Search

Many damage detection methods that use data obtained from contact sensors physically attached to structures have been developed. However, damage-sensitive features such as the modal properties of steel and reinforced concrete are sensitive to environmental conditions such as temperature and humidity. These uncertainties are difficult to address with a regression model or any other temperature compensation method, and these uncertainties are the primary causes of false alarms. A vision-based remote sensing system can be an option for addressing some of the challenges inherent in traditional sensing systems because it provides information about structural conditions. Using bolted connections is a common engineering practice, but very few vision-based techniques have been developed for loosened bolt detection. Thus, this article proposes a fully automated vision-based method for detecting loosened civil structural bolts using the Viola–Jones algorithm and support vector machines. Images of bolt connections for training were taken with a smartphone camera. The Viola–Jones algorithm was trained on two datasets of images with and without bolts to localize all the bolts in the images. The localized bolts were automatically cropped and binarized to calculate the bolt head dimensions and the exposed shank length. The calculated features were fed into a support vector machine to generate a decision boundary separating loosened and tight bolts. We tested our method on images taken with a digital single-lens reflex camera. [ABSTRACT FROM AUTHOR]

Abstract: Computer vision‐based techniques were developed to overcome the limitations of visual inspection by trained human resources and to detect structural damage in images remotely, but most methods detect only specific types of damage, such as concrete or steel cracks. To provide quasi real‐time simultaneous detection of multiple types of damages, a Faster Region‐based Convolutional Neural Network (Faster R‐CNN)‐based structural visual inspection method is proposed. To realize this, a database including 2,366 images (with 500 × 375 pixels) labeled for five types of damages—concrete crack, steel corrosion with two levels (medium and high), bolt corrosion, and steel delamination—is developed. Then, the architecture of the Faster R‐CNN is modified, trained, validated, and tested using this database. Results show 90.6%, 83.4%, 82.1%, 98.1%, and 84.7% average precision (AP) ratings for the five damage types, respectively, with a mean AP of 87.8%. The robustness of the trained Faster R‐CNN is evaluated and demonstrated using 11 new 6,000 × 4,000‐pixel images taken of different structures. Its performance is also compared to that of the traditional CNN‐based method. Considering that the proposed method provides a remarkably fast test speed (0.03 seconds per image with 500 × 375 resolution), a framework for quasi real‐time damage detection on video using the trained networks is developed. [ABSTRACT FROM AUTHOR]