基于深度学习的表面缺陷检测算法文献研究

打开文本图片集

中图分类号：TP391文献标识码：A

Literature Research on Deep Learning-based Algorithms for Surface Defect Detection

HUANG Keteng1，WANG Yuqi1，WANG Qing1， JU Junwei²，BAI Shuowei1 （1. 266O71，China; 2. （Shong） ）

Abstract： Surface defect detection is a key aspect quality inspection industrial components. Aiming at the lack systematic literature research on surface defect detection algorithms for industrial parts，China National Knowledge Infrastructure （CNKI） WOS（Web Science） core ensemble databases are selected as data sources between 2Ol7 2O23. With the help CiteSpace visual analysis stware，the research line surface defect detection algorithms in the field industrial components inspection is analysed by the number annual publications keyword clustering. The current state research on deep learning-based algorithms for detecting surface defects on industrial components is systematically presented，as well as the practical applications single-stage two-stage target detection algorithms. The key problems current surface defect detection algorithms for industrial components the corresponding solution strategies are summarized. The future development surface defect detection algorithms for in

dustrial components is also discussed.

Keywords： deep learning; surface defect detection; industrial components; visualization analysis; CiteSpace

在工业零件的生产过程中，由于加工工艺、生产原材料或生产环境等多因素的影响，零件表面会出现划痕、压伤、黑皮以及凸起等缺陷[]，不仅影响零件的整体性能和质量，还对其使用寿命造成不可逆的损害，甚至可能引发生产事故，给企业带来难以估量的损失。（剩余13554字）