A point cloud semantic segmentation method for nuclear power reactors based on RandLA-Net Model
doi: 10.51393/j.jamst.2023010
A point cloud semantic segmentation method for nuclear power reactors based on RandLA-Net Model
-
摘要:
The detection of foreign objects in nuclear power plant reactor is a key task in the operation and maintenance of nuclear power plants. Loose and falling foreign objects such as bolts can lead to fuel component damage and unplanned shutdown, posing serious hazards. Therefore, we propose a point cloud semantic segmentation method for foreign objects in nuclear power plant reactor based on the RandLA-Net model. Considering the correlation between point cloud collection error and curvature, the data augmentation method is improved to reduce the risk of model overfitting. By treating the boundary points of different classes as the hard examples, the hard example mining is designed to improve model generalization performance. Adding an improved test time augmentation method during model inference, the more reliable segmentation results are performed by multiple prediction on points. The experimental results indicate that the proposed method can achieve high-accuracy reactor point cloud semantic segmentation with mIoU of 0.992 and mAcc of 0.997.
Abstract:The detection of foreign objects in nuclear power plant reactor is a key task in the operation and maintenance of nuclear power plants. Loose and falling foreign objects such as bolts can lead to fuel component damage and unplanned shutdown, posing serious hazards. Therefore, we propose a point cloud semantic segmentation method for foreign objects in nuclear power plant reactor based on the RandLA-Net model. Considering the correlation between point cloud collection error and curvature, the data augmentation method is improved to reduce the risk of model overfitting. By treating the boundary points of different classes as the hard examples, the hard example mining is designed to improve model generalization performance. Adding an improved test time augmentation method during model inference, the more reliable segmentation results are performed by multiple prediction on points. The experimental results indicate that the proposed method can achieve high-accuracy reactor point cloud semantic segmentation with mIoU of 0.992 and mAcc of 0.997.
-
[1] . Ho M, Obbard E, Burr P A, et al. A review on the development of nuclear power reactors. Energy Procedia 2019;160:459–466. [2] . Tan J, Zhang Z, Zheng H, et al. Corrosion fatigue model of austenitic stainless steels used in pressurized water reactor nuclear power plants. J Nucl Mater 2020;541:152407. [3] . Zhang P, Zhong K, Li Z, J et al. High dynamic range 3D measurement based on structured light: A review. J Adv Manuf Sci Technol 2021;1(2):2021004. [4] . Xu X, Zhang L, Yang J, et al. Object detection based on fusion of sparse point cloud and image information. IEEE Trans Instrum Meas 2021;70:1–2. [5] . Wang D, Cao H. A comprehensive review on crack modeling and detection methods of aero-engine disks. J Adv Manuf Sci Technol 2022;2(3):2022012. [6] . Hou J, Ma B, Liang L, et al. An early warning method for mechanical fault detection based on adversarial auto-encoders. J Adv Manuf Sci Technol 2022;2(2):2022006. [7] . Zhang Y, Yan Z, Zhu J, et al. A review of foreign object detection (FOD) for inductive power transfer systems. eTransportation 2019;1:100002. [8] . Xu L, Song Y, Zhang W, et al. An efficient foreign objects detection network for power substation. Image Vis Comput 2021;109:104159. [9] . Zhang K, Wang W, Lv Z, et al. Computer vision detection of foreign objects in coal processing using attention CNN. Eng Appl Artif Intell 2021;102:104242. [10] . Ishii A, Mizuta T, Todo S. Detection of foreign substances mixed in a plastic bottle of medicinal solution using real-time video image processing. In: 14th International Conference on Pattern Recognition. 1998. p.1646–1650. [11] . Chao M, Kai C, Zhang Z. Research on tobacco foreign body detection device based on machine vision. Trans Inst Meas Control 2020; 42(15):2857–2871. [12] . Chen S, Jang J, Chang Y, et al. An automatic foreign matter detection and sorting system for PVC Powder. Appl Sci 2022;12(12):6276. [13] . Pan Y, Kong X, Yuan Y, et al. Detecting the foreign matter defect in lithium-ion batteries based on battery pilot manufacturing line data analyses. Energy 2023;262:125502. [14] . Lazarek J, Pryczek M. A review on point cloud semantic segmentation methods. J Appl Comput Sci 2018; 1507–0360. [15] . Deng S, Dong Q. GA-NET: Global attention network for point cloud semantic segmentation. IEEE Signal Process Lett 2021;28:1300–1304. [16] . Qi C, Yi L, Su H, et al. Pointnet++: deep hierarchical feature learning on point sets in a metric space. Adv Neural Inf Process Syst 2017;30. [17] . Wang Y, Sun Y, Liu Z, et al. Dynamic graph cnn for learning on point clouds. ACM Trans Graph 2019;38(5):1-12. [18] . Zhang Z, Hua B, Yeung S. Shellnet: efficient point cloud convolutional neural networks using concentric shells statistics. In: 2019 IEEE/CVF InTernational Conference on Computer Vision.2019. p.1607-1616. [19] . Liu J, Ni B, Li C, et al. Dynamic points agglomeration for hierarchical point sets learning. In: 2019 IEEE/CVF International Conference on Computer Vision. 2019. p.7546-7555. [20] . Zhang W, Xiao C. PCAN: 3D attention map learning using contextual information for point cloud based retrieval. In: 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2019. p.12436-12445. [21] . Wu B, Wan A, Yue X, et al. SqueezeSeg: convolutional neural nets with recurrent CRF for real-time road-object segmentation from 3D lidar point cloud. In: 2018 IEEE International Conference on Robotics and Automation. 2018. p.1887–1893. [22] . Sharma C, Kaul M. Self-supervised few-shot learning on point clouds. Adv Neural Inf Process Syst 2020;33:7212–7221. [23] . Landrieu L, Simonovsky M. Large-scale point cloud semantic segmentation with superpoint graphs. In: IEEE Conference on Computer Vision and Pattern Recognition. 2018. p.4558–4567. [24] . Hu Q, Yang B, Xie L, et al. RandLA-Net: efficient semantic segmentation of large-scale point clouds. In: IEEE Conference on Computer Vision and Pattern Recognition. 2020. p.11108–11117. [25] . Li C, Guo C, Ren W, et al. An underwater image enhancement benchmark dataset and beyond. IEEE Trans Image Process 2019;29:4376–4389. [26] . Li C, Anwar S, Porikli F. Underwater scene prior inspired deep underwater image and video enhancement. Pattern Recognit 2020;98: 107038. [27] . Xie H, Li W, Jiang C, et al. Pose error estimation using a cylinder in scanner-based robotic belt grinding. IEEE-ASME Trans Mechatron 2020;26(1):515–526. [28] . Wang G, Li W, Jiang C, et al. Trajectory planning and optimization for robotic machining based on measurement point cloud. IEEE Trans Robot 2023;28(3):1621-1637. [29] . Wang C, Zheng L, Pan S. Research on feature extraction of point cloud data based on contrastive learning. In: 2022 IEEE International Conference on Networking, Sensing and Control. 2022. p.1-5. [30] . Huang X, Qu W, Zuo Y, et al. IMFNet: interpretable multimodal fusion for point cloud registration. IEEE Robot Autom Lett 2022;7(4):12323-12330. [31] . Saiti E, Theoharis T. Multimodal registration across 3D point clouds and CT-volumes. Comput Graph 2022;106:259-266. [32] . Rosas-Cervantes V, Hoang Q, Woo S, et al. Mobile robot 3D trajectory estimation on a multilevel surface with multimodal fusion of 2D camera features and a 3D light detection and ranging point cloud. Int J Adv Robot Syst 2022;19(2):17298806221089198. -
下载:
点击查看大图
计量
- 文章访问数: 1014
- HTML全文浏览量: 455
- PDF下载量: 52
- 被引次数: 0
京公网安备 11010502036328号 京ICP备17033152号