In this study, we investigate a novel idea of using synthetic images of bolts which are generated from a graphical model to train a deep learning model for loosened bolt detection. Firstly, a framework for bolt-loosening detection using image-based deep learning and computer graphics is proposed. Next, the feasibility of the proposed framework is demonstrated through the bolt-loosening monitoring of a lab-scaled bolted joint model. For practicality, the proposed idea is evaluated on the real-scale bolted connections of a historical truss bridge in Danang, Vietnam. The results show that the deep learning model trained by the synthesized images can achieve accurate bolt recognitions and looseness detections. The proposed methodology could help to reduce the time and cost associated with the collection of high-quality training data and further accelerate the applicability of vision-based deep learning models trained on synthetic data in practice.

Recognition of spalling on surface of concrete wall is crucial in building condition survey. Early detection of this form of defect can help to develop cost‐effective rehabilitation methods for maintenance agencies. This study develops a method for automatic detection of spalled areas. The proposed approach includes image texture computation for image feature extraction and a piecewise linear stochastic gradient descent logistic regression (PL‐SGDLR) used for pattern recognition. Image texture obtained from statistical properties of color channels, gray‐level cooccurrence matrix, and gray‐level run lengths is used as features to characterize surface condition of concrete wall. Based on these extracted features, PL‐SGDLR is employed to categorize image samples into two classes of “nonspall” (negative class) and “spall” (positive class). Notably, PL‐SGDLR is an extension of the standard logistic regression within which a linear decision surface is replaced by a piecewise linear one. This improvement can enhance the capability of logistic regression in dealing with spall detection as a complex pattern classification problem. Experiments with 1240 collected image samples show that PL‐SGDLR can help to deliver a good detection accuracy (classification accuracy rate = 90.24%). To ease the model implementation, the PL‐SGDLR program has been developed and compiled in MATLAB and Visual C# .NET. Thus, the proposed PL‐SGDLR can be an effective tool for maintenance agencies during periodic survey of buildings.