) model in source domain to other domains is of great importance since deploying a ReID new scenario common practical applications. Most existing unsupervised adaptation methods for employ framework pre-training domain, and clustering fine-tuning target domain. However, how reduce intra-domain variations narrow inter-domain gaps far from solved remains challenging problem under this framework. In paper, we address these...
Previous chapter Next Full AccessProceedings Proceedings of the 2013 SIAM International Conference on Data Mining (SDM)A Hierarchical Probabilistic Model for Low Sample Rate Home-Use Energy DisaggregationBingsheng Wang, Haili Dong, Arnold P. Boedihardjo, Feng Chen, and Chang-Tien LuBingsheng Lupp.704 - 712Chapter DOI:https://doi.org/10.1137/1.9781611972832.78PDFBibTexSections ToolsAdd to favoritesExport CitationTrack CitationsEmail SectionsAboutAbstract crisis climate change have caused a...