Computer vision-based structural deformation monitoring techniques were studied in a large number of applications in the field of structural health monitoring (SHM). Numerous laboratory tests and short-term field applications contributed to the formation of the basic framework of computer vision deformation monitoring systems towards developing long-term stable monitoring in field environments. The major contribution of this paper was to analyze the influence mechanism of the measuring accuracy of computer vision deformation monitoring systems from two perspectives, the physical impact, and target tracking algorithm impact, and provide the existing solutions. Physical impact included the hardware impact and the environmental impact, while the target tracking algorithm impact included image preprocessing, measurement efficiency and accuracy. The applicability and limitations of computer vision monitoring algorithms were summarized.

Deep-learning- and unmanned aerial vehicle (UAV)-based methods facilitate structural crack detection for tall structures. However, contemporary datasets are generally established using images taken with handheld or vehicle-mounted cameras. Thus, these images might be different from those taken by UAVs in terms of resolution and lighting conditions. Considering the difficulty and complexity of establishing a crack image dataset, making full use of the current datasets can help reduce the shortage of UAV-based crack image datasets. Therefore, the performance evaluation of existing crack image datasets in training deep neural networks (DNNs) for crack detection in UAV images is essential. In this study, four DNNs were trained with different architectures based on a publicly available dataset and tested using a small UAV-based crack image dataset with 648 +pixel-wise annotated images. These DNNs were first tested using the four indices of precision, recall, mIoU, and F1, and image tests were also conducted for intuitive comparison. Moreover, a field experiment was carried out to verify the performance of the trained DNNs in detecting cracks from raw UAV structural images. The results indicate that the existing dataset can be useful to train DNNs for crack detection from UAV images; the TransUNet achieved the best performance in detecting all kinds of structural cracks.

In the application of a bridge weigh-in-motion (WIM) system, the collected data may be temporarily or permanently lost due to sensor failure or system transmission failure. The high data loss rate weakens the distribution characteristics of the collected data and the ability of the monitoring system to conduct assessments on bridge condition. A deep learning-based model, or generative adversarial network (GAN), is proposed to reconstruct the missing data in the bridge WIM systems. The proposed GAN in this study can model the collected dataset and predict the missing data. Firstly, the data from stable measurements before the data loss are provided, and then the generator is trained to extract the retained features from the dataset and the data lost in the process are collected by using only the responses of the remaining functional sensors. The discriminator feeds back the recognition results to the generator in order to improve its reconstruction accuracy. In the model training, two loss functions, generation loss and confrontation loss, are used, and the general outline and potential distribution characteristics of the signal are well processed by the model. Finally, by applying the engineering data of the Hangzhou Jiangdong Bridge to the GAN model, this paper verifies the effectiveness of the proposed method. The results show that the final reconstructed dataset is in good agreement with the actual dataset in terms of total vehicle weight and axle weight. Furthermore, the approximate contour and potential distribution characteristics of the original dataset are reproduced. It is suggested that the proposed method can be used in real-life applications. This research can provide a promising method for the data reconstruction of bridge monitoring systems.

In this study, we determined the factors that influence of the temperature on an asphalt pavement by developing a two-dimensional unsteady temperature numerical calculation model using the finite difference method and Matlab. Based on the temperatures obtained by a buried sensor in a construction project, we collected the temperatures at different depths in the pavement structure in real time, and we then compared and analyzed the calculated and measured data. The results showed that the temperature in the asphalt pavement structure was significantly correlated with meteorological factors, such as the air temperature, but it also exhibited obvious hysteresis. Compared with the measured data, the maximum deviation in the numerical model based on the variations in the atmospheric temperature and solar radiation was 3 °C. Thus, it is necessary to effectively optimize the selection of asphalt pavement materials by simulating the temperature conditions in the asphalt pavement structure.

This study has introduced a method for fatigue performance evaluation of steel orthotropic bridge decks based on filed measurements for hotspot stress. The long-term stress of the orthotropic bridge deck of Jinchang Road Canal Bridge in Hangzhou, China, was observed by sensors. The fatigue damage and fatigue reliability of the bridge deck were analyzed and evaluated according to the stress. The results show that: (a) the fatigue life of the bridge deck is larger than the design value, indicating that the design of the orthotropic bridge deck is reasonable; (b) the fatigue loads acting at three hotspots can be described by normal distribution; and (c) for the current traffic volume, the fatigue reliability index of the three hotspots is greater than two in the design period. The fatigue reliability index of the three hotspots decreases significantly with the increase of traffic flow, at which time the fatigue damage becomes considerable.