Miter gates are structures that act as both the damming surface and doorway of lock chambers and are commonly found on U.S. rivers. The geometry of a miter gate is such that the structure is similar to a very deep, cantilevered beam with a channel-like cross-section. Generally, a miter gate is assumed to have very little torsional stiffness, which is an issue due to the torsional loads imparted on the gate as it swings open and closed through the river. Moreover, the channel-like cross-section of the gates means that the gate will torsionally deflect due to its own weight. Thus, long, slender, post-tensioned steel members (termed diagonals) are attached across the diagonal dimensions of the miter gate to address the lack of torsional stiffness and counteract the twist of the gate. Maintaining an appropriate level of tension in these diagonals is of critical importance to both the structural performance and serviceability of the miter gate; however, the number of diagonals that need to be monitored, the difficulty of accessing the diagonals, and the geographic distribution of miter gates makes traditional contact sensors economically impractical for monitoring the tension in diagonals. Accordingly, this study investigates non-contact methods for monitoring the tension in miter gate diagonals. The expected low frequency of vibration of the diagonals makes vision-based methods attractive. Initial efforts of this study are focused on the feasibility of an approach wherein a region of interest of the frames of a video of a vibrating diagonal is used as a virtual sensor from which the displacement of the diagonal can be estimated by means of optical flow measurements. Using the time history of displacement obtained from the virtual sensor, the dominant frequencies of vibration of the diagonal are estimated and used to determine the tension in the diagonal using beam theory. The efficacy of the approach is demonstrated by means of a scale-model lab experiment.