Special purpose computer for digital holographic particle tracking velocimetry

Area of Science:

• Fluid dynamics
• Optical engineering
• Computational imaging

Background:

• Digital holographic particle tracking velocimetry (DHPTV) is a powerful technique for fluid flow analysis.
• Efficient image reconstruction is crucial for real-time DHPTV applications.
• Current computational methods can be time-consuming, limiting experimental throughput.

Purpose of the Study:

• To design and present a specialized computer system for accelerating DHPTV image reconstruction.
• To implement a fast Fourier transform (FFT)-based pipeline on field-programmable gate arrays (FPGAs).
• To evaluate the performance of the developed system for holographic data processing.

Main Methods:

• Development of a dedicated computer system incorporating four FPGA chips.
• Implementation of an object intensity calculation pipeline using FFT on FPGAs.
• Testing the system with 256x256-grid holograms.

Main Results:

• The system successfully calculates object intensity from holograms using an FPGA-accelerated FFT pipeline.
• The system can generate 100 reconstructed images from a 256x256-grid hologram in just 266 milliseconds.
• Achieved a significant speed-up in holographic image reconstruction.

Conclusions:

• The designed FPGA-based system offers a substantial improvement in computational speed for DHPTV.
• This advancement is expected to significantly enhance the efficiency of DHPTV data analysis.
• The system provides a viable solution for high-throughput particle tracking velocimetry.