Accelerating multi-class defect detection of building façades using knowledge distillation of DCNN-based model

Kisu Lee; Sanghyo Lee; Hayoung Kim

doi:10.22712/susb.20210008

All Issue

2021 Vol.12, Issue 2 Next Page

General Article

Accelerating multi-class defect detection of building façades using knowledge distillation of DCNN-based model

30 June 2021. pp. 80-95

PDF XML

Abstract

This paper proposes a high-speed detection method for multi-class defects in residential building façades. Automated deep learning-based defect detection systems have been developed to compensate for various problems in existing human-oriented defect management methods for building façades. However, the superior performance of deep learning-based models occasionally causes a trade-off with the inference time. In other words, using a lightweight model results in performance degradation, which we propose to prevent through a knowledge distillation (KD) method. This study was conducted using approximately 10,000 building façade images, which were obtained using drones. Using these data, we compared the performances of the lightweight model trained simply and the model trained with a KD method. As a result, mean average precision (mAP) increased by approximately 20% and inference time decreased by approximately 2.5x.

Keywords

building façade defects

multi-class defect detection

deep learning

knowledge distillation

model compression

References

G. Atkinson, A century of defects. Building. (1987), pp. 54-55.

C. Pereira, J. de Brito, and J.D. Silvestre, Contribution of humidity to the degradation of façade claddings in current buildings. Engineering Failure Analysis. 90 (2018), pp. 103-115. 10.1016/j.engfailanal.2018.03.028

J. Guo, Q. Wang, Y. Li, and P. Liu, Façade defects classification from imbalanced dataset using meta learning‐based convolutional neural network. Computer‐Aided Civil and Infrastructure Engineering. 35(12) (2020), pp. 1403-1418. 10.1111/mice.12578

A. Aïssani, A. Chateauneuf, J.P. Fontaine, and P. Audebert, Quantification of workmanship insulation defects and their impact on the thermal performance of building facades. Applied Energy. 165 (2016), pp. 272-284. 10.1016/j.apenergy.2015.12.040

E. Bauer, E. Kraus, M.N. Silva, and V.A. Zanoni, Evaluation of damage of building facades in Brasília. In XIII DBMC-International Conference on Durability of Building Materials and Components. 1 (2014), pp. 535-542.

J. Zhang, C. Lu, J. Wang, L. Wang, and X.G. Yue, Concrete cracks detection based on FCN with dilated convolution. Applied Sciences. 9(13) (2019), pp. 2686. 10.3390/app9132686

K. Lee, G. Hong, L. Sael, S. Lee, and H.Y. Kim, MultiDefectNet: Multi-Class Defect Detection of Building Façade Based on Deep Convolutional Neural Network. Sustainability. 12(22) (2020), pp. 9785. 10.3390/su12229785

S. Kim, D.M. Frangopol, and M. Soliman, Generalized probabilistic framework for optimum inspection and maintenance planning. Journal of Structural Engineering. 139(3) (2013), pp. 435-447. 10.1061/(ASCE)ST.1943-541X.0000676

A. Santos, M. Vicente, J. De Brito, I. Flores-Colen, and A. Castelo, Inspection, diagnosis, and rehabilitation system of door and window frames. Journal of Performance of Constructed Facilities. 31(3) (2017), pp. 04016118. 10.1061/(ASCE)CF.1943-5509.0000992

R. Bortolini and N. Forcada, A probabilistic performance evaluation for buildings and constructed assets. Building Research & Information. 48(8) (2020), pp. 838-855. 10.1080/09613218.2019.1704208

Y.J. Yan, L. Cheng, Z.Y. Wu, and L.H. Yam, Development in vibration-based structural damage detection technique. Mechanical systems and signal processing. 21(5) (2007), pp. 2198-2211. 10.1016/j.ymssp.2006.10.002

W.L. Bayissa, N. Haritos, and S. Thelandersson, Vibration-based structural damage identification using wavelet transform. Mechanical systems and signal processing. 22(5) (2008), pp. 1194-1215. 10.1016/j.ymssp.2007.11.001

X. Kong, C.S. Cai, and J. Hu, The state-of-the-art on framework of vibration-based structural damage identification for decision making. Applied Sciences. 7(5) (2017), pp. 497. 10.3390/app7050497

H. Sun and O. Büyüköztürk, Optimal sensor placement in structural health monitoring using discrete optimization. Smart Materials and Structures. 24(12) (2015), pp. 125034. 10.1088/0964-1726/24/12/125034

S. Nishiyama, N. Minakata, T. Kikuchi, and T. Yano, Improved digital photogrammetry technique for crack monitoring. Advanced Engineering Informatics. 29(4) (2015), pp. 851-858. 10.1016/j.aei.2015.05.005

K.W. Liao and D.R. Cheng, Restoration of the distorted color to detect the discoloration status of a steel bridge coating using digital image measurements. Advanced Engineering Informatics. 33 (2017), pp. 96-111. 10.1016/j.aei.2017.04.005

Y.J. Cha, W. Choi, and O. Büyüköztürk, Deep learning-based crack damage detection using convolutional neural networks. Computer‐Aided Civil and Infrastructure Engineering. 32(5) (2017), pp. 361-378. 10.1111/mice.12263

C.M. Yeum and S.J. Dyke, Vision‐based automated crack detection for bridge inspection. Computer‐Aided Civil and Infrastructure Engineering. 30(10) (2015), pp. 759-770. 10.1111/mice.12141

M. Wang and J.C. Cheng, A unified convolutional neural network integrated with conditional random field for pipe defect segmentation. Computer‐Aided Civil and Infrastructure Engineering. 35(2) (2020), pp. 162-177. 10.1111/mice.12481

M. Wang, S.S. Kumar, and J.C. Cheng, Automated sewer pipe defect tracking in CCTV videos based on defect detection and metric learning. Automation in Construction. 121 (2021), pp. 103438. 10.1016/j.autcon.2020.103438

G. Hinton, O. Vinyals, and J. Dean, Distilling the knowledge in a neural network. arXiv preprint arXiv:1503. 02531. (2015).

A. Romero, N. Ballas, S.E. Kahou, A. Chassang, C. Gatta, and Y. Bengio, Fitnets: Hints for thin deep nets. arXiv preprint arXiv:1412.6550. (2014).

G. Chen, W. Choi, X. Yu, T. Han, and M. Chandraker, Learning efficient object detection models with knowledge distillation. In Advances in neural information processing systems. (2017), pp. 742-751.

J. Redmon, S. Divvala, R. Girshick, and A. Farhadi, You only look once: Unified, real-time object detection. In Proceedings of the IEEE conference on computer vision and pattern recognition. (2016), pp. 779-788. 10.1109/CVPR.2016.91

R. Girshick, J. Donahue, T. Darrell, and J. Malik, Rich feature hierarchies for accurate object detection and semantic segmentation. In Proceedings of the IEEE conference on computer vision and pattern recognition. (2014), pp. 580-587. 10.1109/CVPR.2014.81

Zero to Hero: Guide to Object Detection using Deep Learning: Faster R-CNN, YOLO, SSD, CV-Tricks.com. https://cv-tricks.com/object-detection/faster-r-cnn-yolo-ssd/.

S. Ren, K. He, R. Girshick, and J. Sun, Faster r-cnn: Towards real-time object detection with region proposal networks. IEEE transactions on pattern analysis and machine intelligence. 39(6) (2016), pp. 1137-1149. 10.1109/TPAMI.2016.2577031

K. He, X. Zhang, S. Ren, and J. Sun, Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition. (2016), pp. 770-778. 10.1109/CVPR.2016.90

S. Ruder, An overview of gradient descent optimization algorithms. arXiv preprint arXiv:1609.04747. (2016).

T.Y. Lin, P. Dollár, R. Girshick, K. He, B. Hariharan, and S. Belongie. Feature pyramid networks for object detection. In Proceedings of the IEEE conference on computer vision and pattern recognition, (2017), pp. 2117-2125. 10.1109/CVPR.2017.106

M. Sandler, A. Howard, M. Zhu, A. Zhmoginov, and L.C. Chen, Mobilenetv2: Inverted residuals and linear bottlenecks. In Proceedings of the IEEE conference on computer vision and pattern recognition, (2018), pp. 4510-4520. 10.1109/CVPR.2018.00474

Information

Publisher :Sustainable Building Research Center (ERC) Innovative Durable Building and Infrastructure Research Center
Publisher(Ko) :건설구조물 내구성혁신 연구센터
Journal Title :International Journal of Sustainable Building Technology and Urban Development
Volume : 12
No :2
Pages :80-95
Received Date : 2021-02-02
Accepted Date : 2021-03-17
DOI :https://doi.org/10.22712/susb.20210008

International Journal of Sustainable Building Technology and Urban DevelopmentISSN:2093-761X(Print) 2093-7628(Online)건설구조물 내구성혁신 연구센터

All Issue