Techno Press
You logged in as Techno Press

Smart Structures and Systems
  Volume 31, Number 4, April 2023 , pages 383-392
DOI: https://doi.org/10.12989/sss.2023.31.4.383
 


Automatic assessment of post-earthquake buildings based on multi-task deep learning with auxiliary tasks
Zhihang Li, Huamei Zhu, Mengqi Huang, Pengxuan Ji, Hongyu Huang and Qianbing Zhang

 
Abstract
    Post-earthquake building condition assessment is crucial for subsequent rescue and remediation and can be automated by emerging computer vision and deep learning technologies. This study is based on an endeavour for the 2nd International Competition of Structural Health Monitoring (IC-SHM 2021). The task package includes five image segmentation objectives - defects (crack/spall/rebar exposure), structural component, and damage state. The structural component and damage state tasks are identified as the priority that can form actionable decisions. A multi-task Convolutional Neural Network (CNN) is proposed to conduct the two major tasks simultaneously. The rest 3 sub-tasks (spall/crack/rebar exposure) were incorporated as auxiliary tasks. By synchronously learning defect information (spall/crack/rebar exposure), the multi-task CNN model outperforms the counterpart single-task models in recognizing structural components and estimating damage states. Particularly, the pixel-level damage state estimation witnesses a mIoU (mean intersection over union) improvement from 0.5855 to 0.6374. For the defect detection tasks, rebar exposure is omitted due to the extremely biased sample distribution. The segmentations of crack and spall are automated by single-task U-Net but with extra efforts to resample the provided data. The segmentation of small objects (spall and crack) benefits from the resampling method, with a substantial IoU increment of nearly 10%.
 
Key Words
    building assessment; CNN; multi-task deep learning; semantic segmentation; small object detection
 
Address
(1) Zhihang Li, Huamei Zhu, Mengqi Huang, Pengxuan Ji, Qianbing Zhang:
Department of Civil Engineering, Monash University, Wellington Road Clayton, Victoria 3800, Australia;
(2) Hongyu Huang:
Institute of Geotechnical Engineering, Zhejiang University, Hangzhou 310058, China.
 

Techno-Press: Publishers of international journals and conference proceedings.       Copyright © 2024 Techno Press
P.O. Box 33, Yuseong, Daejeon 305-600 Korea, Tel: +82-42-828-7996, Fax : +82-42-828-7997, Email: admin@techno-press.com