Search papers, labs, and topics across Lattice.
This study enhances seam segmentation for autonomous robotic welding by improving a BiSeNetV2 backbone through transfer learning and a hybrid Cross-Entropy--Lov谩sz loss, specifically targeting challenges posed by harsh illumination and reflective surfaces. The proposed framework achieves a Joint IoU of 81.76% and a mean IoU of 90.73%, marking a significant improvement of 22.36 percentage points over the baseline while maintaining efficiency in terms of FLOPs and inference speed. Additionally, it effectively addresses severe zero-IoU failures in reflective conditions, demonstrating superior seam continuity and robustness in demanding environments.
Achieving a 22.36 percentage point improvement in Joint IoU without increasing model complexity could revolutionize seam segmentation for robotic welding in challenging environments.
Reliable seam segmentation is essential for autonomous robotic welding in construction, where harsh illumination, specular reflections, and thin weld geometries often degrade segmentation performance. This study proposes a reflection-robust seam segmentation framework that enhances a BiSeNetV2 backbone through transfer learning and a hybrid Cross-Entropy--Lov谩sz loss. Rather than increasing architectural complexity, the proposed framework improves reflection robustness through learning-stability-oriented optimization. Experimental results show that the proposed method achieves 81.76\% Joint IoU and 90.73\% mIoU, improving Joint IoU by +22.36 percentage points over the OHEM-based baseline while maintaining identical FLOPs, parameter count, and inference speed. The proposed approach also recovers 96.33\% of severe zero-IoU failure cases under reflective conditions. Comparative experiments across BiSeNetV2, DeepLabV3+, UNet, and SegFormer further demonstrate that the proposed optimization strategy is particularly effective for lightweight real-time segmentation architectures. Qualitative analyses additionally show improved seam continuity and reflection robustness in challenging welding environments. These findings suggest that the proposed framework provides a practical and lightweight perception solution for robotic welding applications involving reflective metallic surfaces.
