Pipeline girth weld defect detection and recognition based on YOLO V3
-
摘要: 文中将end-to-end的目标识别算法YOLO V3引人到焊缝缺陷检测领域。根据焊缝缺陷的小面积且不规则特点,采用K-means算法针对焊缝缺陷库进行聚类获取新的目标候选框和GIou,将其作为目标框损失函数的2种策略改进原YOLO V3网络结构。最后在焊缝缺陷数据集上进行原YOLO V3算法、改进YOLO V3算法的对比试验。对比分析各个算法模型的训练过程中的损失值和检测过程中的均值平均精度。试验结果表明,采用2种策略改进算法相较原YOLO V3算法在收敛速度有很大提升,在管道缺陷识别效果有较好的表现,尤其在裂纹、未熔合、未焊透的类别上平均精度有较大提升。Abstract: The end-to-end target recognition algorithm YOLO V3 was introduced into the field of weld defect detection in this paper. According to characteristics of small area and irregularity of the weld defects, K-means algorithm was used for the weld defect database to cluster and obtain the new target candidate frame and GIou, which acted two strategies of the target frame loss function to improve the original YOLO V3 network structure. Finally, a comparative test of the original YOLO V3 algorithm and the improved YOLO V3 algorithm was carried out on the weld defect data set. The loss value in the training process and the mean average accuracy in the detection process of each algorithm model were compared and analyzed. The experimental results showed that compared with the original YOLO V3 algorithm, the improved algorithm with the two strategies had a great improvement in convergence speed and better performance in pipeline defect identification, especially in average accuracy for the category of crack, incomplete fusion, incomplete penetration.
-
Keywords:
- pipeline defect /
- defect detection /
- clustering /
- YOLO V3
-
-
[1] 杨锋平,许彦,曹国飞,等X80管线钢环焊缝力学性能及缺陷成因分析[J].压力容器, 2020, 37(1):68-73. [2] 胡文刚,陆云鹏,郭世雄,等.基于DR数字射线成像技术的铝合金焊缝缺陷检测[J].焊接, 2021(2):46-51. [3] Zapata J, Vilar R, Ruiz R. Performance evaluation of an automatic inspection system of weld defects in radiographic images based on neuro-classifers[J]. Expert Systems with Applications, 2011, 38(7):8812-8824.
[4] Lin B T, Huang K M. An adaptive-network-based fuzy in- ference system for predicting spring back of U-bending[J]. Transactions of the Canadian Society for Mechanical Engineering, 2013, 37(3):335-344.
[5] 谷静,王琦雯,张敏,等.基于DenseNet网络的焊缝缺陷检测识别[J].传感器与微系统, 2020, 39(9):129- 131. [6] 陈本智,方志宏,夏勇,等.基于X射线图像的厚钢管焊缝中气孔缺陷的自动检测[J].计算机应用,2017, 37(3):849-853. [7] 费凡,周永涛,周顺,等.管道环焊缝缺陷智能识别技术探究与实现[J].石油管材与仪器,2020, 6(5):1- 6. [8] 刘涵,郭润元.基于X射线图像和卷积神经网络的石油钢管焊缝缺陷检测与识别[J].仪器仪表学报, 2018, 39(4):247-256. [9] 陈立潮,解丹,张睿,等.基于渐进式卷积神经网络的焊缝缺陷识别[J].计算机工程与设计, 2020, 41(9): 2611-2615. [10] Redmon J, Divvala S, Girshick R, et al. You only look once:unified,real-time object detection[C]//2016 IEEE Conference on Computer Vision and Patterm Recognition, Las Vegas, NV, USA, 2016:779-788.
[11] Redmon J, Farhadi A. Y0L09000:better, faster, stronger [C]//2017 IEEE Conference on Computer Vision and Pattem Recognition, Honolulu, HI, USA, 2017:6517- 6525.
[12] Redmon J, Farhadi A. Y0L0 V3:an incremental improvement:arXiv:1804. 02767[R]. 2018.
[13] Lin T Y, Goyal P, Girshick R, et al. Focal loss for dense object detection[C]//2017 IEEE Intermational Conference on Computer Vision, Venice, ltaly, 2017:2999-3007.
[14] Ren S, He K, Girshick R, et al. Faster R-CNN:towards real-time object detection with region proposal net-works [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39:1137-1149..
[15] He K, Zhang X, Ren S, et al. Deep residual learning for image recognition[C]//2016 IEE Conference on Computer Vision and Pattemn Recognition, Las Vegas, NV, USA, 2016:770-778.
[16] Martinez-Zaruela M, Diaz-Pemas F J, Anton-Rodriguez M, et al. Multi-scale neural texture clasifcation using the GPU as a stream processing engine[J]. Machine Vision and Applications, 2011, 22(6):947-966.
[17] Ioffe s, Szegedy C. Batch normalzation:accelerating deep network training by reducing intermal covariate shift[J]. Proceedings of the 32nd Intemational Conference on Intemational Conference on Machine Learning, 2015, 37:448-456..
[18] Abu Alhaija H, Mustikovela S K, MeschederL, et al. Aug-mentedreality meets computer vision:efficient data generation for urban driving scenes[J]. International Joumal of Computer Vision, 2017,126(2):1-12.
[19] Rezatofighi H, Tsoi N, Gwak J Y, et al. Generalized intersection over union:a metric and a loss for bounding box regression[C]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, CA, USA, 2019:1-9.
[20] 王兵,李文璟,唐欢.改进YOL0 V3算法及其在安全帽检测中的应用[J].计算机工程与应用, 2020, 56(9): 33-40. [21] 韦若禹,李舒婷,吴松荣,等.基于改进Y0L0 V3算法的轨道扣件缺陷检测[J].铁道标准设计, 2020, 64(12):30-36.
计量
- 文章访问数: 53
- HTML全文浏览量: 3
- PDF下载量: 17
下载:
