Answers are grounded in this paper and cite their sources.
Demonstrates improved precision and speed in automating wire bundle disassembly with laser sensing technology, indicating significant safety benefits.
In response to the challenges of high labor intensity and safety risks in manually disassembling wire bundles from pipes and bars, this study proposes an automated disassembly method using laser vision sensing. A novel laser stripe centerline extraction algorithm based on unilateral tracking and midpoint prediction is introduced, featuring three key contributions: (1) a unilateral boundary tracking strategy that adaptively constructs the region of interest, reducing computational load by approximately 50%; (2) a hybrid center estimation approach that switches between the grayscale centroid method in clean regions and least-squares prediction in interference-affected regions based on stripe width deviation, achieving a superior speed–accuracy trade-off under industrial conditions; and (3) selective Hessian matrix-based sub-pixel refinement applied only at initial center points, preserving positioning precision while avoiding redundant computation. Experimental validation demonstrates that the proposed algorithm achieves a mean root mean square error of 0.66 ± 0.05 pixels under severe noise (signal-to-noise ratio = 8.52 dB), outperforming the geometric center, grayscale centroid, Steger, and improved U-Net methods by factors of 65.19, 8.89, 5.76, and 1.91, respectively. Processing speed is improved by 3.96–10.52 times over conventional methods. Standard gauge block measurements over 30 repetitions confirm a standard deviation as low as 0.0321–0.0532 mm, demonstrating excellent repeatability. The extracted centerline enables reliable identification of the maximum bundle seam and wire positions, providing robust guidance for robot-assisted automated disassembly.
No takes yet. Share an insight, caveat, or question.
Li et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: