Recasting Tomo-PIV Reconstruction as Constrained and L1-Regularized Non-Linear Least Squares Problem

Kurzfassung

Tomo-PIV is a volumetric flow measurement system. It is able to obtain instantaneous 3D-3C velocity fields using multiple cameras that observe the same volume of two subsequently illuminated particle distributions from different angles. Part of the measurement technique is the 3D reconstruction of the particle volume from 2D images. This work presents a promising alternative approach to the popular reconstruction algorithms MART and SMART. Simulations and experimental data of a turbulent free jet at Ma = 0.7 are used in order to assess and demonstrate the performance of the newly developed reconstruction algorithm based on constrained least squares strategies and L1-regularization.