| 摘 要: 针对钢铁工业中线材表面缺陷深度难以精准量化的问题，该文基于激光轮廓传感器设计了一个线材表面缺陷深度测量系统。系统通过线激光传感器与精密直线运动模组协同采集线材表面点云数据，并提出一种分区域约束抽样的改进RANSAC圆拟合算法，结合三次多项式全局平滑优化策略，有效构建连续稳定的理想基准面。在此基础上，通过深度偏差计算与三维连通域分析实现缺陷区域的识别与量化。实验结果表明，该系统可以有效应用于复杂线材表面缺陷深度测量，划痕、凹坑、局部凸起等缺陷的最大深度测量值与真实最大深度的偏差在0.01mm左右，满足工业检测精度需求，可为线材质量评估提供数据支撑。 |
| 关键词: 线材，三维点云，缺陷检测，RANSA，系统设计，激光传感器 |
|
中图分类号:
文献标识码:
|
|
| Design of a Wire Surface Defect Depth Measurement System Based on Line Laser Sensing Technology |
|
LuDeyun, HuaYunsong
|
USST
|
| Abstract: Aiming at the problem that the depth of wire surface defects in the iron and steel industry is difficult to accurately quantify, this paper designs a wire surface defect depth measurement system based on a laser profile sensor. The system collaboratively collects point cloud data of the wire surface through a line laser sensor and a precision linear motion module. An improved RANSAC circle fitting algorithm with regional constraint sampling is proposed, combined with a cubic polynomial global smoothing optimization strategy, to effectively construct a continuous and stable ideal datum plane. On this basis, the identification and quantification of defect regions are realized through depth deviation calculation and 3D connected component analysis. Experimental results show that the system can be effectively applied to the depth measurement of complex wire surface defects. The deviation between the measured maximum depth and the true maximum depth of defects such as scratches, pits, and local protrusions is about 0.01 mm, which meets the requirements of industrial detection accuracy and can provide data support for wire quality assessment. |
| Keywords: wire material, 3D point cloud, defect detection, RANSAC, system design, laser sensor |
