Loading...

Concrete Crack Detection Using Convolutional Neural Networks Based on Deep Learning

Mousavi Sarasia, Mohammad | 2021

217 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 54776 (09)
  4. University: Sharif University of Technology
  5. Department: Civil Engineering
  6. Advisor(s): Bakhshi, Ali
  7. Abstract:
  8. Crack detection is a critical task in monitoring and inspection of civil engineering structures. This study proposes a deep-learning-based model for automatic crack detection on the concrete surface. The proposed model is an encoder-decoder model which uses EfficientNetB7, the state-of-the-art convolutional neural network, as encoder and the U Net’s expansion path as decoder. To minimize the training time and maximizing the accuracy, we use transfer learning in our approach. We train our model with a novel training strategy on images from an open-source dataset and achieve 96.44% F1-score for unseen test data. To compare the performance of the proposed method, we evaluate our model on CFD dataset and achived 96.13% F1-score which outperforms all the previous crack detection methods. The robustness of proposed model is investigated by various number of train data and the optimized data size for training this model is presented. The results show that although deep-leaning models acquire large number of data, this model works with limited data, without losing its performance. Also the novel training strategy used in this study lead to a perfect performance on detection of fine cracks as well as ordinary cracks
  9. Keywords:
  10. Crack Detection ; Deep Learning ; Convolutional Neural Network ; Semantic Segmentation ; Structural Health Monitoring ; Vision Based Inspection

 Digital Object List

 Bookmark

No TOC