Digital Inline Holographic PTV using Regularized Inverse Volume Reconstruction

We demonstrate an improved algorithm for digital inline holographic PTV (DIH-PTV) measurements. We utilize an inverse problem formulation whereby the 3D optical field best explaining the recorded hologram is iteratively reconstructed utilizing the sparsity and spatial smoothness of the volume to regularize the solution. The reconstruction is substantially noise-free with dramatically improved axial resolution and increased maximum tracer particle concentration relative to prior DIH-PTV approaches. The use of sparsity regularization enables a sparse data representation which reduces memory requirements and enables processing very large holographic images (5k x 5k) while simplifying the identification and tracking of individual particles. Using synthetic data, we show a 3x improvement in localization accuracy and a similar reduction in the RMS velocity fluctuation in addition to a threefold increase in the allowable tracer concentration. We further present experimental demonstration cases measuring nanofiber dynamics, swimming behaviors of algae, and turbulent channel flow.