复杂场景下多类型路面病害分割方法

doi:10.13474/j.cnki.11-2246.2023.0360

摘要/Abstract

摘要： 多类型路面病害的自动分割对于提升公路养护管理水平具有重要意义。大规模、多场景、多类型路面病害训练数据的缺乏,降低了深度学习模型在复杂场景下的泛化能力,限制了路面病害提取算法工程化应用的发展。为此,本文收集并建立了一个用于多类型路面病害分割的数据集CPCD,旨在从CCD图像中分割出高精度的路面病害。新数据集涵盖了7种常见路面病害类型,以及带有纹理相似噪声的负样本,合计6967张。在此基础上,提出了一种基于注意力机制与高分辨率网络相融合的病害分割卷积神经网络CBAM-HRNet,用于提升细小病害分割的精度。试验结果表明,本文模型在CPCD测试集上的F1、mIoU分别为91.30%、84.64%,其表现明显优于其他串联结构的主流卷积神经网络模型。研究成果将为我国公路自动化检测研究和可持续发展提供一定的支撑。

关键词: 深度学习, 复杂场景, CPCD, 语义分割, 路面病害

Abstract: Automatic segmentation of various types of pavement distresss is of great significance for improving the level of highway maintenance management. The lack of large-scale, diverse-scenario, and multi-type training data for pavement distresss has reduced the generalization ability of deep learning models in complex scenarios, and limited the development of engineering applications for pavement distress extraction algorithms. To address this issue, this paper collects and establishes a dataset called CPCD for the segmentation of multiple types of pavement distresss, aiming to accurately segment pavement distresss from CCD images. The new dataset covers 7 common types of pavement distresss, including negative samples with texture-similar noise, totaling 6967 images. Based on this dataset, a distress segmentation convolutional neural network called CBAM-HRNet, which combines attention mechanisms with high-resolution networks is proposed to improve the accuracy of fine-grained distress segmentation. Experimental results show that the F1 score and mIoU of the proposed model on the CPCD dataset are 91.30% and 84.64% respectively. Its performance is significantly better than other mainstream convolutional neural network models with sequential structures. The research findings will provide certain support for the research on highway automation detection and sustainable development in our country.

Key words: deep learning, complex scenarios, CPCD, semantic segmentation, pavement distresss

中图分类号:

P258

张在岩, 宋伟东, 陈兆雪. 复杂场景下多类型路面病害分割方法[J]. 测绘通报, 2023, 0(12): 63-69.

ZHANG Zaiyan, SONG Weidong, CHEN Zhaoxue. Multi-type pavement distress segmentation methods in complex scenarios[J]. Bulletin of Surveying and Mapping, 2023, 0(12): 63-69.

参考文献

[1] 经济日报.《“十四五”公路养护管理发展纲要》印发—发挥公路资产最大功用[EB/OL].2022-05-11.https://www.gov.cn/zhengce/2022-05/11/content_5689580.
[2] 晏班夫,徐观亚,栾健,等. 基于Faster R-CNN与形态法的路面病害识别[J]. 中国公路学报,2021,34(9): 181-193.
[3] SONG Weidong,JIA Guohui,JIA Di,et al. Automatic pavement crack detection and classification using multiscale feature attention network[J]. IEEE Access,2019,7: 171001-171012.
[4] MAJIDIFARD H,ADU-GYAMFI Y,BUTTLAR W G. Deep machine learning approach to develop a new asphalt pavement condition index[J]. Construction and Building Materials,2020,247: 118513.
[5] STRICKER R,AGANIAN D,SESSELMANN M,et al. Road surface segmentation-pixel-perfect distress and object detection for road assessment[C]//Proceedings of 2021 IEEE International Conference on Automation Science and Engineering (CASE). Lyon: IEEE,2021: 1789-1796.
[6] ZHANG Ce,NATEGHINIA E,MIRANDA-MORENO L F,et al. Pavement distress detection using convolutional neural network (CNN): a case study in Montreal,Canada[J]. International Journal of Transportation Science and Technology,2022,11(2): 298-309.
[7] XU Zhengsen,GUAN Haiyan,KANG Jian,et al. Pavement crack detection from CCD images with a locally enhanced transformer network[J]. International Journal of Applied Earth Observation and Geoinformation,2022,110: 102825.
[8] SHI Yong,CUI Limeng,QI Zhiquan,et al. Automatic road crack detection using random structured forests[J]. IEEE Transactions on Intelligent Transportation Systems,2016,17(12): 3434-3445.
[9] YANG Fan,ZHANG Lei,YU Sijia,et al. Feature pyramid and hierarchical boosting network for pavement crack detection[J]. IEEE Transactions on Intelligent Transportation Systems,2020,21(4): 1525-1535.
[10] CHAMBON S,MOLIARD J M. Automatic road pavement assessment with image processing: review and comparison[J]. International Journal of Geophysics,2011,2011: 1-20.
[11] ZOU Qin,ZHANG Zheng,LI Qingquan,et al. DeepCrack: learning hierarchical convolutional features for crack detection[J]. IEEE Transactions on Image Processing,2019,28(3): 1498-1512.
[12] 翁飘,陆彦辉,齐宪标,等. 基于改进的全卷积神经网络的路面裂缝分割技术[J]. 计算机工程与应用,2019,55(16): 235-239.
[13] 朱苏雅,杜建超,李云松,等. 采用U-Net卷积网络的桥梁裂缝检测方法[J]. 西安电子科技大学学报,2019,46(4): 35-42.
[14] 张志华,邓砚学,张新秀. 基于改进SegNet的沥青路面病害提取与分类方法[J]. 交通信息与安全,2022,40(3): 127-135.
[15] SUN Ke,XIAO Bin,LIU Dong,et al. Deep high-resolution representation learning for human pose estimation[C]//Proceedings of 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).Long Beach,CA:IEEE,2019:5686-5696.
[16] PERJU V,CASASENT D,MARDARE I. Image complexity matrix for pattern and target recognition based on Fourier spectrum analysis[C]// Proceedings of 2009 SPIE Defense,Security and Sensing. Orlando,Florida: 2009,7340: 209-217.
[17] RONNEBERGER O,FISCHER P,BROX T. U-net: convolutional networks for biomedical image segmentation[M]//Lecture Notes in Computer Science. Cham: Springer International Publishing,2015: 234-241.
[18] ZHAO Hengshuang,SHI Jianping,QI Xiaojuan,et al. Pyramid scene parsing network[C]// Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Honolulu,HI: IEEE,2017: 6230-6239.
[19] CHEN L C,ZHU Yukun,PAPANDREOU G,et al. Encoder-decoder with atrous separable convolution for semantic image segmentation[M]//Computer Vision - ECCV 2018. Cham: Springer International Publishing,2018: 833-851.