Volume 1,Issue 8
一种基于区域分割的钢板表面缺陷自动检测技术
钢板作为零件加工制造的常用材料之一,在机械领域具有广泛用途。由于加工环境的复杂性,其表面常出现划痕、腐蚀、裂纹等缺陷,存在尺度不一、类型多样、背景复杂等特点,在基于深度学习的缺陷检测领域中具有较强的挑战性。针对目前深度学习算法对钢板[3]表面缺陷检测精度低的问题,结合实际生产需求,本论文提出钢板区域自动分割技术,以缺陷分类模式提高钢板表面缺陷检测精度。为实现钢板区域自动分割,提出随机采样一致性算法提取钢板边缘直线,继而获得钢板在采集图像中的位姿,并通过图像的平移、旋转技术钢板摆正区域,实现钢板任意姿态下的区域自动分割。
[1]储茂祥.钢板表面缺陷检测关键技术研究[D].东北大学,2014.
[2]Neogi N, Mohanta D K, Dutta P K. Review of vision-based steel surface inspection systems[J]. EURASIP Journal on Image and Video Processing, 2014, 2014(1): 1-19.
[3]代晓林,刘梦玫,生群,等.基于改进Swin Transformer的钢板表面缺陷检测方法[J].装备制造技术,2022,(04):88-91.
[4]Ruzavina I, Theis L S, Lemeer J, et al. SteelBlastQC: Shot-blasted Steel Surface Dataset with Interpretable Detection of Surface Defects[J]. arXiv preprint arXiv:2504.20510, 2025.
[5]Fu J, Zhu X, Li Y. Recognition of surface defects on steel sheet using transfer learning[J]. arXiv preprint arXiv:1909.03258, 2019.
[6]Damacharla P, Rao A, Ringenberg J, et al. TLU-net: a deep learning approach for automatic steel surface defect detection[C]//2021 International Conference on Applied Artificial Intelligence (ICAPAI). IEEE, 2021: 1-6.
[7]Tan M, Le Q. Efficientnet: Rethinking model scaling for convolutional neural networks[C]//International conference on machine learning. PMLR, 2019: 6105-6114.
[8]He K, Zhang X, Ren S, et al. Deep residual learning for image recognition[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2016: 770-778.
[9]Zagoruyko S, Komodakis N. Wide residual networks[J]. arXiv preprint arXiv:1605.07146, 2016.
[10]吴禄慎,李彧雯,陈华伟,等.基于图像区域划分的轨道缺陷自动检测技术研究[J].激光与红外,2012,42(05):594-599.