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.

Numerous damage detection methods that use data obtained from contact sensors, physically attached to structures and human inspection methods have been developed. However, damage sensitive features used for these methods such as modal properties of steel and reinforced concrete structures are sensitive to environmental conditions such as temperature and humidity. Besides, human inspection is cost, labor extensive, and is controlled by the technical understanding of an individual. The uncertainties of the contact sensor methods are difficult to address with a regression model or any other temperature compensation method, and these are primary causes of false alarms. In order to address some of these challenges of the traditional sensing system, a vision-based remote sensing system can be one of the alternatives as it gives the explicit intuitions of structural conditions. In addition, bolted connections are common engineering practices, and very few vision-based techniques are developed for loosened bolt detection. Thus, this thesis proposes an automated vision-based method for detecting loosened structural bolts using the Viola-Jones algorithm and support vector machines. The test images of bolt connections are taken with a digital single lens reflex camera. The Viola-Jones algorithm is trained on two datasets of images with and without bolts. The trained algorithm localizes and crops all the bolts on test images. The SVM is trained on another dataset of loose and tight bolts to generate decision boundary for classification of the loosened and tight bolts. The cropped bolt images are binarized to calculate bolt features as head dimensions and exposed shank length. The extracted features are fed into a trained support vector machine to classify the loosened and tight bolts. We test our method on images taken by digital single lens reflex and smartphone cameras.iii A special feeling of gratitude to my loving parents. Their words of encouragement, support, and prayers helped me to come this far.iv