Advances in Computer Vision-Based Civil Infrastructure Inspection and Monitoring

Artificial intelligence in the AEC industry : scientometric analysis and visualization of research activities

Digital twin and its implementations in the civil engineering sector

A review of digital twin in product design and development

Diagnosis algorithms for indirect structural health monitoring of a bridge model via dimensionality reduction

A vision and learning-based indoor localization and semantic mapping framework for facility operations and management

A system and method for large-scale lane marking detection using multimodal sensor data are disclosed. A particular embodiment includes: receiving image data from an image generating device mounted on a vehicle; receiving point cloud data from a distance and intensity measuring device mounted on the vehicle; fusing the image data and the point cloud data to produce a set of lane marking points in three-dimensional (3D) space that correlate to the image data and the point cloud data; and generating a lane marking map from the set of lane marking points.

System and method for large-scale lane marking detection using multimodal sensor data

Track geometry is one of the most important health indices in the maintenance of rail tracks. Visual inspection and inspection using a track-geometry car are two common approaches to inspect track geometry. Presently, using accelerations from in-service trains has become a popular track inspection approach, because it is a low-cost way to monitor the rail tracks more frequently. However, due to the noise presented in the collected accelerations, detecting anomalies using manually designed features often results in many false alarms. In this paper, we propose a learning-based anomaly detection approach for monitoring the longitude elevation of track geometry from the dynamic response of an in-service train. We consider the track geometry with a sudden change as an anomaly, measured by the signal energy of slopes of the track geometry. The proposed approach uses a variational autoencoder (VAE) to detect the anomaly. The VAE takes accelerations as input and learns a mapping from the frequency-domain representation of acceleration signals to a low-dimensional latent space that represents the distribution of the observed data. The reconstruction probability, which measures the variability of the distribution of the input data, is used as an anomaly score for indicating how well the input follows the normal pattern. Compared to distance- and density-based anomaly detection methods, such as K-nearest neighbor and clustering, the VAE-based anomaly detection is robust to measurement noise and prevents overfitting because it captures the underlying distribution of the data in a low-dimensional space. Furthermore, the VAE-based method does not require model-specific thresholds for detecting anomalies because it uses a probabilistic measurement instead of reconstruction error as the anomaly score. We validate the proposed VAE-based approach on the vibration dataset from an in-service train. We show that this approach outperforms a baseline model (an autoencoder-based anomaly detection method) in terms of recall, precision, and F1-score. The proposed method also successfully addresses the overfitting problem presenting in recurrent neural network-based methods. The results make the proposed approach a strong candidate for low-cost and frequent track geometry inspection.

Detecting anomalies in longitudinal elevation of track geometry using train dynamic responses via a variational autoencoder

With rapid developments in 3D imaging technology, as well as the evolving need in Architecture/Engineering/Construction/Facility Management (AEC/FM) industry to understand various field aspects from a 3D perspective, several technologies, such as laser scanning, camera and RGBD camera, are becoming important components of civil engineers’ and architects’ toolbox. With improvements in efficiency and reduction in operational cost, new avenues are opening in leveraging 3D imaging sensors for construction and infrastructure management. In the light of this, there is a need to assess the achievements to date, especially with respect to unique challenging use case scenarios and requirements that construction and infrastructure management domains provide, and consequently identify potential research challenges that still need to be tackled. This paper targets doing such an assessment around several challenges faced when using 3D imaging to support construction and infrastructure management. It specifically discusses to what extent these challenges have been addressed and several approaches that are used in addressing them. It also presents current 3D imaging development trends and discusses briefly some challenges that are emerging due to increased application of 3D imaging techniques to highlight future research needs in this domain.

3D Imaging in Construction and Infrastructure Management: Technological Assessment and Future Research Directions

Conventional demolition approaches of razing a building at the end of its life-cycle generate a large amount of mingled debris, which is difficult to reuse and recycle. Compared to demolition, deconstruction involves disassembling a building systematically and it is a more environmentally friendly alternative. Recent research studies have focused on the transition from demolition to deconstruction to minimize the amount of generated waste and maximize the amount of recycling and reusing material. However, due to tight schedule requirements, extra labor cost, and the lack of drawings and design information, it is difficult for an owner to estimate the cost and duration of deconstruction ahead of time. 3D imaging technologies, such as laser scanning and image-based 3D reconstruction, provide an opportunity to obtain data about as-is conditions at a job site and hence can potentially help in identifying quantities of materials that will be recycled. Existing 3D imaging workflows have two primary limitations: visibility and appearance ambiguity. First, 3D imaging can only capture visible objects before a deconstruction process starts. Also, data captured before deconstruction or at different times during deconstruction can only include a subset of all building components. Second, building components with similar appearances can be made from different materials, resulting in misclassification and errors in quantity estimation. Only a few case studies have discussed how visibility and appearance ambiguity can affect the usage of 3D imaging in deconstruction waste management. In this paper, the authors aim to illustrate the application of 3D imaging during a small-scale deconstruction project in Pittsburgh. Specifically, the authors documented the waste generated during deconstruction manually and by using two different 3D imaging technologies: laser scanning and image-based registration. We then quantified the number of invisible objects and objects with ambiguous appearances at different stages of deconstruction. Through the comparison between the quantity takeoffs from 3D imagery and the ground truth, the paper aims at providing insights on the following questions: 1) How accurate are the quantity estimation and documentation of two 3D imaging technologies (laser scanning and imagery) compared to the actual waste generated? 2) Does 3D imaging capture all components of interest during

Yujie Wei

Papers

A vision and learning-based indoor localization and semantic mapping framework for facility operations and management

System and method for large-scale lane marking detection using multimodal sensor data

Detecting anomalies in longitudinal elevation of track geometry using train dynamic responses via a variational autoencoder

3D Imaging in Construction and Infrastructure Management: Technological Assessment and Future Research Directions

Supporting Deconstruction Waste Management through 3D Imaging: A Case Study