Youyi Feng

Date of Graduation


Document Type


Degree Type



Statler College of Engineering and Mineral Resources


Civil and Environmental Engineering

Committee Chair

Fei Dai

Committee Co-Chair

Roger Chen

Committee Member

Radhey S Sharma

Committee Member

Yoojung Yoon


Maintaining the integrity and safety of civil infrastructures such as bridges, dams, tunnels and high-rise buildings is an essential task for civil engineers. Collapse or damage of these civil infrastructures may lead to a tremendous amount of injuries and casualties. To alleviate this situation, a real-time surveillance method enabled by visual sensing techniques is proposed in this thesis. The advances of applying visual sensing techniques, for instance, are allowing practical deployment for large extended systems in a more cost-effective way. Also, the image or video data can be easily used for long-term condition assessments.;The proposed method entails applying visual sensing techniques to measure in-plane deflections and strains of structural members for civil infrastructure applications. In specific, it employs visual sensors (digital/industrial cameras) to capture and record a series of continuous image frames of the targets. Then automated feature detection and matching algorithms are applied to detect and match object features in the consecutive image frames. Based on the location information of the detected features, the in-plane object displacement can be accurately calculated through keeping tracking those features in the continuous image frames. Next, an optimized interpolation procedure is conducted to obtain dense displacement field for the object. And the strains can be consequently recovered from the displacement field through computing its derivatives.;In this research, firstly, the work of evaluating the optimum feature detection and matching algorithm is reported, which is the key task to achieve accurate surveillance. A series of experiments were conducted to compare the three algorithms: DIC (Digital Image Correlation), SIFT (Scale Invariant Feature Transform), and SURF (Speeded-Up Robust Features). The experimental result indicated that the DIC algorithm reveals superiority among the three algorithms and holds the most potential for measuring in-plane deflections and strains of civil infrastructures. To further validate our method, we employed high-speed industrial camera (Manta G223B) to capture a series of continuous image frames of deformed real-world scenarios. The DIC algorithm was adopted for the feature detection and matching process. As the output, the displacement and strains were calculated and then compared with the ground truth in order to evaluate the accuracy performance of the method. Colored strain maps were generated by using different colors to reflect different strain levels in an intuitive way. The experimental result indicated that our method can achieve highly accurate measuring performance of computing in-plane displacements and strains for civil infrastructure applications. The proposed method has several advantages when compared to pre-existing methods (such as sensor networks). It can generate accurate full-field deflections and strains of the target. Besides, the cost-effective equipment and much more convenient set-up procedures will enable engineers to operate periodically and apply for different scales of civil infrastructure applications.