Probabilistic Evaluation and Filtering of Image Velocimetry Measurements

The recent technological advances in remote sensing (e.g., unmanned aerial vehicles, digital image acquisition, etc.) have vastly improved the applicability of velocimetry hydrological studies. Thus, has become an established technique with acceptable error for practical applications (the can be lower than 10%). main source errors been attributed to incomplete intrinsic and extrinsic camera calibration, non-constant frame rate spurious low velocities due moving objects that are irrelevant streamflow. Some researchers even employed probabilistic approaches (Monte Carlo simulations) analyze uncertainty introduced during calibration procedure. On other hand, endogenous algorithms per se received little attention. In this study, a approach is systematically uncertainty. It argued analysis may not only improve performance methods but it also provide information regarding impact video recording conditions density features, oblique angle, resolution, on accuracy estimated values. suggested method tested six case studies which data previously made publicly available by independent researchers.