Answers are grounded in this paper and cite their sources.
Large-scale particle image velocimetry (LSPIV) is an efficient and powerful technique for measuring river surface velocities. The raw information for LSPIV consists of images of the river flow recorded from a nearby-elevated position (i.e., bridge, neighboring building). For such situations, image distortion inherently occurs due to the large imaged area and to the oblique recording angle. Images need to be corrected before the conventional image processing is applied to extracted the velocity information. The reliability of the LSPIV measurements significantly depends on the accuracy of the mapping relation, especially when the viewing angle is small. The authors have previously developed a mapping algorithm between the physical plane and the image plane to correct the image distortion. The paper presents a new image transformation procedure which considerably minimizes errors associated with the mapping relation. The efficiency of the newly-developed method is verified by applying it to measurements of a river flow in non flood condition. A novel feature of these measurements is that no artificial (added) seeding was used to trace the water free surface. Evaluation of the spatial velocity resolution is also presented.
No takes yet. Share an insight, caveat, or question.
FUJITA et al. (2000) studied this question.