Perpustakaan Digital ITB

This undergraduate thesis discuss the development of a python-based image processing code to measure velocity fields from given sets of particle images obtained from digital particle image velocimetry (DPIV). Implementation of window shifting and deformation technique as advanced DPIV image processing method is also done to enhance the image processing accuracy. The image processing code also included with addition of stereoscopic calibration, in order to measure three velocity component within two-dimensional space (3C2D). Rotational displacement system is chosen for the stereoscopic con_guration, while the reconstruction method uses 2-d calibration based method. A new vorticity calculation method using Rectangular-Bilinear Interpolation is added to improve the code itself. Several evaluations are performed using particle images for either the DPIV image processing code, stereoscopic calibration and the vorticity calculation to assess their performance in comparison with the previous researches. The results are quite fascinating, where the results from both DPIV image processing and stereoscopic calibration show suffcient accuracy. Unfortunately, the Rectangular-Bilinear Interpolation is proven to be able to enhance the accuracy of vorticity calculation only within several requirements. The whole stereoscopic DPIV image processing program is already implemented with python-based multi-processing module in order to accelerate the computational time and the result is, remarkable.