Research Article
BibTex RIS Cite

Detection of Concrete Surface Cracks with Deep Learning Architectures

Year 2022, , 318 - 325, 30.12.2022
https://doi.org/10.46460/ijiea.1098046

Abstract

The most basic problem for concrete surfaces is the presence of cracks. These cracks should be detected and repaired as soon as possible to ensure safety. At the present time, detection of cracks is done by human power. Although a lot of effort is spent in the determinations made with manpower, the error rate is high. The aim of this study is to ensure more accurate and faster detection of cracks. For this, an autonomous system is needed. Some Convolutional Neural Networks (CNN) have been used in the detection of concrete surface cracks. The image data used in this study were collected from the Middle East Technical Universty (METU) campus buildings. There are 20000 Negative and 20000 Positive data in this data set. Image data was trained using deep CNN architectures such as ResNet-50, VGG-16, Inception-V3, Xeption and lightweight CNN architectures such as MobileNet, ShuffleNet, EfficientNet. By comparing the data obtained as a result of the training, it was observed how the accuracy changed when fewer parameters were used.

References

  • [1] B. Y. N. S. P. K. R. &. A. P. R. Kim, «Surface crack detection using deep learning with shallow CNN architecture for enhanced computation.,» Neural Computing and Applications, 2021.
  • [2] L. A. F. J. H. A. G. M. K. W. &. S. M. A. Ali, «"Performance evaluation of deep CNN-based crack detection and localization techniques for concrete structures.",» Sensors , 2021.
  • [3] C. D. G. V. M. D. C. A. M. v. L. S. L. Attard, «"Automatic Crack Detection using Mask R-CNN,",» %1 içinde International Symposium on Image and Signal Processing and Analysis (ISPA), 2019.
  • [4] K. &. B. H. B. Hacıefendioğlu, « Concrete road crack detection using deep learning-based faster R-CNN method.,» Iranian Journal of Science and Technology, Transactions of Civil Engineering, 2021.
  • [5] X. Z. M. S. P. R. R. H. X. W. X. &. Y. H. Xu, «Crack Detection and Comparison Study Based on Faster R-CNN and Mask R-CNN,» Sensors, 2022.
  • [6] J. Y. Y. Z. D. T. W. C. R. Z. Q. &. L. Q. Liao, «Automatic Tunnel Crack Inspection Using an Efficient Mobile Imaging Module and a Lightweight CNN,» IEEE Transactions on Intelligent Transportation Systems, 2022.
  • [7] C. k. dedektörü, «Wikipedia,» 24 01 2022. [Çevrimiçi]. Available: https://en.wikipedia.org/wiki/Canny_edge_detector#:~:text=The%20Canny%20edge%20detector%20is,explaining%20why%20the% 20technique%20works..
  • [8] S. a. V. S. Niu, « "Simulation trained CNN for accurate embedded crack length, location, and orientation prediction from ultrasound measurements.",» International Journal of Solids and Structures, 2022.
  • [9] Ç. F. Özgenel, «Concrete Crack Images for Classification,» Mendeley Data, 2019.
  • [10] L. F. Y. Y. D. Z. a. Y. J. Z. Zhang, « “Road crack detection using deep convolutional neural network.”,» IEEE International Conference on Image Processing (ICIP), 2016.
  • [11] C. &. W. G. &. P. F. Pelletier, «Temporal Convolutional Neural Network for the Classification of Satellite Image Time Series.,» Remote Sensing, 2019.
  • [12] C. Szegedy, «"Rethinking the inception architecture for computer vision.",» %1 içinde Proceedings of the IEEE conference on computer vision and pattern recognition, 2016.
  • [13] A. Pujara, «Image Classification With MobileNet,» Medium, 4 7 2020. [Çevrimiçi]. Available: https://medium.com/analytics-vidhya/image-classification-with-mobilenet-cc6fbb2cd470.
  • [14] L. S. Y. X. S. L. X. H. a. R. F. M. L. Li, «Fuzzy Multilevel Image Thresholding Based on Improved Coyote Optimization Algorithm,» IEEE Access, 2021.
  • [15] A. M. T. D. M. F. L. B. V. V. &. J. T. Baloch, «Hardware Synthesize and Performance Analysis of Intelligent Transportation Using Canny Edge Detection Algorithm.,» International Journal of Engineering and Manufacturing, 2021.
  • [16] P. Broberg, «Surface crack detection in welds using thermography.,» NDT & E International, 2013.
  • [17] B. Z. W. G. P. Z. Y. &. W. Z. Wang, «Crack damage detection method via multiple visual features and efficient multi-task learning model,» Sensors, 2018.
  • [18] S. S. N. &. S. K. Mohamed, « A Hybrid Image Enhancement Algorithm for Effective Concrete Surface Crack Classification.,» Journal of University of Shanghai for Science and Technology, 2021.
  • [19] X. Meng, «Concrete Crack Detection Algorithm Based on Deep Residual Neural Networks.,» Scientific Programming , 2021.
  • [20] F. Chollet, «Xception: Deep learning with depthwise separable convolutions,» %1 içinde Proceedings of the IEEE conference on computer vision and pattern recognition, 2017.

Beton Yüzey Çatlaklarının Tespitinde Derin Öğrenme Mimarilerin Kullanılması

Year 2022, , 318 - 325, 30.12.2022
https://doi.org/10.46460/ijiea.1098046

Abstract

Beton yüzeyler için en temel problem çatlakların varlığıdır. Bu çatlaklar, güvenliğin sağlanabilmesi için mümkün olan en kısa sürede tespit edilip onarılmalıdır. Günümüzde çatlakların tespit edilmesi insan gücüyle gerçekleştirilmektedir. İnsan gücü ile yapılan tespitlerde fazla emek olmasına karşın hata oranı yüksektir. Bu çalışmanın amacı, çatlakların daha doğru ve hızlı tespit edilmesini sağlamaktır. Bunun için ise otonom bir sisteme ihtiyaç duyulmaktadır. Beton yüzey çatlaklarının tespitinde bazı Evrişimsel Sinir Ağları (CNN) kullanılmıştır. Bu çalışmada kullanılan görüntü verisi Orta Doğu Teknik Üniversitesi (ODTÜ) kampüs binalarından toplanmıştır. Bu veri setinde 20000 Negatif ve 20000 Pozitif veri bulunmaktadır. Görüntü verileri, ResNet-50, VGG-16, Inception-V3, Xeption gibi derin CNN mimarileri ve MobileNet, ShuffleNet, EfficientNet gibi hafif CNN mimarilerini kullanarak eğitildi. Eğitim sonucunda elde edilen veriler karşılaştırılarak, daha az parametre kullanıldığında doğruluğun nasıl değiştiği gözlemlenmiştir.

References

  • [1] B. Y. N. S. P. K. R. &. A. P. R. Kim, «Surface crack detection using deep learning with shallow CNN architecture for enhanced computation.,» Neural Computing and Applications, 2021.
  • [2] L. A. F. J. H. A. G. M. K. W. &. S. M. A. Ali, «"Performance evaluation of deep CNN-based crack detection and localization techniques for concrete structures.",» Sensors , 2021.
  • [3] C. D. G. V. M. D. C. A. M. v. L. S. L. Attard, «"Automatic Crack Detection using Mask R-CNN,",» %1 içinde International Symposium on Image and Signal Processing and Analysis (ISPA), 2019.
  • [4] K. &. B. H. B. Hacıefendioğlu, « Concrete road crack detection using deep learning-based faster R-CNN method.,» Iranian Journal of Science and Technology, Transactions of Civil Engineering, 2021.
  • [5] X. Z. M. S. P. R. R. H. X. W. X. &. Y. H. Xu, «Crack Detection and Comparison Study Based on Faster R-CNN and Mask R-CNN,» Sensors, 2022.
  • [6] J. Y. Y. Z. D. T. W. C. R. Z. Q. &. L. Q. Liao, «Automatic Tunnel Crack Inspection Using an Efficient Mobile Imaging Module and a Lightweight CNN,» IEEE Transactions on Intelligent Transportation Systems, 2022.
  • [7] C. k. dedektörü, «Wikipedia,» 24 01 2022. [Çevrimiçi]. Available: https://en.wikipedia.org/wiki/Canny_edge_detector#:~:text=The%20Canny%20edge%20detector%20is,explaining%20why%20the% 20technique%20works..
  • [8] S. a. V. S. Niu, « "Simulation trained CNN for accurate embedded crack length, location, and orientation prediction from ultrasound measurements.",» International Journal of Solids and Structures, 2022.
  • [9] Ç. F. Özgenel, «Concrete Crack Images for Classification,» Mendeley Data, 2019.
  • [10] L. F. Y. Y. D. Z. a. Y. J. Z. Zhang, « “Road crack detection using deep convolutional neural network.”,» IEEE International Conference on Image Processing (ICIP), 2016.
  • [11] C. &. W. G. &. P. F. Pelletier, «Temporal Convolutional Neural Network for the Classification of Satellite Image Time Series.,» Remote Sensing, 2019.
  • [12] C. Szegedy, «"Rethinking the inception architecture for computer vision.",» %1 içinde Proceedings of the IEEE conference on computer vision and pattern recognition, 2016.
  • [13] A. Pujara, «Image Classification With MobileNet,» Medium, 4 7 2020. [Çevrimiçi]. Available: https://medium.com/analytics-vidhya/image-classification-with-mobilenet-cc6fbb2cd470.
  • [14] L. S. Y. X. S. L. X. H. a. R. F. M. L. Li, «Fuzzy Multilevel Image Thresholding Based on Improved Coyote Optimization Algorithm,» IEEE Access, 2021.
  • [15] A. M. T. D. M. F. L. B. V. V. &. J. T. Baloch, «Hardware Synthesize and Performance Analysis of Intelligent Transportation Using Canny Edge Detection Algorithm.,» International Journal of Engineering and Manufacturing, 2021.
  • [16] P. Broberg, «Surface crack detection in welds using thermography.,» NDT & E International, 2013.
  • [17] B. Z. W. G. P. Z. Y. &. W. Z. Wang, «Crack damage detection method via multiple visual features and efficient multi-task learning model,» Sensors, 2018.
  • [18] S. S. N. &. S. K. Mohamed, « A Hybrid Image Enhancement Algorithm for Effective Concrete Surface Crack Classification.,» Journal of University of Shanghai for Science and Technology, 2021.
  • [19] X. Meng, «Concrete Crack Detection Algorithm Based on Deep Residual Neural Networks.,» Scientific Programming , 2021.
  • [20] F. Chollet, «Xception: Deep learning with depthwise separable convolutions,» %1 içinde Proceedings of the IEEE conference on computer vision and pattern recognition, 2017.
There are 20 citations in total.

Details

Primary Language Turkish
Subjects Engineering
Journal Section Articles
Authors

Arzu Sevinç 0000-0002-1614-117X

Fatih Özyurt 0000-0002-8154-6691

Publication Date December 30, 2022
Submission Date April 5, 2022
Published in Issue Year 2022

Cite

APA Sevinç, A., & Özyurt, F. (2022). Beton Yüzey Çatlaklarının Tespitinde Derin Öğrenme Mimarilerin Kullanılması. International Journal of Innovative Engineering Applications, 6(2), 318-325. https://doi.org/10.46460/ijiea.1098046
AMA Sevinç A, Özyurt F. Beton Yüzey Çatlaklarının Tespitinde Derin Öğrenme Mimarilerin Kullanılması. ijiea, IJIEA. December 2022;6(2):318-325. doi:10.46460/ijiea.1098046
Chicago Sevinç, Arzu, and Fatih Özyurt. “Beton Yüzey Çatlaklarının Tespitinde Derin Öğrenme Mimarilerin Kullanılması”. International Journal of Innovative Engineering Applications 6, no. 2 (December 2022): 318-25. https://doi.org/10.46460/ijiea.1098046.
EndNote Sevinç A, Özyurt F (December 1, 2022) Beton Yüzey Çatlaklarının Tespitinde Derin Öğrenme Mimarilerin Kullanılması. International Journal of Innovative Engineering Applications 6 2 318–325.
IEEE A. Sevinç and F. Özyurt, “Beton Yüzey Çatlaklarının Tespitinde Derin Öğrenme Mimarilerin Kullanılması”, ijiea, IJIEA, vol. 6, no. 2, pp. 318–325, 2022, doi: 10.46460/ijiea.1098046.
ISNAD Sevinç, Arzu - Özyurt, Fatih. “Beton Yüzey Çatlaklarının Tespitinde Derin Öğrenme Mimarilerin Kullanılması”. International Journal of Innovative Engineering Applications 6/2 (December 2022), 318-325. https://doi.org/10.46460/ijiea.1098046.
JAMA Sevinç A, Özyurt F. Beton Yüzey Çatlaklarının Tespitinde Derin Öğrenme Mimarilerin Kullanılması. ijiea, IJIEA. 2022;6:318–325.
MLA Sevinç, Arzu and Fatih Özyurt. “Beton Yüzey Çatlaklarının Tespitinde Derin Öğrenme Mimarilerin Kullanılması”. International Journal of Innovative Engineering Applications, vol. 6, no. 2, 2022, pp. 318-25, doi:10.46460/ijiea.1098046.
Vancouver Sevinç A, Özyurt F. Beton Yüzey Çatlaklarının Tespitinde Derin Öğrenme Mimarilerin Kullanılması. ijiea, IJIEA. 2022;6(2):318-25.