Issue 65

L. Wang, Frattura ed Integrità Strutturale, 65 (2023) 289-299; DOI: 10.3221/IGF-ESIS.65.19

Automatic detection of concrete cracks from images using Adam SqueezeNet deep learning model

Lin Wang Hefei University, Hefei, Anhui Province, 230061, China linw@smu.edu, http://orcid.org/0000-0002-6481-2409

A BSTRACT . Cracks in the concrete surfaces are typically clear warning signs of a potential threat to the integrity and serviceability of the structure. The techniques based on image processing can effectively detect cracks in digital images. These techniques, however, are generally susceptible to user-driven heuristic thresholds and irrelevant distractors. Inspired by the recent success of artificial intelligence, a deep learning-based automated crack detection system named CrackSN was presented. This proposed deep learning model, built on the Adam-SqueezeNet architecture, automatically learns the discriminative feature directly from the labeled and augmented patches. An image dataset of concrete surfaces is collected by smartphone and carefully prepared in order to develop and train the CrackSN system. The hyper parameters of SqueezeNet are tuned with Adam optimization through the training and validation procedures. The fine-tuned CrackSN system outperforms state-of-the-art models in recent literature by correctly classifying 97.3% of the cracked patches in the image dataset. The success of CrackSN, demonstrated by its light network design and outstanding performance, provides a crucial step toward automated damage inspection and health evaluation for infrastructure. K EYWORDS . Concrete crack, Automated damage inspection, SqueezeNet, Adam optimization, Deep learning.

Citation: Wang, L., Automatic detection of concrete crack from images using Adam SqueezeNet deep learning model, Frattura ed Integrità Strutturale, 65 (2023) 289-299.

Received: 07.04.2023 Accepted: 18.06.2023 Online first: 19.06.2023 Published: 01.07.2023

Copyright: © 2023 This is an open access article under the terms of the CC-BY 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

I NTRODUCTION oncreted civil infrastructure such as buildings, bridges, and highways sustain various external loads throughout their service period. Earthquake vibration, wind, temperature, or vehicle-made excitation initiate structural damage during their lifetime and , subsequently, trigger catastrophic failure [1, 2]. Regular inspection and maintenance are vital to ensuring the integrity of concrete infrastructure for public safety, economic running, and national security. Cracks on concrete surfaces are essential indicators of the potential damage attached to the infrastructure [1-5]. In this regard, C

289