Recognition, location, measurement, and 3D reconstruction of concealed cracks using convolutional neural networks

Abstract Concealed cracks in asphalt pavement are the cracks that originate below the surface of the pavement. These cracks are a major contributing factor to pavement damage, in addition to being a major contributing factor to the formation of reflection cracks. The detection of a concealed crack is considered challenging because the location of the crack is, by definition, difficult to find. Therefore, the research on the utilization of ground penetrating radar (GPR) to locate concealed cracks has gained significant interest in recent years. However, the manually processed GPR image used for the recognition, location, and measurement of concealed cracks is inefficient and inaccurate. This project presents an application of convolutional neural networks (CNNs) to GPR images that automatically recognizes, locates, measures, and produces a 3D reconstruction of concealed cracks. In this project, three different CNNs (recognition, location, and feature extraction) were established to accomplish the aforementioned tasks automatically. Each CNN is developed through processes of structural design, training, and testing. The recognition CNN was designed to distinguish concealed cracks from other types of damage in a GPR image, the location CNN determined the location and length measurement of concealed crack images based on the results provided by the recognition CNN, and crack feature points were extracted by the feature extraction CNN to establish the 3D reconstruction models of the concealed cracks. The 3D reconstruction models were then used to calculate crack volume and predict the growth tendency of cracks. The results indicated that the recognition CNN is able to distinguish concealed cracks from other types of damages in 6482 GPR images with zero errors. In addition, the length recognition results calculated from the location CNN possess a 0.2543 cm mean squared error, a 0.978 cm maximum length error, and a 0.504 cm average error in the test samples. Meanwhile, the feature extraction CNN is able to provide feature points for a 3D reconstruction model. The results of this study suggest that the CNNs could be accurately used for the recognition, location, and 3D reconstruction of concealed cracks in asphalt pavement in real-world applications.

[1]  Zhang Xiong,et al.  3D object retrieval with stacked local convolutional autoencoder , 2015, Signal Process..

[2]  Qingshan Liu,et al.  FaceHunter: A multi-task convolutional neural network based face detector , 2016, Signal Process. Image Commun..

[3]  Iveta Mrázová,et al.  Can Deep Neural Networks Discover Meaningful Pattern Features? , 2012, Complex Adaptive Systems.

[4]  Yuan Dong,et al.  Automatic age estimation based on deep learning algorithm , 2016, Neurocomputing.

[5]  Xiang Bai,et al.  Script identification in the wild via discriminative convolutional neural network , 2016, Pattern Recognit..

[6]  Imad L. Al-Qadi,et al.  Development of an analytic approach utilizing the extended common midpoint method to estimate asphalt pavement thickness with 3-D ground-penetrating radar , 2016 .

[7]  Weiguang Zhang,et al.  A statistical based framework for predicting field cracking performance of asphalt pavements: Application to top-down cracking prediction , 2016 .

[8]  Serge Andrianov,et al.  Comparison of Regularization Methods for ImageNet Classification with Deep Convolutional Neural Networks , 2014 .

[9]  Hui Luo,et al.  The feasibility of continuous construction of the base and asphalt layers of asphalt pavement to solve the problem of reflective cracks , 2016 .

[10]  Xiaogang Wang,et al.  Deep Convolutional Network Cascade for Facial Point Detection , 2013, 2013 IEEE Conference on Computer Vision and Pattern Recognition.

[11]  Li Xiu,et al.  Practical methods for detection of concealed cracks in highway pavement using ground penetration radar data , 2007 .

[12]  Chalavadi Krishna Mohan,et al.  Human action recognition using genetic algorithms and convolutional neural networks , 2016, Pattern Recognit..

[13]  Edward K. Wong,et al.  Dual many-to-one-encoder-based transfer learning for cross-dataset human action recognition , 2016, Image Vis. Comput..

[14]  Ching Y. Suen,et al.  Detecting predatory conversations in social media by deep Convolutional Neural Networks , 2016, Digit. Investig..

[15]  Cécile Barat,et al.  String representations and distances in deep Convolutional Neural Networks for image classification , 2016, Pattern Recognit..

[16]  Motoyuki Sato,et al.  in situ measurement of pavement thickness and dielectric permittivity by GPR using an antenna array , 2014 .

[17]  Yuan Dong,et al.  Adaptive Cascade Deep Convolutional Neural Networks for face alignment , 2015, Comput. Stand. Interfaces.

[18]  P. Szymczyk,et al.  Non-destructive building investigation through analysis of GPR signal by S-transform , 2015 .

[19]  Yan Wang,et al.  Dense crowd counting from still images with convolutional neural networks , 2016, J. Vis. Commun. Image Represent..

[20]  Anant Madabhushi,et al.  A Deep Convolutional Neural Network for segmenting and classifying epithelial and stromal regions in histopathological images , 2016, Neurocomputing.

[21]  V. Pérez-Gracia,et al.  GPR evaluation of the damage found in the reinforced concrete base of a block of flats: A case study , 2008 .

[22]  Wei Shen,et al.  Spatial-temporal convolutional neural networks for anomaly detection and localization in crowded scenes , 2016, Signal Process. Image Commun..

[23]  Higinio González-Jorge,et al.  Approach to identify cracking in asphalt pavement using GPR and infrared thermographic methods: Preliminary findings , 2014 .

[24]  Miguel Ángel Guevara-López,et al.  Representation learning for mammography mass lesion classification with convolutional neural networks , 2016, Comput. Methods Programs Biomed..

[25]  YuanFang,et al.  Integration of image quality and motion cues for face anti-spoofing , 2016 .

[26]  Daniel Dajun Zeng,et al.  Mining opinion summarizations using convolutional neural networks in Chinese microblogging systems , 2016, Knowl. Based Syst..

[27]  Mônica Batista Leite,et al.  Prediction of compressive strength of concrete containing construction and demolition waste using artificial neural networks , 2013 .

[28]  Sang Woo Kim,et al.  Localization of the slab information in factory scenes using deep convolutional neural networks , 2017, Expert Syst. Appl..

[29]  Ligang Liu,et al.  Upright orientation of 3D shapes with Convolutional Networks , 2016, Graph. Model..

[30]  Rui Li,et al.  Preparation and characterization of novel light induced self-healing materials for cracks in asphalt pavements , 2016 .

[31]  Wenyu Liu,et al.  Traffic sign detection and recognition using fully convolutional network guided proposals , 2016, Neurocomputing.

[32]  Jin Wu,et al.  GPR detection of several common subsurface voids inside dikes and dams , 2010 .

[33]  Jun Wu,et al.  Pornographic image detection utilizing deep convolutional neural networks , 2016, Neurocomputing.

[34]  Dar-Hao Chen,et al.  Estimating the hotmix asphalt air voids from ground penetrating radar , 2014 .

[35]  Lai-Man Po,et al.  Integration of image quality and motion cues for face anti-spoofing: A neural network approach , 2016, J. Vis. Commun. Image Represent..

[36]  Nectaria Diamanti,et al.  Ground-penetrating radar for assessing winter roads , 2016 .