基于智能型面分析的抛光表面缺陷检测研究  

Polished Surface Defect Detection Based on Intelligent Surface Analysis

作　　者：李子豪 房丰洲[1,2] 任仲贺 侯高峰 Li Zihao;Fang Fengzhou;Ren Zhonghe;Hou Gaofeng(State Key Laboratory of Precision Measuring Technology and Instrument,Tianjin University,Tianjin 300072,China;Labotatory of Micro/Nano Manufacturing Technology(MNMT),Tianjin 300072,China)

机构地区：[1]天津大学精密测试技术及仪器国家重点实验室,天津300072 [2]微纳制造实验室,天津300072

出　　处：《激光与光电子学进展》2023年第24期202-213,共12页Laser & Optoelectronics Progress

基　　金：国家自然科学基金(52035009)。

摘　　要：工件的表面质量对零件可靠性、质量和使用寿命的影响至关重要。尽管各种基于计算机视觉的目标检测框架已经被广泛应用于工业表面缺陷检测场景,但由于面型的影响以及缺陷之间的混叠性,超精加工工件表面缺陷检测仍然具有挑战性。因此,提出了一种频率嵌入双分支参数预测网络来预测滤波参数,滤除掉型面信息从而使得缺陷特征更加显著。基于智能型面分析的预处理后,提出了一种基于级联区域神经网络感受野增强缺陷检测网络,将可变形卷积间隔地替换到高效网络的卷积模块中,有效地提高了主干网络特征提取的能力,然后重新选择特征图组成新的特征金字塔网络以提高效率,进一步提高网络性能。此外,还构建了具有滤波参数标注信息的滤波参数数据集UPP-CLS和具有缺陷类别及位置的缺陷检测数据集UPP-DET。模型在UPP-CLS上达到了85.36%的准确性,相较于现有网络提升3~5个百分点;在UPP-DET上达到了0.862的平均精度,相较于现有网络提升5.3%~7.8%。模型整体性能优于主流网络结构。源代码将在https://gitee.com/zihaodl/detect_app上开源。Surface quality of the workpiece is critical for part reliability,quality and service life.Although various visionbased target detection frameworks have been widely applied to industrial surface defect detection scenarios,surface defect detection of ultra-precise machining workpieces is still challenging due to the influence of face shape and the confounding nature between defects.Therefore,we propose a frequency-embedded two-branch parametric prediction network to predict the filtering parameters and filter out the profile information to make the defect features more significant.After preprocessing based on intelligent type surface analysis,a cascaded regional neural network-based perceptual field enhancement defect detection network is proposed.It replaces the deformable convolution intervals into the convolution module of the EfficientNet,which effectively improves the feature extraction capability of the backbone network.Then,the feature map is reselected to form a new feature pyramid network to improve the efficiency and further improve the network performance.In addition,the filter parameter dataset ultra precision polishing(UPP-CLS)with filter parameter labelling information and the defect detection dataset UPP-DET with defect category and location are constructed.The model achieves 85.36%accuracy on UPP-CLS,which is 3 to 5 percentage points higher than that of the existing networks,and 0.862 average precision on UPP-DET,which is 5.3%‒7.8%higher than that of the existing networks.The overall performance of the model is better than the mainstream network architecture.The source code and dataset will be available at https://gitee.com/zihaodl/detect_app.

关 键 词：超精密加工 计算机视觉 缺陷检测 型面分析 

分 类 号：O436[机械工程—光学工程]