基于改进Mask R-CNN和LSD的图纸管道检测方法

doi:10.3969/j.issn.1006-2475.2024.10.007

摘要/Abstract

摘要： 针对核电轴测视图中因管道特征不明显、管道尺度差异大及管道相交导致的管道识别精度差及错检、漏检等问题，提出一种基于改进的Mask R-CNN及LSD方法检测图纸中的管道。首先，针对管道特征不明显问题，将识别目标由管道调整为管道及其尺寸标注线，增加目标几何特征；其次，改进Mask R-CNN网络，引入双向加权特征金字塔结构，提高不同尺度目标特征提取能力，将原非极大值抑制改为DIoU-NMS，提高相交管道识别精度；最后，通过LSD算法检测出目标图像中的直线，再经过条件约束筛选及最小二乘拟合得到管道直线，实现管道的准确检测。实验结果表明，改进的Mask R-CNN算法有效减少了管道错检、漏检问题，精度可达90.04%，结合LSD直线检测、条件约束及最小二乘拟合算法得到管道直线，满足图纸管道检测的要求。

关键词: 管道检测, 实例分割, Mask R-CNN算法, LSD, 条件约束

Abstract: Aiming at the problems of poor precision of pipelines detection， false detection and missed detection caused by indistinct pipelines features， large differences in pipeline scales and pipeling intersections in the nuclear power axonometric drawings，an method for pipelines detection based on improved Mask R-CNN and LSD is proposed. Firstly， aiming at the problems of indistinct pipelines features， the recognition target is adjusted from the pipelines to the pipelines and its dimensioned lines. The recognition target geometry features are added. Secondly， the Mask R-CNN network is improved， and the BiFPN structure is used to enhance the ability to extract target features at different scales. We change the original NMS to DIoU-NMS to improve the accuracy of intersecting pipelines detection. Finally， the LSD algorithm is used to detect the lines in the target image， and then the pipeline lines are obtained by conditional constraint filtering and least square fitting. The experimental results show that the improved Mask R-CNN algorithm can well solve the problems of missed detection and false detection， and its accuracy recognition rate reaches 90.04%. Combining LSD line detection， conditional constraint， and least squares fitting algorithm， pipeline lines are obtained， which meets the requirements of pipelines detection in the drawings.

Key words: , pipelines detection, instance segmentation, Mask R-CNN algorithm, LSD, conditional constraint

中图分类号:

TU17

黄杉杉1, 吴巍2, 徐雨晴1, 魏婕1. 基于改进Mask R-CNN和LSD的图纸管道检测方法[J]. 计算机与现代化, 2024, 0(10): 42-48.

HUANG Shanshan1, WU Wei2, XU Yuqing1, WEI Jie1. Pipelines in Drawings Detection Method Based on Improved Mask R-CNN and LSD [J]. Computer and Modernization, 2024, 0(10): 42-48.

参考文献

［1］徐建军，王浩，张翼. 核电厂全生命周期数字化转型研究与设计［J］. 仪器仪表用户，2022，29（8）：63-68.
［2］钟方慧，刘勇，黄岁锋，等. 基于BIM技术的3D作业指导书创新与应用［J］. 中国核电，2022，15（6）：842-848.
［3］ MOLOO R K， DAWOOD M A S， AULEEAR A S. 3-phase recognition approach to pseudo 3D building generation from 2D floor plan［J］. International Journal of Computer Graphics & Animation， 2011，1（2）：13-27.
［4］ ZHAO Y F， DENG X Y， LAI H H. A YOLO-based method to recognize structural components from 2D drawings［C］// 2020 ASCE Construction Research Congress. ASCE， 2020. DOI：10.1061/9780784482865.080.
［5］ 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 Pattern Re-cognition（CVPR）. IEEE， 2016：779-788.
［6］杨墨逸，赵云凡，邓雪原. 基于改进Cascade RCNN的风管平面二维图纸识别［J］. 土木工程与管理学报， 2022，39（4）：114-123
［7］ CAI Z W， VASCONCELOS N. Cascade R-CNN： Delving into high quality object detection［C］// 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. IEEE， 2018：6154-6162.
［8］ T/GDASE 0004—2020，工业管道检验与安全评价用轴测图编制规范［S］. 广州：广东省特种设备行业协会， 2020.
［9］ GB/T 18229—2000， CAD工程制图规则［S］. 北京：中国质检出版社， 2000.
［10］张红颖，王徐泳，彭晓雯. 结合前景分割的多特征融合行人重识别［J］. 中国图象图形学报， 2023，28（5）：1360-1371.
［11］ REDMON J，FARHADI A. YOLO9000：Better，faster，stronger［C］// 2017 IEEE Conference on Computer Vision and Pattern Recognition（CVPR）. IEEE， 2017：6517-6525.
［12］ REDMON J ， FARHADI A. YOLOv3： An incremental improvement［C］// 2018 Conference on Computer Vision and Pattern Recognition. IEEE， 2018. DOI: 10.48550/arXiv.18
04.02767.
［13］ ZHU X K， LYU S C， WANG X， et al. TPH-YOLOv5： Improved YOLOv5 based on transformer prediction head for object detection on drone-captured scenarios［C］// Proceedings of the 2021 IEEE/CVF International Conference on Computer Vision Workshops. IEEE， 2021：2778-2788.
［14］ LIU W， ANGUELOV D， ERHAN D， et al. SSD： Single shot multiBox detector［C］// Computer Vision-ECCV 2016：14th European Conference. Springer， 2016：21-37.
［15］ GIRSHICK R，DONAHUE J，DARRELL T，et al. Rich feature hierarchies for accurate object detection and semantic segmentation［C］// 2014 IEEE Conference on Computer Vision and Pattern Recognition. IEEE， 2014：580-587.
［16］ GIRSHICK R. Fast R-CNN［C］// 2015 IEEE International Conference on Computer Vision（ICCV）. IEEE， 2015：1440-1448.
［17］ REN S Q， HE K M， GIRSHICK R， et al. Faster R-CNN： Towards real-time object detection with region proposal networks［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence， 2017，39（6）：1137-1149.
［18］ HE K M，GKIOXARI G，DOLLAR P，et al. Mask R-CNN［C］// 2017 IEEE International Conference on Computer Vision（ICCV）. IEEE， 2017：2980-2988.
［19］ TAN M X， LE Q V. EfficientNet： Rethinking model scaling for convolutional neural networks［C］// Proceedings of the 2019 International Conference on Machine Learning. PMLR， 2019：6105-6114.
［20］ ZHENG Z H， WANG P， LIU W， et al. Distance IoU loss：Faster and better learning for bounding box regression［C］// Proceedings of the 2020 AAAI Conference on Artificial Intelligence. AAAI， 2020，34（7）：12993-13000.
［21］ AGGARWAL N， KARL W C. Line detection in images through regularized Hough transform［J］. IEEE Transactions on Image Processing， 2006，15（3）：582-591.
［22］ VON GIOI R G， JAKUBOWICZ J， MOREL J M， et al. LSD： A line segment detector［J］. Image Processing on Line， 2012，2：35-55.
［23］李瑾，范佳能，刘屹然. 复杂环境基于YOLACT电缆识别与定位［J］. 电子测量技术， 2023，46（4）：114-120.
［24］石孟曦，张文红，张磊，等. 基于LSD和改进YOLOv2的压力表识读方法［J］. 计算机仿真， 2023，40（5）：284-288.
［25］ BOLYA D， ZHOU C， XIAO F Y， et al. YOLACT： Real-time instance segmentation［C］// Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision. IEEE， 2019：9156-9165.
［26］ BOLYA D， ZHOU C， XIAO F Y， et al. YOLACT++ bett-er real-time instance segmentation［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence， 2022，44（2）：1108-1121.
［27］ WANG X L， KONG T， SHEN C H， et al. SOLO： Segmenting objects by locations［C］// 2020 European Conference on Computer Vision. Springer， 2020：649-665.
［28］ WANG X L， ZHANG R F， SHEN C H， et al. SOLO： A simple framework for instance segmentation［J］. IEEE Transactions on Pattern Analysis & Machine Intelligence， 2022，44（11）： 8587-8601.
［29］ LIN T Y， DOLLÁR P， GIRSHICK R， et al. Feature pyramid networks for object detection［C］// Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition. IEEE， 2017：936-944.
［30］ ZHAO Y F，DENG X Y，LAI H H. Reconstructing BIM from 2D structural drawings for existing buildings［J］. Automation in Construction， 2021，128.DOI： 10.1016/j.autcon.2021.103750.