MétaCan
Menu
Back to cohort
Record W4313567559 · doi:10.1016/j.iintel.2022.100024

Automated multiclass structural damage detection and quantification using augmented reality

2023· article· en· W4313567559 on OpenAlex

Why this work is in the frame

A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.

affAt least one author lists a Canadian institution in the pinned OpenAlex snapshot.
fundA Canadian funder is recorded on the work.

Bibliographic record

VenueJournal of Infrastructure Intelligence and Resilience · 2023
Typearticle
Languageen
FieldEngineering
TopicInfrastructure Maintenance and Monitoring
Canadian institutionsWestern University
FundersNatural Sciences and Engineering Research Council of CanadaMinistry of Colleges and Universities
KeywordsComputer scienceStructural health monitoringVisualizationVisual inspectionIdentification (biology)SegmentationAugmented realityArtificial intelligenceScale (ratio)Engineering

Abstract

fetched live from OpenAlex

Civil infrastructure worldwide is ageing and enduring increasingly adverse weather conditions. Traditional structural health monitoring (SHM) involves the expensive and time-consuming installation of contact sensors. For example, inspectors use costly large-scale equipment to reach a certain area of the structure and at different heights to inspect it, which can pose a risk to the inspector's safety. Moreover, the inspectors rely only on the batch data acquired during the inspection period, which are analyzed by engineers at a later time due to the limited availability of a real-time visualization approach for structural inspection within the traditional mode of SHM. To address these timely challenges, an Augmented Reality (AR)-based automated multiclass damage identification and quantification methodology is proposed in this paper. The interactive visualization framework of AR is integrated with the autonomous decision-making of Artificial Intelligence (AI) in a unified fashion to incorporate human-sensor interaction. The proposed system uses an AI model that is trained and optimized using the YOLOv5 architecture to detect and classify four different types of anomalies/damages (i.e., cracks, spalls, pittings, and joints). The AI model is then updated to quantify the length, area, and perimeter of any damage using segmentation to further assess its severity. Once the model is developed, the model is embedded with the AR device and tested through its interactive environment for SHM of various structures. The paper concludes that the proposed approach successfully classifies four types of damage with an accuracy of more than 90% for up to 2 ​m, and it also quantifies the length, area, and perimeter with less than 2% of error.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.000
metaresearch head score (Gemma)0.000
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: Simulation or modeling
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.303
Threshold uncertainty score0.597

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0000.000
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.000
Science and technology studies0.0000.000
Scholarly communication0.0000.000
Open science0.0000.000
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0000.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.017
GPT teacher head0.288
Teacher spread0.271 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it