HoloD3 suite

HoloD3 suite is a set of Python tools for holographic particle measurement and tracking. Its three packages work together to turn time-series holograms into 3D particle trajectories and velocity estimates.

The workflow has three stages: measure particles → link trajectories → smooth trajectories and estimate velocities. Each package passes its results to the next through CSV files.

HoloD3

HoloD3 detects particles in minimum-intensity projections (MinIP) of reconstructed holograms and estimates each particle’s depth and diameter. It produces per-frame 3D particle measurements and an interactive 3D visualization, with command-line, Python, and local web interfaces.

GitHub: dainakai/HoloD3

HoloD3-Linker

HoloD3-Linker connects the per-frame measurements from HoloD3 into temporal trajectories. It combines motion continuity, particle size, and depth to associate particles across frames, assigning track IDs and exporting trajectories for further analysis.

GitHub: dainakai/HoloD3-Linker

HoloD3-Smoother

HoloD3-Smoother smooths the trajectories from HoloD3-Linker and estimates particle velocities. It uses robust polynomial fitting to reduce the influence of depth outliers, automatically selects smoothing parameters, and provides an interactive offline report for comparing raw and fitted trajectories. It runs on the CPU without Julia or CUDA.

GitHub: dainakai/HoloD3-Smoother

Installation instructions and usage examples are available in each repository.