[1]陈婷婷.U-Net的桥梁裂缝智能检测方法改进研究[J].大众科技,2023,25(1):18-21.
　Research on Improvement U-Net Intelligent Detection Method for Bridge Cracks[J].Popular Science & Technology,2023,25(1):18-21.
点击复制

U-Net的桥梁裂缝智能检测方法改进研究()
分享到：

《大众科技》[ISSN:1008-1151/CN:45-1235/N]

卷:
25
期数:
2023年1
页码:
18-21
栏目:
信息技术与通信
出版日期:
2023-01-20

文章信息/Info

Title:
Research on Improvement U-Net Intelligent Detection Method for Bridge Cracks
作者:
陈婷婷　
（长安大学工程机械学院，陕西 西安 710000）
关键词:
U-Net; 桥梁裂缝检测; 神经网络; 图像分割
Keywords:
U-Net bridge crack detection neural network image segmentation
文献标志码:
A
摘要:
针对桥梁裂缝检测准确率与精度有待提高的问题，提出了基于图像分割技术的U-Net网络进行桥梁裂缝检测。采用西安市的桥梁裂缝数据集，通过人工标定完成了对数据集的标注，用于进行监督学习。针对这一数据集采用了数据增强、空洞卷积、批次归一化等方法提高识别精度，减少过拟合现象；将交叉熵损失与Dice损失相结合，提高了模型训练的速度以及识别精度。与广泛应用的图像分割方法进行比较，实验结果表明该模型在桥梁裂缝数据集上的分割表现结果具有优越性。
Abstract:
Aiming at the problem that the accuracy and precision of bridge crack detection need to be improved, a U-Net network based on image segmentation technology is proposed for bridge crack detection. The bridge crack data set in Xian city is used to mark the data set through manual calibration for supervision and learning. For this data set, data enhancement, void convolution, batch normalization and other methods are used to improve the recognition accuracy and reduce the over fitting phenomenon the cross entropy loss and Dice loss are combined to improve the speed of model training and the recognition accuracy. Compared with the widely used image segmentation methods, the experimental results show that the model has advantages in the segmentation performance of the bridge crack dataset.

参考文献/References:

[1] 郭素明. 道路桥梁养护中常见病害与维护方法探析[J]. 工程技术研究，2018(15): 247-248. [2] LONG J, SHWLHAMER E, DARRELL T. Fully convolutional networks for semantic segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 39(4): 640-651. [3] RONNEBERGER O, FISCHER P, BROX T. U-Net: Convolutional networks for biomedical image segmentation[C]. International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer International Publishing, 2015: 234-241. [4] CHEN L C, ZHU Y , PAPANDREOU G, et al. Encoder-decoder with atrous separable convolution for semantic image segmentation[C]. Computer Vision-ECCV 2018 15th European Conference, Munich, Germany, 2018: 833-851. [5] YU F, KOLTUN V. Multi-scale context aggregation by dilated convolutions[J]. Computer Science, 2016, 2016: 17127188. [6] MILLETARI F , NAVAB N, AHMADI S A. V-Net: Fully convolutional neural networks for volumetric medical image segmentation[J]. IEEE, 2016, 2016: 4797.

备注/Memo

备注/Memo:
【收稿日期】2022-09-10 【作者简介】陈婷婷（1999－），女，长安大学工程机械学院在读硕士研究生，研究方向为机器学习、图像处理。
更新日期/Last Update: 2023-03-30