计算机应用 ›› 2020, Vol. 40 ›› Issue (4): 1056-1061.DOI: 10.11772/j.issn.1001-9081.2019091546

• 人工智能 • 上一篇    下一篇

基于卷积神经网络的桥梁裂缝识别和测量方法

梁雪慧1,2, 程云泽2, 张瑞杰2, 赵菲2   

  1. 1. 天津市复杂系统控制理论与应用重点实验室(天津理工大学), 天津 300384;
    2. 天津理工大学 电气电子工程学院, 天津 300384
  • 收稿日期:2019-09-09 修回日期:2019-11-01 出版日期:2020-04-10 发布日期:2019-11-18
  • 通讯作者: 程云泽
  • 作者简介:梁雪慧(1970-),女,河北藁城人,副教授,硕士,主要研究方向:智能控制;程云泽(1996-),男,河北唐山人,硕士研究生,主要研究方向:深度学习、图像处理;张瑞杰(1993-),男,甘肃金昌人,硕士研究生,主要研究方向:智能控制;赵菲(1994-),女,河北石家庄人,硕士研究生,主要研究方向:深度学习、图像处理。
  • 基金资助:
    天津市科技特派员项目(14JCTPJC00510)。

Bridge crack classification and measurement method based on deep convolutional neural network

LIANG Xuehui1,2, CHENG Yunze2, ZHANG Ruijie2, ZHAO Fei2   

  1. 1. Tianjin Key Laboratory for Control Theory and Application in Complicated Systems(Tianjin University of Technology), Tianjin 300384, China;
    2. School of Electrical and Electronic Engineering, Tianjin University of Technology, Tianjin 300384, China
  • Received:2019-09-09 Revised:2019-11-01 Online:2020-04-10 Published:2019-11-18
  • Supported by:
    This work is partially supported by the Tianjin Science and Technology Commissioner Project(14JCTPJC00510).

摘要: 为了提高桥梁裂缝检测水平,解决目前手工检测费时费力和传统图像处理方法需要人工设定参数的问题,提出一种基于改进GoogLeNet的桥梁裂缝检测算法。首先,构建了一个较大规模的桥梁裂缝数据集RLH(Retinex-Laplace-Histogram equalization)用于模型的训练和测试。其次,基于原始的GoogLeNet模型,采用归一化的卷积核改进了inception模块,采用三种改进方案修改网络开头,去掉第七个及以后的inception层,建立桥梁裂缝特征图像分类系统。最后,利用滑动窗口精准定位裂缝并结合骨架提取算法计算裂缝的长度和宽度。实验结果表明,改进的GoogLeNet网络与原始GoogLeNet网络相比,识别准确率提升了3.13%,训练时间减少为原来的64.6%。另外,骨架提取算法能够考虑裂缝的走势,计算宽度更加准确,且最大宽度和平均宽度都能计算。综上所述,所提分类和测量方法具有准确度高、速度快、定位准确、测量准确等特点。

关键词: 图像处理, 深度学习, GoogLeNet深度卷积模型, 骨架提取算法, 桥梁裂缝

Abstract: In order to improve the detection level of bridge cracks,and solve the time-consuming and laborious problem in manual detection and the parameters to be set manually in traditional image processing methods,an improved bridge crack detection algorithm was proposed based on GoogLeNet. Firstly,a large-scale bridge crack Retinex-Laplace-Histogram equalization(RLH)dataset was constructed for model training and testing. Secondly,based on the original GoogLeNet model,the inception module was improved by using the normalized convolution kernel,three improved schemes were used to modify the beginning of the network,the seventh and later inception layers were removed,and a bridge crack feature image classification system was established. Finally,the sliding window was used to accurately locate the cracks and the lengths and widths of the cracks were calculated by the skeleton extraction algorithm. The experimental results show that compared with the original GoogLeNet network,the improve-GoogLeNet network increased the recognition accuracy by 3. 13%, and decreased the training time to the 64. 6% of the original one. In addition,the skeleton extraction algorithm can consider the trend of the crack,calculate the width more accurately,and the maximum width and the average width can be calculated. In summary,the classification and measurement method proposed in this paper have the characteristics of high accuracy,fast speed,accurate positioning and accurate measurement.

Key words: image processing, deep learning, GoogLeNet deep convolution model, skeleton extraction algorithm, bridge creak

中图分类号: