Institute of Fundamental Technological Research

Computer vision (CV) methods for measurement of structural vibration are less expensive, and their application is more straightforward than methods based on sensors that measure physical quantities at particular points of a structure. However, CV methods produce significantly more measurement errors. Thus, computer vision-based structural health monitoring (CVSHM) requires appropriate methods of damage assessment that are robust with respect to highly contaminated measurement data. In this paper a complete CVSHM framework is proposed, and three damage assessment methods are tested. The first is the augmented inverse estimate (AIE), proposed by Peng et al. in 2021. This method is designed to work with highly contaminated measurement data, but it fails with a large noise provided by CV measurement. The second method, as proposed in this paper, is based on the AIE, but it introduces a weighting matrix that enhances the conditioning of the problem. The third method, also proposed in this paper, introduces additional constraints in the optimization process; these constraints ensure that the stiff ness of structural elements can only decrease. Both proposed methods perform better than the original AIE. The latter of the two proposed methods gives the best results, and it is robust with respect to the selected coefficients, as required by the algorithm.

computer vision,structural health monitoring,physics-based graphical models,augmented inverse estimate,model updating,non-negative least square method

The area affected by the earthquake is vast and often difficult to entirely cover, and the earthquake itself is a sudden event that causes multiple defects simultaneously, that cannot be effectively traced using traditional, manual methods. This article presents an innovative approach to the problem of detecting damage after sudden events by using interconnected set of deep machine learning models organized in a single pipeline and allowing for easy modification and swapping models seamlessly. Models in the pipeline were trained with a synthetic dataset and were adapted to be further evaluated and used with unmanned aerial vehicles (UAVs) in real-world conditions. Thanks to the methods presented in the article, it is possible to obtain high accuracy in detecting buildings defects, segmenting constructions into their components and estimating their technical condition on the basis of a single drone flight.

Seismic measurements, Safety, Training data, Earthquakes, Computer vision, Machine learning, Autonomous aerial vehicles, Drones, Synthetic data

In the present study a comparison of frequently used computer vision (CV)-based methods for structural health monitoring of truss structures is shown. The attention is paid to template matching methods that can be classified into one of two groups: area-based and feature-based methods. Synthetic but realistic video is used in this study. Results of the comparison are reliable due to the fact that the exact displacements are known from the finite element model of the investigated structure. From the variety of tested CV methods, the Kanade–Lucas–Tomasi algorithm with FREAK-based repetitive correction outperforms the remaining tested methods in terms of the computation time with a negligibly greater estimation error.

computer vision, structural health monitoring, physics-based graphics models (PBGM), IC-SHM 2021, benchmark test

Despite significant advances in structural health monitoring (SHM), the design of contact sensor networks and their power supply for large-scale structures is still expensive and difficult. Due to the recent progress in computer vision (CV) it is possible to monitor structural components or even whole structures with the aid of digital cameras that allow to avoid the use of the contact sensors. However, CV-based measurements have a significantly lower accuracy than the techniques based on the contact sensors. Moreover, the amount of benchmark data available for development, testing and comparison of CV-based methods is limited. This problem has been partially overcome in recent years by the use of the physics-based graphical models (PBGM) in generation of synthetic but realistic video data. In this work, a comparison of two popular methods of CV-based object tracking applicable in SHM is discussed. PBGM-based videos used in this study are a part of The 2nd International Competition for Structural Health Monitoring'. Exact structural displacements are available due to the fact that PBGM-based video are generated using the structural model. Hence, calculation of the error metrics is straightforward and reliable. The PBGM-based videos show a spatial truss subjected to an unknown excitation.

In the present study a benchmark test of selected methods of template matching-bated methods for computer vision-based object tracking is performed. The attention is paid to compare these methods in terms of estimation of nodal displacements in a flexible truss structure, aiming at assessment of their reliability in Structural Health Monitoring (SHM) applications. Thanks to the use of synthetic but realistic videos generated with the aid of physics-based graphics models (PBGM), exact displacement of tracked structural nodes are known. Therefore, reliable assessment of the accuracy of the examined methods is possible.

computer vision, structural health monitoring, physics-based graphics models (PBGM)

The uNET neural network architecture has shown very promising results when applied to semantic segmentation of biomedical images. The aim of this work is to check whether this architecture is equally applicable to semantic segmentation distinguishing the structural elements of railway viaducts. Artificial images generated by a computer graphics program rendering the 3D model of the viaduct in a photorealistic manner will be used as data sets. This approach produces a large number of
images that provide a solid training set for machine learning model.

Computer vision, deep learning, semantic segmentation