Wenjie Su
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Wenjie Su has a total of 11 co-authors. Here are the top co-authors by number of shared publications:
- En Fan
School of Intelligent Engineering, Shaoxing University, Shaoxing, China
1 shared publication Hao Yu
Department of Computer Science and Engineering, Institute of Artificial Intelligence, Shaoxing University, Shaoxing 312000, China
1 shared publication- Jilong Wang
Zhejiang Ningwei Machinery Technology Co., Ltd., Shaoxing 311814, China
1 shared publication - Junlei Chen
School of Intelligent Engineering, Shaoxing University, Shaoxing, China
1 shared publication - Panjun Sun
Department of Computer Science and Engineering, Institute of Artificial Intelligence, Shaoxing University, Shaoxing 312000, China
1 shared publication - Qiushuo Chen
School of Intelligent Engineering, Shaoxing University, Shaoxing 312000, China
1 shared publication
Latest Publications
- Open Access
A Hardware–Software Integrated PCB Image Registration Method Based on Local Adaptive KNN and SIFT
Solder-joint detection and localization on large, complex printed circuit boards (PCBs) remain challenging because PCB images often contain unevenly distributed features, nonuniform illumination, specular reflection, scale variation and geometric distortion. Conventional SIFT-based registration methods usually use fixed matching parameters and therefore cannot adapt well to regions with different component densities. To address this limitation, this study proposes a hardware–software integrated PCB image registration framework that combines a robotic end effector with a locally adaptive K-nearest neighbor (LAKNN) strategy and SIFT descriptors. The hardware platform provides stable image acquisition through a lifting mechanism and ring-light illumination, while the software module adjusts the neighbor-search space according to local feature density and matching confidence. Experimental results show that the proposed LAKNN method achieves 90.8% inlier-match accuracy in the ablation experiment and improves registration robustness under height, region and viewpoint variations. The proposed framework provides a practical basis for automated PCB inspection and robotic soldering alignment.
Open AccessA Hardware–Software Integrated PCB Image Registration Method Based on Local Adaptive KNN and SIFT
Solder-joint detection and localization on large, complex printed circuit boards (PCBs) remain challenging because PCB images often contain unevenly distributed features, nonuniform illumination, specular reflection, scale variation and geometric distortion. Conventional SIFT-based registration methods usually use fixed matching parameters and therefore cannot adapt well to regions with different component densities. To address this limitation, this study proposes a hardware–software integrated PCB image registration framework that combines a robotic end effector with a locally adaptive K-nearest neighbor (LAKNN) strategy and SIFT descriptors. The hardware platform provides stable image acquisition through a lifting mechanism and ring-light illumination, while the software module adjusts the neighbor-search space according to local feature density and matching confidence. Experimental results show that the proposed LAKNN method achieves 90.8% inlier-match accuracy in the ablation experiment and improves registration robustness under height, region and viewpoint variations. The proposed framework provides a practical basis for automated PCB inspection and robotic soldering alignment.
- 1 April 2026
- Published by Elsevier in Neural Networks
- Vol. 201, 108927
Privacy emotion recognition in artificial intelligence: Research frontiers, key challenges, and prospects
- 1 April 2026
- Published by Elsevier in Neural Networks
- Vol. 201, 108927
Privacy emotion recognition in artificial intelligence: Research frontiers, key challenges, and prospects
- Open Access
- 3 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of Industry Applications
- Vol. 7, 474-484
Multiobjective Layered Optimization Analysis of Noncontact Integrated Permanent Magnet Vernier Motor for Deep-Sea Joint Motor
This paper proposes a non-contact integrated permanent magnet vernier motor (NCIPMVM) to address the challenges of implementing a high-performance direct-drive actuator within the extremely constrained and pressure-sealed environment of deep-sea joint motors. The core concept involves forming an axially rigid structure with a common inner rotor through the integration of a vernier motor and a magnetic gear (MG). To improve the effectiveness of the integrated design, an optimized layering strategy is introduced, which considers the demagnetization characteristics of permanent magnets (PMs). Sensitivity analysis and hierarchical parameter optimization are conducted using cross-parameter analysis in combination with the Pearson correlation coefficient. Design of experiment (DOE) models are then established across different levels, while NSGA-III and PSO algorithms are employed to perform parameter optimization. Layered optimization is further refined through iterative conditional constraints, with the goal of satisfying output performance requirements. Based on the optimized parameters, a prototype is fabricated. Finally, the validity and accuracy of both the proposed NCIPMVM and the optimization strategy are verified through finite element analysis (FEA) and experimental testing.
- 3 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of Industry Applications
- Vol. 7, 474-484
Open AccessMultiobjective Layered Optimization Analysis of Noncontact Integrated Permanent Magnet Vernier Motor for Deep-Sea Joint Motor
This paper proposes a non-contact integrated permanent magnet vernier motor (NCIPMVM) to address the challenges of implementing a high-performance direct-drive actuator within the extremely constrained and pressure-sealed environment of deep-sea joint motors. The core concept involves forming an axially rigid structure with a common inner rotor through the integration of a vernier motor and a magnetic gear (MG). To improve the effectiveness of the integrated design, an optimized layering strategy is introduced, which considers the demagnetization characteristics of permanent magnets (PMs). Sensitivity analysis and hierarchical parameter optimization are conducted using cross-parameter analysis in combination with the Pearson correlation coefficient. Design of experiment (DOE) models are then established across different levels, while NSGA-III and PSO algorithms are employed to perform parameter optimization. Layered optimization is further refined through iterative conditional constraints, with the goal of satisfying output performance requirements. Based on the optimized parameters, a prototype is fabricated. Finally, the validity and accuracy of both the proposed NCIPMVM and the optimization strategy are verified through finite element analysis (FEA) and experimental testing.