Assessment and evaluation of damage in civil infrastructure is most often conducted visually, despite its subjectivity and qualitative nature in locating and verifying damaged areas. This study aims to present a new workflow to analyze non-temporal point clouds to objectively identify surface damage, defects, cracks, and other anomalies based solely on geometric surface descriptors that are irrespective of point clouds’ underlying geometry. Non-temporal, in this case, refers to a single dataset, which is not relying on a change detection approach. The developed method utilizes vertex normal, surface variation, and curvature as three distinct surface descriptors to locate the likely damaged areas. Two synthetic datasets with planar and cylindrical geometries with known ground truth damage were created and used to test the developed workflow. In addition, the developed method was further validated on three real-world point cloud datasets using lidar and structure-from-motion techniques, which represented different underlying geometries and exhibited varying severity and mechanisms of damage. The analysis of the synthetic datasets demonstrated the robustness of the proposed damage detection method to classify vertices as surface damage with high recall and precision rates and a low false-positive rate. The real-world datasets illustrated the scalability of the damage detection method and its ability to classify areas as damaged and undamaged at the centimeter level. Moreover, the output classification of the damage detection method automatically bins the damaged vertices into different confidence intervals for further classification of detected likely damaged areas. Moving forward, the presented workflow can be used to bolster structural inspections by reducing subjectivity, enhancing reliability, and improving quantification in surface-evident damage.

Aerial data collection is well known as an efficient method to study the impact following extreme events. While datasets predominately include images for post-disaster remote sensing analyses, images alone cannot provide detailed geometric information due to a lack of depth or the complexity required to extract geometric details. However, geometric and color information can easily be mined from three-dimensional (3D) point clouds. Scene classification is commonly studied within the field of machine learning, where a workflow follows a pipeline operation to compute a series of engineered features for each point and then points are classified based on these features using a learning algorithm. However, these workflows cannot be directly applied to an aerial 3D point cloud due to a large number of points, density variation, and object appearance. In this study, the point cloud datasets are transferred into a volumetric grid model to be used in the training and testing of 3D fully convolutional network models. The goal of these models is to semantically segment two areas that sustained damage after Hurricane Harvey, which occurred in 2017, into six classes, including damaged structures, undamaged structures, debris, roadways, terrain, and vehicles. These classes are selected to understand the distribution and intensity of the damage. The point clouds consist of two distinct areas assembled using aerial Structure-from-Motion from a camera mounted on an unmanned aerial system. The two datasets contain approximately 5000 and 8000 unique instances, and the developed methods are assessed quantitatively using precision, accuracy, recall, and intersection over union metrics.

Efficient and rapid data collection techniques are necessary to obtain transitory information in the aftermath of natural hazards, which is not only useful for post-event management and planning, but also for post-event structural damage assessment. Aerial imaging from unpiloted (gender-neutral, but also known as unmanned) aerial systems (UASs) or drones permits highly detailed site characterization, in particular in the aftermath of extreme events with minimal ground support, to document current conditions of the region of interest. However, aerial imaging results in a massive amount of data in the form of two-dimensional (2D) orthomosaic images and three-dimensional (3D) point clouds. Both types of datasets require effective and efficient data processing workflows to identify various damage states of structures. This manuscript aims to introduce two deep learning models based on both 2D and 3D convolutional neural networks to process the orthomosaic images and point clouds, for post windstorm classification. In detail, 2D convolutional neural networks (2D CNN) are developed based on transfer learning from two well-known networks AlexNet and VGGNet. In contrast, a 3D fully convolutional network (3DFCN) with skip connections was developed and trained based on the available point cloud data. Within this study, the datasets were created based on data from the aftermath of Hurricanes Harvey (Texas) and Maria (Puerto Rico). The developed 2DCNN and 3DFCN models were compared quantitatively based on the performance measures, and it was observed that the 3DFCN was more robust in detecting the various classes. This demonstrates the value and importance of 3D datasets, particularly the depth information, to distinguish between instances that represent different damage states in structures.

Assessment and evaluation of damage in cultural heritage structures are conducted primarily using nondestructive and noncontact methods. One common deployment is laser scanners or ground-based lidar scanners that produce a point cloud containing information at the centimeter to the millimeter level. This type of data allows for detecting surface damage, defects, cracks, and other anomalies based only on geometric surface descriptors using a single dataset, which does not rely on a change detection approach. Moreover, geometric features are not influenced by color, which is essential for heritage structures because they are nonuniform in color due to anthropologic and environmental effects (e.g., painting or moisture). In this work, a damage detection method developed based on local geometric features is evaluated and expanded for crack detection within the example fresco walls of Sala degli Elementi in the Palazzo Vecchio. The workflow’s performance is then compared in a qualitative manner to that of manual crack mapping results identified using images.

Perishable surveying, mapping, and post-disaster damage data typically require efficient and rapid field collection techniques. Such datasets permit highly detailed site investigation and characterization of civil infrastructure systems. One of the more common methods to collect, preserve, and reconstruct three-dimensional scenes digitally, is the use of an unpiloted aerial system (UAS), commonly known as a drone. Onboard photographic payloads permit scene reconstruction via structure-from-motion (SfM); however, such approaches often require direct site access and survey points for accurate and verified results, which may limit its efficiency. In this paper, the impact of the number and distribution of ground control points within a UAS SfM point cloud is evaluated in terms of error. This study is primarily motivated by the need to understand how the accuracy would vary if site access is not possible or limited. In this paper, the focus is on two remote sensing case studies, including a 0.75 by 0.50-km region of interest that contains a bridge structure, paved and gravel roadways, vegetation with a moderate elevation range of 24 m, and a low-volume gravel road of 1.0 km in length with a modest elevation range of 9 m, which represent two different site geometries. While other studies have focused primarily on the accuracy at discrete locations via checkpoints, this study examines the distributed errors throughout the region of interest via complementary light detection and ranging (lidar) datasets collected at the same time. Moreover, the international roughness index (IRI), a professional roadway surface standard, is quantified to demonstrate the impact of errors on roadway quality parameters. Via quantification and comparison of the differences, guidance is provided on the optimal number of ground control points required for a time-efficient remote UAS survey.

Unmanned aerial vehicles (UAVs) have transitioned from a futuristic research concept to becoming a reality for practical, safe, cost-effective bridge inspections. Several studies have used UAVs to capture images from bridges and infrastructure to assess their condition. However, measuring the dynamic responses of bridges using a UAV involves the integration of UAV, sensors, and the use of dynamic equations. Measurement of dynamic transverse displacement especially is a difficult task in the field given the actual constraints of bridges, flights, and sensing under loading events. If transverse displacements could be measured easily, bridge owners could prioritize maintenance operations more cost-effectively by selecting to repair those bridges that move the most under trains. This paper discusses new requirements and solutions for fabricating an enhanced UAV to obtain dynamic transverse displacement benefiting from experiences gained from several field tests. This work follows the regulations of railroad bridge inspection guidelines and considers the aspects for an implementable system in its development. The paper first introduces the preliminary system that has been developed to this end and discusses potential improvements to this system that are identified through multiple field tests. The preliminary UAV system was developed using an algorithm combining the signals from sensors mounted on the UAV to measure dynamic displacements. This paper explains the step-by-step improvement of the existing system which resulted in a successful field test on a real bridge. Subsequently, some modifications and enhancements for the algorithm are proposed which are compatible with the new system data.