Improved Image Based Super Resolution and Concrete Crack Prediction Using Pre-Trained Deep Learning Models

Document Type : Regular Article

Authors

1 Assistant Professor, Department of CSE, Coimbatore Institute of Technology, Coimbatore, India

2 UG Student, Coimbatore Institute of Technology, Coimbatore, India

Abstract

Detection and prediction of cracks play a vital role in the maintenance of concrete structures. The manual instructions result in having images captured from different sources wherein the acquisition of such images into the network may cause an error. The errors are rectified by a method to increase the resolution of those images and are imposed through Super-Resolution Generative Adversarial Network (SRGAN) with a pre-trained model of VGG19. After increasing the resolution then comes the prediction of crack from high resolution images through Convolutional Neural Network (CNN) with a pre-trained model of ResNet50 that trains a dataset of 40,000 images which consists of both crack and non-crack images. This work makes a comparative analysis of predicting the crack after and before the super-resolution method and their performance measure is compared. Compared with other methods on super-resolution and prediction, the proposed method appears to be more stable, faster and highly effective. For the dataset used in this work, the model yields an accuracy of 98.2%, proving the potential of using deep learning for concrete crack detection.

Highlights

Google Scholar

Keywords

Main Subjects


[1]     Li S, Zhao X. Image-Based Concrete Crack Detection Using Convolutional Neural Network and Exhaustive Search Technique. Adv Civ Eng 2019;2019:1–12. doi:10.1155/2019/6520620.
[2]     Yang X, Li H, Yu Y, Luo X, Huang T, Yang X. Automatic Pixel-Level Crack Detection and Measurement Using Fully Convolutional Network. Comput Civ Infrastruct Eng 2018;33:1090–109. doi:10.1111/mice.12412.
[3]     Miko┼éajczyk A, Grochowski M. Data augmentation for improving deep learning in image classification problem. 2018 Int Interdiscip PhD Work, IEEE; 2018, p. 117–22.
[4]     Perez L, Wang J. The effectiveness of data augmentation in image classification using deep learning. ArXiv Prepr ArXiv171204621 2017.
[5]     Inoue H. Data augmentation by pairing samples for images classification. ArXiv Prepr ArXiv180102929 2018.
[6]     Goodfellow I, Pouget-Abadie J, Mirza M, Xu B, Warde-Farley D, Ozair S, et al. Generative adversarial nets. Adv Neural Inf Process Syst, 2014, p. 2672–80.
[7]     Wang C, Xu C, Yao X, Tao D. Evolutionary Generative Adversarial Networks. IEEE Trans Evol Comput 2019;23:921–34. doi:10.1109/TEVC.2019.2895748.
[8]     Yang W, Zhang X, Tian Y, Wang W, Xue J-H, Liao Q. Deep Learning for Single Image Super-Resolution: A Brief Review. IEEE Trans Multimed 2019;21:3106–21. doi:10.1109/TMM.2019.2919431.
[9]     Hoang N-D. Detection of Surface Crack in Building Structures Using Image Processing Technique with an Improved Otsu Method for Image Thresholding. Adv Civ Eng 2018;2018:1–10. doi:10.1155/2018/3924120.
[10]    Wang M, Chen Z, Wu QMJ, Jian M. Improved face super-resolution generative adversarial networks. Mach Vis Appl 2020;31:22. doi:10.1007/s00138-020-01073-6.
[11]    Zhang T. Research and Improvement of Single Image Super-Resolution Based on Generative Adversarial Network. J Phys Conf Ser, vol. 1237, IOP Publishing; 2019, p. 32046.
[12]    Simonyan K, Zisserman A. Very Deep Convolutional Networks for Large-Scale Image Recognition. ArXiv Prepr ArXiv14091556 2014.
[13]    Mikami H, Suganuma H, Tanaka Y, Kageyama Y. Imagenet/resnet-50 training in 224 seconds. ArXiv Prepr ArXiv181105233 2018:1–8.
[14]    Dhankhar P. ResNet-50 and VGG-16 for recognizing Facial Emotions. Int J Innov Eng Technol 2019;13:126–30.
[15]    Panella F, Boehm J, Loo Y, Kaushik A, Gonzalez D. Deep learning and image processing for automated crack detection and defect measurement in underground structures. ISPRS - Int Arch Photogramm Remote Sens Spat Inf Sci 2018;XLII–2:829–35. doi:10.5194/isprs-archives-XLII-2-829-2018.
[16]    Nandepu R. Understanding and implementation of Residual Networks (ResNets) 2019. https://medium.com/analytics-vidhya/understanding-and-implementation-of-residual-networks-resnets-b80f9a507b9c (accessed October 31, 2019).
[17]    He K, Zhang X, Ren S, Sun J. Deep Residual Learning for Image Recognition. 2016 IEEE Conf Comput Vis Pattern Recognit, IEEE; 2016, p. 770–8. doi:10.1109/CVPR.2016.90.
[18]    Zhang K, Cheng H-D, Gai S. Efficient Dense-Dilation Network for Pavement Cracks Detection with Large Input Image Size. 2018 21st Int Conf Intell Transp Syst, IEEE; 2018, p. 884–9. doi:10.1109/ITSC.2018.8569958.
[19]    Zhang K, Cheng HD, Zhang B. Unified Approach to Pavement Crack and Sealed Crack Detection Using Preclassification Based on Transfer Learning. J Comput Civ Eng 2018;32:04018001. doi:10.1061/(ASCE)CP.1943-5487.0000736.
[20]    Ledig C, Theis L, Huszár F, Caballero J, Cunningham A, Acosta A, et al. Photo-realistic single image super-resolution using a generative adversarial network. Proc IEEE Conf Comput Vis pattern Recognit, 2017, p. 4681–90.
[21]    Zhu X, Zhang L, Zhang L, Liu X, Shen Y, Zhao S. GAN-Based Image Super-Resolution with a Novel Quality Loss. Math Probl Eng 2020;2020:1–12. doi:10.1155/2020/5217429.
[22]    Cha Y-J, Choi W, Büyüköztürk O. Deep Learning-Based Crack Damage Detection Using Convolutional Neural Networks. Comput Civ Infrastruct Eng 2017;32:361–78. doi:10.1111/mice.12263.
[23]    Sultana F, Sufian A, Dutta P. Advancements in Image Classification using Convolutional Neural Network. 2018 Fourth Int Conf Res Comput Intell Commun Networks, IEEE; 2018, p. 122–9. doi:10.1109/ICRCICN.2018.8718718.
[24]    Zhang L, Yang F, Daniel Zhang Y, Zhu YJ. Road crack detection using deep convolutional neural network. 2016 IEEE Int Conf Image Process, IEEE; 2016, p. 3708–12. doi:10.1109/ICIP.2016.7533052.