Learning‐based image scale estimation using surface textures for quantitative visual inspection of regions‐of‐interest

Abstract

A major shortfall of vision‐based inspection solutions is the lack of scale information, required to resolve inspection regions to a physical scale. To address this challenge, a learning‐based scale estimation technique is proposed. The underlying assumption is that the surface texture of structures, captured in images, contains enough information to estimate scale for each corresponding image (e.g., pixel/mm). This permits the training of a regression model to establish the relationship between surface textures, captured in images, and scales. A convolutional neural network is trained to extract scale‐related features from textures captured in images. Then, the trained model can be exploited to estimate scales for all images that are captured from a structure's surfaces with similar textures. The capability of the proposed technique is demonstrated using data collected from surface textures of three different structures. An average scale estimation error, from images of each structure, is less than 15%.

Autopilot model for shield tunneling machines using support vector regression and its application to previously constructed tunnels

“…In recent years, expectations for machine learning have increased (Bui et al., 2022; Choi et al., 2022; T. Gao et al., 2022; Lin et al., 2022; Park et al, 2021; Wu et al., 2022; Żarski et al., 2022). Research using machine learning has been conducted in the field of tunneling (Xue et al., 2022; Z. Zhou et al., 2022; Zhu et al., 2021).…”

Automatic detection method of tunnel lining multi‐defects via an enhanced You Only Look Once network

“…Recently, the DL algorithms were adopted to solve challenging problems in the field of infrastructure (Martins et al., 2020; Rafiei & Adeli, 2016, 2017a), that is, earthquake early warning (Rafiei & Adeli, 2017b), prediction of ground settlements (K. Zhang et al., 2020), construction cost estimation (Rafiei & Adeli, 2018), and structural surface defects detection (Park et al., 2021). Convolutional neural network (CNN) is the most used detection algorithm for structural surface defects.…”

A graph‐based method for quantifying crack patterns on reinforced concrete shear walls

“…A. Park et al., 2021; S. E. Park et al., 2020). The central idea is to quantify the features of the structure and relate these features to the severity of the structural damage.…”

“…To quantify the extent of damage to a concrete structure, researchers, including the authors, have developed different damage assessment methods (Ebrahimkhanlou et al, 2016a;Momeni & Dolatshahi, 2019;J. A.…”