Design of an endoscopic 3D Particle-Tracking Velocimetry system and its application in flow measurements within a gravel layer

In this thesis a novel method for 3-D flow measurements within a permeable gravel layer is developed. Two fiberoptic endoscopes are used in a stereoscopic arrangement to acquire image sequences of the flow field within a single gravel pore. The images are processed by a 3-D Particle-Tracking Velocimetry (3-D PTV) algorithm, which yields the three-dimensional reconstruction of Lagrangian particle trajectories. The underlying image processing algorithms are significantly enhanced and adapted to the special conditions of endoscopic imagery. This includes methods for image preprocessing, robust camera calibration, image segmentation and particle-tracking. After a performance and accuracy analysis, the measurement technique is applied in extensive systematic investigations of the flow within a gravel layer in an experimental flume at the Federal Waterways Engineering and Research Institute in Karlsruhe. In addition to measurements of the pore flow within three gravel pores, an extended experimental setup enables the simultaneous observation of the near-bed 3-D flow field in the turbulent open-channel flow above the gravel layer and of grain motions in a sand layer beneath the gravel layer. The interaction of the free surface flow and the pore flow can be analyzed for the first time with a high temporal and spatial resolution. The experiments are part of a research project initiated by an international cooperation called Filter and Erosion Research Club (FERC). The longterm goal of this project is to quantify the influence of turbulent velocity and pressure fluctuations on the bed stability of waterways. The obtained experimental data provide new insight into the damping behaviour of a gravel bed and can be used for comparison with numerical, analytical and phenomenological models.