Manufacture Assembly Fault Detection Method based on Deep Learning and Mixed Reality

Methodology to integrate augmented reality and pattern recognition for crack detection

“…‐F. Liu et al., 2020; Wang et al., 2018). The proposed method can serve as the first step in crack dimensional characterization and condition evaluation with AR headsets.…”

“…Several recent studies utilized different deep learning algorithms for crack detection (Aravind et al., 2021; Çelik & König, 2022; Cha et al., 2017; Chen & He, 2022; Dung & Anh, 2019; Le et al., 2021; C. Liu & Xu, 2022; Miao & Srimahachota, 2021; Ni et al., 2019; Piyathilaka et al., 2020; Żarski et al., 2022; Zhang & Yuen, 2021; Zheng et al., 2022; Zhou et al., 2022; Zou et al., 2022). Alternatively, several studies have focused on pattern recognition crack detection as another approach to image‐based methods (Dorafshan et al., 2019; Iyer & Sinha, 2005; Li et al., 2014; Miao et al., 2020; Safaei et al., 2022; W. Wang et al., 2018; Y. Wang et al., 2019). Table 1 shows the benefits and limitations of different artificial intelligence approaches used for structural health monitoring (SHM).…”

“…AR headsets can deploy a crack detection/characterization application for humans, enabling hands‐free visual inspections while automatically obtaining useful data via computer‐environment interaction (Karaaslan et al., 2019). Past studies explored the use of AR technology for object recognition (Bahri et al., 2019; Corneli et al., 2019; Farasin et al., 2020; L. Liu et al., 2019; Pepe et al., 2018; Shuai Wang et al., 2018). Evaluation of infrastructure inspection with AR headsets is among the recent interests in SHM and maintenance methods (Maharjan et al., 2021; Mascareñas et al., 2021).…”