Qin Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Qin Zhang has a total of 74 co-authors. Here are the top co-authors by number of shared publications:
- Zhenmin Wang
School of Mechanical and Automotive Engineering, South China University of Technology, Guangzhou 510641, China
12 shared publications - Peng Chi
School of Mechanical and Automotive Engineering, South China University of Technology, Guangzhou, China
11 shared publications Haipeng Liao
School of Marine Science and Engineering, South China University of Technology, Guangzhou, People’s Republic of China
9 shared publicationsZhenmin Wang
School of Mechanical & Automotive Engineering, South China University of Technology, Guangzhou, People’s Republic of China
9 shared publications- Haipeng Liao
School of Mechanical & Automotive Engineering, South China University of Technology, Guangzhou, China
8 shared publications - Ting Li
School of Mechanical and Automotive Engineering, South China University of Technology, Guangzhou, China
8 shared publications
Latest Publications
- 15 August 2026
- Published by Springer Nature in The International Journal of Advanced Manufacturing Technology
- p. 1-17
A novel topology for a high-frequency pulsed arc-welding power supply
- 15 August 2026
- Published by Springer Nature in The International Journal of Advanced Manufacturing Technology
- p. 1-17
A novel topology for a high-frequency pulsed arc-welding power supply
- 15 August 2026
- Published by Springer Nature in Journal of Materials Engineering and Performance
- Vol. 35 (39) , 41053-41072
Effect of Heat Treatment on Precipitation Behavior and Mechanical Properties of Inconel 718 Welded by FFP-TIG Welding
- 15 August 2026
- Published by Springer Nature in Journal of Materials Engineering and Performance
- Vol. 35 (39) , 41053-41072
Effect of Heat Treatment on Precipitation Behavior and Mechanical Properties of Inconel 718 Welded by FFP-TIG Welding
- 22 March 2026
- Published by Springer Nature in The Visual Computer
- Vol. 42 (5) , 216
Multi-prior-guided deep learning framework for hepatic vessel segmentation
- 22 March 2026
- Published by Springer Nature in The Visual Computer
- Vol. 42 (5) , 216
Multi-prior-guided deep learning framework for hepatic vessel segmentation
- 9 March 2026
- Published by Springer Nature in Progress in Additive Manufacturing
- Vol. 11 (4) , 4513-4534
Influence of process parameters in fast-frequency single-pulse wire arc additive manufacturing on the microstructure and mechanical properties of Ti-6Al-4 V alloy
- 9 March 2026
- Published by Springer Nature in Progress in Additive Manufacturing
- Vol. 11 (4) , 4513-4534
Influence of process parameters in fast-frequency single-pulse wire arc additive manufacturing on the microstructure and mechanical properties of Ti-6Al-4 V alloy
- 9 March 2026
- Published by Springer Nature in Journal of Materials Science
- Vol. 61 (16) , 11353-11380
Microstructure, mechanical properties, and corrosion resistance of EH36 steel welds fabricated by fast-frequency double-pulsed TIG welding
- 9 March 2026
- Published by Springer Nature in Journal of Materials Science
- Vol. 61 (16) , 11353-11380
Microstructure, mechanical properties, and corrosion resistance of EH36 steel welds fabricated by fast-frequency double-pulsed TIG welding
- 30 January 2026
- Published by Springer Nature in International Journal of Computer Assisted Radiology and Surgery
- Vol. 21 (4) , 781-792
Digs: diffusion-guided Gaussian Splatting for dynamic occlusion surgical scene reconstruction
Purpose: Accurate 3D reconstruction from endoscopic videos is crucial for advancing computer-assisted minimally invasive surgery. However, existing approaches struggle with dynamic surgical scenes where instrument occlusions cause significant reconstruction artifacts. Although 3D Gaussian Splatting (3DGS) enables rapid reconstruction, it often suffers from incomplete surface recovery due to occlusion-induced missing regions and error propagation from suboptimal initial point clouds during radiance field optimization. This study aims to enhance reconstruction accuracy in dynamically occluded surgical environments. Methods: We propose a diffusion-guided Gaussian Splatting (DiGS) framework comprising two key components: (1) a diffusion-guided surface completion network that incorporates surgical scene priors to restore high-fidelity textures in occluded regions, improving surface completeness; and (2) a lightweight annealed smoothing mechanism designed to mitigate endoscope motion estimation errors, ensuring temporal coherence during continuous frame interpolation and stabilizing radiance field optimization. Results: Extensive experiments on the EndoNeRF and StereoMIS datasets demonstrate the superiority of DiGS over state-of-the-art baselines. On EndoNeRF, DiGS achieves a 61.75% improvement in LPIPS, indicating stronger perceptual alignment in dynamically occluded scenes. On StereoMIS, DiGS delivers an 7.03% PSNR gain and a 40.79% LPIPS improvement, along with consistently higher SSIM scores confirming superior preservation of structural details. Conclusion: The proposed DiGS framework effectively addresses the challenges of dynamic occlusions and motion-induced errors in surgical scene reconstruction, producing more accurate and temporally coherent 3D models. The code is publicly available at https://github.com/IGSResearch/DiGS.
- 30 January 2026
- Published by Springer Nature in International Journal of Computer Assisted Radiology and Surgery
- Vol. 21 (4) , 781-792
Digs: diffusion-guided Gaussian Splatting for dynamic occlusion surgical scene reconstruction
Purpose: Accurate 3D reconstruction from endoscopic videos is crucial for advancing computer-assisted minimally invasive surgery. However, existing approaches struggle with dynamic surgical scenes where instrument occlusions cause significant reconstruction artifacts. Although 3D Gaussian Splatting (3DGS) enables rapid reconstruction, it often suffers from incomplete surface recovery due to occlusion-induced missing regions and error propagation from suboptimal initial point clouds during radiance field optimization. This study aims to enhance reconstruction accuracy in dynamically occluded surgical environments. Methods: We propose a diffusion-guided Gaussian Splatting (DiGS) framework comprising two key components: (1) a diffusion-guided surface completion network that incorporates surgical scene priors to restore high-fidelity textures in occluded regions, improving surface completeness; and (2) a lightweight annealed smoothing mechanism designed to mitigate endoscope motion estimation errors, ensuring temporal coherence during continuous frame interpolation and stabilizing radiance field optimization. Results: Extensive experiments on the EndoNeRF and StereoMIS datasets demonstrate the superiority of DiGS over state-of-the-art baselines. On EndoNeRF, DiGS achieves a 61.75% improvement in LPIPS, indicating stronger perceptual alignment in dynamically occluded scenes. On StereoMIS, DiGS delivers an 7.03% PSNR gain and a 40.79% LPIPS improvement, along with consistently higher SSIM scores confirming superior preservation of structural details. Conclusion: The proposed DiGS framework effectively addresses the challenges of dynamic occlusions and motion-induced errors in surgical scene reconstruction, producing more accurate and temporally coherent 3D models. The code is publicly available at https://github.com/IGSResearch/DiGS.
- 21 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automation Science and Engineering
- Vol. 23 (15455955) , 4167-4181
Dynamic Path Optimization of Humanoid Welding Robots Based on Environmental Modeling and Real-Time Obstacle Detection
Welding plays a vital role in the fabrication and repair of steel structures. However, current operations still rely heavily on manual intervention, resulting in challenges such as inconsistent quality, low efficiency, poor working conditions, and elevated safety risks. Humanoid welding robots (HWR) have emerged as a potential solution, though they face challenges in environment modeling and path planning, particularly in narrow spaces. This paper presents a novel path optimization approach for HWR systems, integrating environment modeling with local obstacle avoidance to enable autonomous exploration of the welding area. First, the three-dimensional (3D) point clouds captured by binocular images are transformed into two-dimensional (2D) grids to identify narrow passages. Then, dynamic obstacles are detected through a 3D point cloud frame matching. The global path planning employs an enhanced A* algorithm, while local path planning utilizes an improved RRT* algorithm to ensure a short, smooth, and safe trajectory. The proposed method’s feasibility and effectiveness are validated through HWR experiments. Finally, the welding results of HWR was provided through right-angle welding, butt welding, and welding patches, further confirming the practicality of the method proposed in this paper.
- 21 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automation Science and Engineering
- Vol. 23 (15455955) , 4167-4181
Dynamic Path Optimization of Humanoid Welding Robots Based on Environmental Modeling and Real-Time Obstacle Detection
Welding plays a vital role in the fabrication and repair of steel structures. However, current operations still rely heavily on manual intervention, resulting in challenges such as inconsistent quality, low efficiency, poor working conditions, and elevated safety risks. Humanoid welding robots (HWR) have emerged as a potential solution, though they face challenges in environment modeling and path planning, particularly in narrow spaces. This paper presents a novel path optimization approach for HWR systems, integrating environment modeling with local obstacle avoidance to enable autonomous exploration of the welding area. First, the three-dimensional (3D) point clouds captured by binocular images are transformed into two-dimensional (2D) grids to identify narrow passages. Then, dynamic obstacles are detected through a 3D point cloud frame matching. The global path planning employs an enhanced A* algorithm, while local path planning utilizes an improved RRT* algorithm to ensure a short, smooth, and safe trajectory. The proposed method’s feasibility and effectiveness are validated through HWR experiments. Finally, the welding results of HWR was provided through right-angle welding, butt welding, and welding patches, further confirming the practicality of the method proposed in this paper.
- Open Access
- 19 January 2026
- Published by Springer Nature in Research Square
A Novel Topology for a High-Frequency Pulsed Arc-Welding Power Supply
Pulsed current can constrict the welding arc and enhance arc force, thereby promoting improved stirring within the molten pool. This effect has been widely studied and applied in the welding of metallic materials. In this paper, a high-frequency pulsed DC power supply is proposed for pulsed arc welding applications. The system comprises a constant-current source, a high-frequency pulse modulation circuit, and an energy recovery circuit. The output current amplitude (0 ~ 140 A) is regulated by the constant-current source, while the pulse frequency (20 ~ 40 kHz) and duty cycle (0 ~ 100%) are independently controlled via the pulse switching circuit. To suppress voltage spikes and recover energy from the output cable, an active clamping technique is applied to the switching devices in the modulation stage. This configuration enables rapid rise and fall times of the pulsed current, with controllable current slew rates. Furthermore, by optimizing the pulse control strategy, zero-voltage switching (ZVS) is achieved for the main switches, improving overall efficiency and reducing switching losses. The operating principles of the proposed topology are analyzed in detail, and key design parameters are calculated and presented. To verify the proposed topology, a laboratory prototype was built and tested, featuring a peak pulse current of 140 A and a maximum pulse repetition frequency of 40 kHz.
- 19 January 2026
- Published by Springer Nature in Research Square
Open AccessA Novel Topology for a High-Frequency Pulsed Arc-Welding Power Supply
Pulsed current can constrict the welding arc and enhance arc force, thereby promoting improved stirring within the molten pool. This effect has been widely studied and applied in the welding of metallic materials. In this paper, a high-frequency pulsed DC power supply is proposed for pulsed arc welding applications. The system comprises a constant-current source, a high-frequency pulse modulation circuit, and an energy recovery circuit. The output current amplitude (0 ~ 140 A) is regulated by the constant-current source, while the pulse frequency (20 ~ 40 kHz) and duty cycle (0 ~ 100%) are independently controlled via the pulse switching circuit. To suppress voltage spikes and recover energy from the output cable, an active clamping technique is applied to the switching devices in the modulation stage. This configuration enables rapid rise and fall times of the pulsed current, with controllable current slew rates. Furthermore, by optimizing the pulse control strategy, zero-voltage switching (ZVS) is achieved for the main switches, improving overall efficiency and reducing switching losses. The operating principles of the proposed topology are analyzed in detail, and key design parameters are calculated and presented. To verify the proposed topology, a laboratory prototype was built and tested, featuring a peak pulse current of 140 A and a maximum pulse repetition frequency of 40 kHz.
- 23 July 2025
- Published by Springer Nature in Welding in the World
- Vol. 70 (8) , 1-18
A magnetic arc blow extraction and quantification model based on YOLOv8n-improvement
- 23 July 2025
- Published by Springer Nature in Welding in the World
- Vol. 70 (8) , 1-18
A magnetic arc blow extraction and quantification model based on YOLOv8n-improvement