Cracks are a common defect in civil infrastructures, and their occurrence is often closely related to structural loading conditions, material properties, design and construction, and other factors. Therefore, detecting and analyzing cracks in civil infrastructures can effectively determine the extent of damage, which is crucial for safe operation. In this paper, Web of Science (WOS) and Google Scholar were used as literature search tools and “crack”, “civil infrastructure”, and “computer vision” were selected as search terms. With the keyword “computer vision”, 325 relevant documents were found in the study period from 2020 to 2024. A total of 325 documents were searched again and matched with the keywords, and 120 documents were selected for analysis and research. Based on the main research methods of the 120 documents, we classify them into three crack detection methods: fusion of traditional methods and deep learning, multimodal data fusion, and semantic image understanding. We examine the application characteristics of each method in crack detection and discuss its advantages, challenges, and future development trends.

Roads are the fundamental elements of transportation, connecting cities and rural areas, as well as people’s lives and work. They play a significant role in various areas such as map updates, economic development, tourism, and disaster management. The automatic extraction of road features from high-resolution remote sensing images has always been a hot and challenging topic in the field of remote sensing, and deep learning network models are widely used to extract roads from remote sensing images in recent years. In light of this, this paper systematically reviews and summarizes the deep-learning-based techniques for automatic road extraction from high-resolution remote sensing images. It reviews the application of deep learning network models in road extraction tasks and classifies these models into fully supervised learning, semi-supervised learning, and weakly supervised learning based on their use of labels. Finally, a summary and outlook of the current development of deep learning techniques in road extraction are provided.

With the gradual maturity of the terrestrial laser scanners (TLS) technology, it is widely used in the field of deformation monitoring due to its fast, automated, and non-contact data acquisition capabilities. The TLS technology has changed the traditional deformation monitoring mode which relies on single-point monitoring. This paper analyzes the application of TLS in deformation monitoring, especially in the field of ground surface, dam, tunnel, and tall constructions. We divide the methods for obtaining ground surface deformation into two categories: the method based on point cloud distance and the method based on displacement field. The advantages and disadvantages of the four methods (M2M, C2C, C2M, M3C2) based on point cloud distance are analyzed and summarized. The deformation monitoring methods and precisions based on TLS for dams, tunnels, and tall constructions are summarized, as well as the various focuses of different monitoring objects. Additionally, their limitations and development directions in the corresponding fields are analyzed. The error sources of TLS point cloud data and error correction models are discussed. Finally, the limitations and future research directions of TLS in the field of deformation monitoring are presented in detail.

This paper presents a novel framework to extract metro tunnel cross sections (profiles) from Terrestrial Laser Scanning point clouds. The entire framework consists of two steps: tunnel central axis extraction and cross section determination. In tunnel central extraction, we propose a slice-based method to obtain an initial central axis, which is further divided into linear and nonlinear circular segments by an enhanced Random Sample Consensus (RANSAC) tunnel axis segmentation algorithm. This algorithm transforms the problem of hybrid linear and nonlinear segment extraction into a sole segmentation of linear elements defined at the tangent space rather than raw data space, significantly simplifying the tunnel axis segmentation. The extracted axis segments are then provided as input to the step of the cross section determination which generates the coarse cross-sectional points by intersecting a series of straight lines that rotate orthogonally around the tunnel axis with their local fitted quadric surface, i.e., cylindrical surface. These generated profile points are further refined and densified via solving a constrained nonlinear least squares problem. Our experiments on Nanjing metro tunnel show that the cross sectional fitting error is only 1.69 mm. Compared with the designed radius of the metro tunnel, the RMSE (Root Mean Square Error) of extracted cross sections’ radii only keeps 1.60 mm. We also test our algorithm on another metro tunnel in Shanghai, and the results show that the RMSE of radii only keeps 4.60 mm which is superior to a state-of-the-art method of 6.00 mm. Apart from the accurate geometry, our approach can maintain the correct topology among cross sections, thereby guaranteeing the production of geometric tunnel model without crack defects. Moreover, we prove that our algorithm is insensitive to the missing data and point density.

Ancient architecture, with its long history, has a high cultural value, artistic achievement, and scientific value. The Nanjing City Wall was constructed in the mid-to-late 14th century, and it ranks first among the world’s city walls in terms of both length and size, whether historically or in the contemporary era. However, these sites are subject to long-term degradation and are sensitive to disturbances from the surrounding landscape, resulting in the potential deterioration of the architecture. Therefore, it is urgent to detect the defects and repair and protect Nanjing City Wall. In this paper, a novel method is proposed to detect the surface defects of the city walls by using the unmanned aerial vehicle (UAV) oblique photogrammetry and terrestrial laser scanning (TLS) data. On the one hand, the UAV oblique photogrammetry was used to collect the image data of the city wall, and a three-dimensional (3D) model of the wall was created using the oblique images. With this model, 43 cracks with lengths greater than 30 cm and 15 shedding surfaces with an area greater than 300 cm² on the wall can be effectively detected. On the other hand, the point cloud data obtained by TLS were firstly preprocessed, and then, the KNN algorithm was used to construct a local neighborhood for each sampling point, and the neighborhood was fitted using the least squares method. Next, five features of the point cloud were calculated, and the results were visualized. Based on the visualization results, surface defects of the wall were identified, and 18 cracks with lengths greater than 30 cm and 5 shedding surfaces with an area greater than 300 cm² on the wall were detected. To verify the accuracy of these two techniques in measuring cracks, the coordinates of some cracks were surveyed using a prism-free total station, and the lengths were calculated. The root mean square error (RMSE) of crack lengths based on the UAV oblique photogrammetry model and TLS point cloud model were calculated to be 0.73 cm and 0.34 cm, respectively. The results of the study showed that both techniques were able to detect the defects on the wall surface, and the measurement accuracy could meet the accuracy requirements of the surface defect detection of the city wall. Considering their low cost and high efficiency, these two techniques provide help for the mapping and conservation of historical buildings, which is of great significance for the conservation and repair of ancient buildings.

LiDAR (Light Detection And Ranging) technology is an important means to obtain three-dimensional information of trees and vegetation. However, due to the influence of scanning mode, environmental occlusion and mutual occlusion between tree canopies and other factors, a tree point cloud often has different degrees of data loss, which affects the high-precision quantitative extraction of vegetation parameters. Aiming at the problem of a tree laser point cloud being missing, an individual tree incomplete point cloud restoration method based on local features of the point cloud is proposed. The L₁-Median algorithm is used to extract key points of the tree skeleton, then the dominant direction of skeleton key points and local point cloud density are calculated, and the point cloud near the missing area is moved based on these features to gradually complete the incomplete point cloud compensation. The experimental results show that the above repair method can effectively repair the incomplete point cloud with good robustness and can adapt to the individual tree point cloud with different geometric structures and correct the branch topological connection errors.