Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

CrackSpot: deep learning for automated detection of structural cracks in concrete infrastructure

CrackSpot: deep learning for automated detection of structural cracks in concrete infrastructure To maintain structural integrity and avoid structural failures that could harm neighboring infrastructure, pollute the environment, and even result in fatalities, routine inspection and repair of concrete infrastructure are required. Throughout the structure’s operational life, routine visual inspections are typically undertaken to detect various problems caused by environmental exposure (such as cracks, loss of material, rusting of metal bindings, etc.). Visual examination can yield a variety of data that may enable the cause of distress to be positively identified. Its effectiveness is subject to human error and depends on the investigator’s skill and experience and because of their size and difficult-to-reach features, huge structures like dams, bridges, and tall skyscrapers can be prohibitively dangerous. The approach presented here uses deep learning techniques to identify the structural cracks on concrete surfaces to achieve easy detection of the cracks and high accuracy. Here, we propose an integrated Tensrflow CNN and image processing-based crack-finding method to detect cracks with high precision. Thousands of photos of cracked and non-cracked structure surface datasets are considered while developing the model. Image features are extracted during pre-processing to increase training effectiveness. The developed model has a 97.11% accuracy rate and an F1-score of 97%. The results show that the designed model is highly precise and effective in identifying cracks in structures and more accurate than many implemented techniques. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Asian Journal of Civil Engineering Springer Journals

CrackSpot: deep learning for automated detection of structural cracks in concrete infrastructure

Download PDF
12 pages

Loading next page...
 
/lp/springer-journals/crackspot-deep-learning-for-automated-detection-of-structural-cracks-4UK42hoQCY

References (39)

Publisher
Springer Journals
Copyright
Copyright © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2023. Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
ISSN
1563-0854
eISSN
2522-011X
DOI
10.1007/s42107-023-00754-7
Publisher site
See Article on Publisher Site

Abstract

To maintain structural integrity and avoid structural failures that could harm neighboring infrastructure, pollute the environment, and even result in fatalities, routine inspection and repair of concrete infrastructure are required. Throughout the structure’s operational life, routine visual inspections are typically undertaken to detect various problems caused by environmental exposure (such as cracks, loss of material, rusting of metal bindings, etc.). Visual examination can yield a variety of data that may enable the cause of distress to be positively identified. Its effectiveness is subject to human error and depends on the investigator’s skill and experience and because of their size and difficult-to-reach features, huge structures like dams, bridges, and tall skyscrapers can be prohibitively dangerous. The approach presented here uses deep learning techniques to identify the structural cracks on concrete surfaces to achieve easy detection of the cracks and high accuracy. Here, we propose an integrated Tensrflow CNN and image processing-based crack-finding method to detect cracks with high precision. Thousands of photos of cracked and non-cracked structure surface datasets are considered while developing the model. Image features are extracted during pre-processing to increase training effectiveness. The developed model has a 97.11% accuracy rate and an F1-score of 97%. The results show that the designed model is highly precise and effective in identifying cracks in structures and more accurate than many implemented techniques.

Journal

Asian Journal of Civil Engineering – Springer Journals

Published: Jun 27, 2023

Keywords: Crack detection; Structural damage; Image processing; Structural load; Feature extraction; Convolutional neural network

There are no references for this article.