Comparative Analysis of Deep Neural Networks for Crack Image Classification

  • Sheerin Sitara Noor MohamedEmail author
  • Kavitha Srinivasan
Conference paper
Part of the Lecture Notes on Data Engineering and Communications Technologies book series (LNDECT, volume 38)


Deep Learning (DL) is widely used in different types of classification problems in real time. DL models can be constructed in two ways namely, Convolutional Neural Network (CNN) and using pre-trained models such as VGG16, VGG19 and Inception ResNet V2. In this paper, an automatic crack image classification approach is proposed and implemented using CNN and pre–trained models. For validation, three types of datasets having both enhanced and without enhanced crack images are used and the result of classification are analysed using appropriate quantitative metrics such as accuracy, precision and recall, from the results, it has been inferred that the proposed CNN, derived the highest accuracy of 99% for dataset 1 whereas Inception ResNet V2 model, derived the highest accuracy of 87% and 94% for dataset 2 and dataset 3 respectively.


Crack classification Convolutional Neural Network Pre-trained models Transfer learning approach Deep Learning 


  1. 1.
    Chen, Y., Mei, T., Wang, X., Li, F.: A bridge crack image detection and classification method based on climbing robot. In: Proceeding 35th Chinese Control Conference, pp. 4037–4042 (2016)Google Scholar
  2. 2.
    Sheerin Sitara, N., Kavitha, S., Raghuraman, G.: Review and analysis of crack detection and classification techniques based on crack types. Int. J. Appl. Eng. Res. 13(8), 6056–6062 (2018)Google Scholar
  3. 3.
    Sheng, P., Chen, L., Tian, J.: Learning-based road crack detection using gradient boost decision tree. In: IEEE Conference on Industrial Electronics and Applications, pp. 1228–1232 (2018)Google Scholar
  4. 4.
    Salari, E., Ouyang, D.: An image-based pavement distress detection and classification. In: IEEE International Conference on Electro/Information Technology, pp. 1–6 (2012)Google Scholar
  5. 5.
    da Silva, W.R.L., de Lucena, D.S.: Concrete crack detection based on deep learning image classification. In: 18th International Conference on Experimental Mechanics, Brussels, Belgium, pp. 1–6 (2018)Google Scholar
  6. 6.
    Zhang, D.-Q.: Image recognition using scale recurrent neural networks. In: Computer Vision and Pattern Recognition, pp. 1–5. (2018)
  7. 7.
    Chinea, A.: Understanding the principles of recursive neural networks: a generative approach to tackle model complexity. In: International Conference on Artificial Neural Networks, pp. 952–963 (2009)CrossRefGoogle Scholar
  8. 8.
    Krizhevsky, A., Sutskever, I., Hinton, G.E.: ImageNet classification with deep convolutional neural networks. In: Neural Information Processing Systems, Tahoe, Nevada, pp. 1–9 (2012)Google Scholar
  9. 9.
    Simonyan, K., Zisserman, A.: Very deep convolutional neural networks for large scale image recognition. In: International Conference on Learning Representations, Banff, Canada, pp. 1–14 (2014)Google Scholar
  10. 10.
    Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., Rabinovich, A.: Going deeper with convolutions. In: Conference on Computer Vision and Pattern Recognition, Columbus, Ohio, pp. 1–9 (2014)Google Scholar
  11. 11.
    He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: ImageNet Large Scale Visual Recognition Challenge, pp. 1–12 (2016)Google Scholar
  12. 12.
    Szegedy, C., Ioffe, S., Vanhoucke, V., Alemi, A.A.: Inception-v4, inception-ResNet and the impact of residual connections on learning. In: Proceedings of the Thirty-First Association for the Advancement of Artificial Intelligence Conference on Artificial Intelligence, pp. 4278–4284 (2017)Google Scholar
  13. 13.
    Özgenel, C.F., Gönenç Sorguç, A.: Performance comparison of pretrained convolutional neural networks on crack detection in buildings. In: 35th International Symposium on Automation and Robotics in Construction, pp. 1–8 (2018)Google Scholar
  14. 14.
    Zhang, L., Yang, F., Zhang, Y.D., Zhu, Y.J.: Road crack detection using deep convolutional neural network. In: IEEE International Conference on Image Processing, pp. 3708–3712 (2016). DatasetGoogle Scholar
  15. 15.
    Maguire, M., Dorafshan, S., Thomas, R.J.: SDNET2018: a concrete crack image dataset for machine learning applications (2018). Browse all Datasets. Paper 48. Accessed Feb 2019. Dataset
  16. 16.
    Dataset: Accessed Aug 2017
  17. 17.

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  • Sheerin Sitara Noor Mohamed
    • 1
    Email author
  • Kavitha Srinivasan
    • 1
  1. 1.Department of Computer Science and EngineeringSSN College of EngineeringChennaiIndia

Personalised recommendations