Abstract:Structural health monitoring (SHM) has been conducted by placing sensors on structures, so-called contact sensors, to measure displacements, strains, and accelerations of engineering structures. Contact sensing is accurate and effective, but suffers from some critical shortcomings, such as high cost, intensive labor, and traffic interruption. With the advancement and wide availability of digital video cameras, compounded with the development and enhancement of computer vision algorithms, vision-based sensing using a video camera as a sensor for SHM has become a viable alternative and a complementary method to remotely capture structural responses. Several computer vision algorithms are proposed, together with specific type of video camera, for capturing and processing the SHM video for the dynamic structural responses, e.g. displacements of selected targets. Up to date, there is a limited number of successful applications using video camera as a sensor in SHM practice. To facilitate the real-world applications of vision-based SHM, this paper presents the integrated methods, and a software tool for analyzing SHM videos using template matching algorithms along with a subpixel method that enhances the accuracy of the vision-based structural responses. The integrated approach is flexible and versatile for not only extracting displacements, strains, velocities, and accelerations, but also detecting damage of the desired multiple points or areas (templates) from videos. The methods and the software tool have been applied to a comprehensive analysis of the lab structural test, which was video-recorded by a third party without any involvement of the authors or any intension in using video camera as sensor. Good agreement for the lab test has been achieved for the structural responses (e.g. displacement and strain) and the damage detection recorded by the conventional sensors and extracted from the video by applying the integrated tool. The approach has been further validated on a highway bridge in the field, where a contact-based SHM system is permanently installed on the bridge and serves a good baseline reference for validating the proposed approach. The vision-based SHM results obtained for the highway bridge application show good correlation with the structural responses captured by conventional sensors and indicate that the developed analysis methods and software tool perform well for highway bridge test and structural health monitoring.

Vision-based fatigue crack detection using global motion compensation and video feature tracking

Vision-based inspection of out-of-plane fatigue cracks in steel structures

Computer Vision and Augmented Reality for Human-Centered Fatigue Crack Inspection

A Three-Step Computer Vision-Based Framework for Concrete Crack Detection and Dimensions Identification

Automated Vision-Based Fatigue Crack Growth Monitoring Based on Background Subtraction

Machine Learning-Based Real-Time Visible Fatigue Crack Growth Detection

Machine Learning Based Features Matching for Fatigue Crack Detection

3D vision technologies for a self-developed structural external crack damage recognition robot

Vision‐based fatigue crack automatic perception and geometric updating of finite element model for welded joint in steel structures

Crack Length Measurement Using Convolutional Neural Networks and Image Processing

An Anti-Occlusion Vision-Based Method for Structural Motion Estimation

Vision-based automatic structural motion estimation in presence of 3D camera motion effects

Propagation Measurement for Visually Thin Fatigue Crack Using Homography Mapping Error and Digital Image Correlation

Real-Time Video Surveillance Based Structural Health Monitoring of Civil Structures Using Artificial Neural Network

Continuous crack detection using the combination of dynamic mode decomposition and connected component-based filtering method

Vision-based concrete crack detection technique using cascade features

Structural Health Monitoring of Fatigue Cracks for Steel Bridges with Wireless Large-Area Strain Sensors

Binocular Video-Based 3D Reconstruction and Length Quantification of Cracks in Concrete Structures

Automated Crack Detection on Concrete Bridges

Development of video analytics with template matching methods for using camera as sensor and application to highway bridge structural health monitoring

Estimating Structural Motions in Extreme Environmental Conditions——a Dynamic Correlation Filter Based Computer Vision Approach