报告人：Dr. Mijia Yang, Associate Professor and Associate Chair,
Department of Civil and Environmental Engineering, North Dakota State University
会议 ID：724 672 603 会议密码：2021
Computer vision is widely adopted in literature to perform structural health monitoring. The core of computer vision technique is on image processing and its accuracy depends on the quality of input images. It is intuitive quality of input images may be affected by motions of camera induced by external factors such as wind. Direct usage of the images captured by camera would cause errors. A new method to filter the effect of camera motions through background templates is proposed here. Moreover, effect of template sizes and brightness on the accuracy of monitored displacement were analyzed and compared. Several experiments on MTS were performed with different frequencies and amplitudes to verify the method. The results show that filtering of vibrations of the camera significantly improves the displacement monitoring accuracy. Adopting scale factors for every frame and different position in each frame also helps improving accuracy of the monitored displacement. With the accurate displacement monitoring tool in hand, condition assessment of the structure can be conducted smoothly through the modified dynamic response method developed by the authors.
Dr. Mijia Yang is the Associate Professor and Associate Chair of Department of Civil and Environmental Engineering, North Dakota State University, USA. Dr. Yang leads a research program addressing the grand challenge of sustainable infrastructure through innovative materials and advanced assessment methods. His research interests include: (1) Explore advanced structural health monitoring techniques for structural condition assessment, extreme event identification through advanced sensing techniques such as fiber optic sensors, high fidelity dynamic responses of structures, and innovative vehicle bridge and pavement interaction. (2) Create new civil engineering materials for sensing, excitation, strength and stiffness enhancement, material self-healing, and energy harvesting. (3) Develop advanced modeling tools in simulating strength degradation due to fatigue and creep, fracture propagation, microstructure controlled behavior, tailored design of composites, and hydro-chemical-stress interaction. (4) Integrate with industrials and DOTs implementing new PCC pavement repair materials (geopolymer) and develop guidelines for epoxy bonded anchor systems. (5) Enhance engineering education through practice related student projects and research outcomes. These endeavours have been funded by over $1 million in research grants, where Dr. Yang has served as lead or co-principal investigator from funding agencies such as USDOT, TxDOT, MnDOT, NDDOT, and many other local and international agencies and consulting firms.