基于改进U-Net网络的隧道裂缝分割算法研究

doi:10.3778/j.issn.1002-8331.2007-0223

摘要/Abstract

摘要：

针对复杂背景下隧道的细小裂缝图像特征难以提取以及裂缝像素类别不平衡等问题，提出了一种改进U-Net网络的隧道裂缝分割算法。将U-Net模型的编码器和解码器与残差模块相结合，使得网络参数共享，并避免出现深层网络梯度消失的问题；在此结构基础上引入挤压和激励（Squeeze and Excitation，SE）模块来提升重要特征，抑制无用特征，加强对裂缝边缘和形状等特征的权重分配；采用组合损失函数来处理裂缝像素正负样本不平衡的问题，进一步获得更加精细的分割结果。在公共隧道裂缝数据集和自制数据集上设计对比实验来验证改进模型的有效性。结果表明：该算法对裂缝的分割精度均优于其他方法，F1-Score分别达到了76.36%和75.46%，并且运行速度也有明显的提升，可以很好地满足实际工程的应用需求。

关键词: 隧道裂缝分割, U-Net网络, 残差模块, SE模块, 组合损失函数

Abstract:

Aiming at the problems of the difficult extraction of small crack image features of tunnels under complex background and the imbalance of crack pixel categories, an improved U-Net network tunnel crack segmentation algorithm is proposed. Firstly the encoder and decoder of the U-Net model are combined with the residual module to share network parameters and avoid the problem of the disappearance of deep network gradients. Secondly, based on this structure, the Squeeze and Excitation（SE） module is introduced to enhance important features, suppress useless features, and strengthen the weight distribution of crack edges, patterns and shapes. Finally, the combined loss function is used to deal with the problem of the imbalance of the crack pixel categories, and further finer segmentation results are obtained. A comparative experiment is designed on the crack data set of the public tunnel and the self-made data set to verify the effectiveness of the improved model. The results show that the algorithm is superior to other methods in segmentation accuracy. F1-Score has reached 76.36% and 75.46% respectively. The running speed has also been significantly improved, which can well meet the actual engineering application requirements.

Key words: tunnel crack segmentation, U-Net network, residual module, SE module, combined loss function

常惠，饶志强，赵玉林，李益晨. 基于改进U-Net网络的隧道裂缝分割算法研究[J]. 计算机工程与应用, 2021, 57(22): 215-222.

CHANG Hui, RAO Zhiqiang, ZHAO Yulin, LI Yichen. Research on Tunnel Crack Segmentation Algorithm Based on Improved U-Net Network[J]. Computer Engineering and Applications, 2021, 57(22): 215-222.

参考文献

[1] SALMAN M，MATHAVAN S，KAMAL K，et al.Pavement crack detection using the Gabor filter[C]//16th International IEEE Conference on Intelligent Transportation Systems，2013：2039-2044.
[2] SHI Y，CUI L，QI Z，et al.Automatic road crack detection using random structured forests[J].IEEE Transactions on Intelligent Transportation Systems，2016，17（12）：3434-3445.
[3] NAGARAJAIAH S，YANG Y.Modeling and harnessing sparse and low rank data structure：a new paradigm for structural dynamics，identification，damage detection，and health monitoring[J].Structural Control & Health Monitoring，2017，24（1）：256-278.
[4] DAI B，GU C，ZHAO E，et al.Improved online sequential extreme learning machine for identifying crack behavior in concrete dam[J].Advances in Structural Engineering，2019，22（2）：402-412.
[5] LI G，ZHAO X，DU K，et al.Recognition and evaluation of bridge cracks with modified active contour model and greedy search-based support vector machine[J].Automation in Construction，2017，78：51-61.
[6] 柴雪松，朱兴永，李健超，等.基于深度卷积神经网络的隧道衬砌裂缝识别算法[J].铁道建筑，2018（6）：65-70.
CHAI X S，ZHU X Y，LI J C，et al.Tunnel lining crack identification algorithm based on deep convolutional neural network[J].Railway Construction，2018（6）：65-70.
[7] 刘新根，陈莹莹，朱爱玺，等.基于深度学习的隧道裂缝识别方法[J].广西大学学报（自然科学版），2018（6）：2243-2251.
LIU X G，CHEN Y Y，ZHU A X，et al.Tunnel crack identification based on deep learning[J].Journal of Guangxi University（Natural Science Edition），2018（6）：2243-2251.
[8] CHENG J，XIONG W，CHEN W，et al.Pixel-level crack detection using U-Net[C]//2018 IEEE Region 10 Conference，2018（8）：462-466.
[9] 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（6）：1-8.
ZHU S Y，DU J C，LI Y S，et al.Method for bridge crack detection based on the U-Net convolutional networks[J].Journal of Xidian University，2019，46（6）：1-8.
[10] LAU S L，WANG X，YANG X，et al.Automated pavement crack segmentation using fully convolutional U-Net with a pretrained ResNet-34 encoder[J].arXiv：2001.01912，2020.
[11] 徐国整，廖晨聪，陈锦剑，等.基于HU-ResNet的混凝土表观裂缝信息提取[J].计算机工程，2020，46（11）：279-285.
XU G Z，LIAO C C，CHEN J J，et al.Extraction of concrete apparent crack imformation based on HU-ResNet[J].Computer Engineering，2020，46（11）：279-285.
[12] LIU Z，CAO Y，WANG Y，et al.Computer vision-based concrete crack detection using U-Net fully convolutional networks[J].Automation in Construction，2019，104：129-139.
[13] 翁飘.复杂背景下路面裂缝分割算法研究[D].郑州：郑州大学，2019.
WENG P.Research on pavement crack segmentation algorithm in complex environment[D].Zhengzhou：Zhengzhou University，2019.
[14] ISLAM M M，KIM J M.Vision-based autonomous crack detection of concrete structures using a fully convolutional encoder-decoder network[J].Sensors，2019，19（1）：4251.
[15] SCHMUGGE S J，RICE L，LINDBERG J，et al.Crack segmentation by leveraging multiple frames of varying illumination[C]//2017 IEEE Winter Conference on Applications of Computer Vision，2017：1045-1053.
[16] REN Y，HUANG J，HONG Z，et al.Image-based concrete crack detection in tunnels using deep fully convolutional networks[J].Construction and Building Materials，2020，234：117-127.
[17] HU J，SHEN L，ALBANIE S，et al.Squeeze-and-excitation networks[C]//2018 IEEE Conference on Computer Vision and Pattern Recognition，2018：7132-7141.
[18] 顾书豪，李小霞，王学渊，等.增强语义信息与多通道特征融合的裂缝检测[J].计算机工程与应用，2021，57（10）：204-210.
GU S H，LI X X，WANG X Y，et al.Pavement crack detection based on enhanced semantic information and multi-channel feature fusion[J].Computer Engineering and Applications，2021，57（10）：204-210.