Search

The condition and hazard monitoring of bridges play important roles in ensuring their service continuity not only throughout their entire lifespan but also under extreme conditions such as those of earthquakes. Advanced structural health monitoring (SHM) systems using vision-based technology, such as surveillance, traffic, or drone cameras, may assist in preventing future impacts due to structural deficiency and are critical to the emergence of sustainable and smart transportation infrastructure. This study evaluates several feature detection and tracking algorithms and implements them in the vision-based SHM of bridges along with their systematic procedures. The proposed procedures are implemented via a two-span accelerated bridge construction (ABC) system undergoing a large-scale shake-table test. The research objectives are to explore the effect of refined matching algorithms on blob-based features in improving their accuracies and to implement the proposed algorithms on large-scale bridges tested under seismic loads using vision-based SHM. The procedure begins by adopting blob-based feature detectors, i.e., the scale-invariant feature transform (SIFT), speeded-up robust features (SURF), and KAZE algorithms, and their stability is compared. The least medium square (LMEDS), least trimmed square (LTS), random sample consensus (RANSAC), and its generalization maximum sample consensus (MSAC) algorithms are applied for model fitting, and their sensitivity for removing outliers is analyzed. The raw data are corrected using mathematical models and scaled to generate displacement data. Finally, seismic vibrations of the bridge are generated, and the seismic responses are compared. The data are validated using target-tracking methods and mechanical sensors, i.e., string potentiometers. The results show a good agreement between the proposed blob feature detection and matching algorithms and target-tracking data and reference data obtained using mechanical sensors. [ABSTRACT FROM AUTHOR]

This work proposes a systematic assessment of stereophotogrammetry and noise-floor tests to characterize and quantify the uncertainty and accuracy of a vision-based tracking system. Two stereophotogrammetry sets with different configurations, i.e., some images are designed and their sensitivity is quantified based on several assessments. The first assessment evaluates the image coordinates, stereo angle and reconstruction errors resulting from the stereophotogrammetry procedure, and the second assessment expresses the uncertainty from the variance and bias errors measured from the noise-floor test. These two assessments quantify the uncertainty, while the accuracy of the vision-based tracking system is assessed from three quasi-static tests on a small-scaled specimen. The difference in each stereophotogrammetry set and configuration, as indicated by the stereophotogrammetry and noise-floor assessment, leads to a significant result hat the first stereophotogrammetry set measures the RMSE of 3.6 mm while the second set identifies only 1.6 mm of RMSE. The results of this work recommend a careful and systematic assessment of stereophotogrammetry and noise-floor test results to quantify the uncertainty before the real test to achieve a high displacement accuracy of the vision-based tracking system. [ABSTRACT FROM AUTHOR]

Monitoring large structures using a vision‐based target‐tracking (TT) system while maintaining the full resolution of high‐speed cameras may limit the data size or the sampling rate selection. Also, similar to wireless sensors networks where data loss often occurs during data transmission, TT signals could possibly lose data due to overexposure. The overall goal of this paper is to demonstrate the validity of compressive sensing (CS) for TT time signal processing when faced with challenges such as data loss. The first part of the study is concerned with signal length where TT signals could be further compressed while original data length is already minimum. Next, CS is investigated for improving and recovering TT signals from the possibility of signal loss as well as enhancing sampling rate for system identification purposes. Two case studies were used to obtain the signals for CS processing. The first case included four field‐monitoring tests of a footbridge that were conducted using different sampling rates TT with limited data length to use full cameras resolution. The second case study was concerned with measuring the shifting of the modal properties of a large‐scale bridge model from white‐noise excitation after earthquake loading. The results show that with limited signal length, lower sampling rates, or data loss, CS techniques can successfully improve TT signals. [ABSTRACT FROM AUTHOR]