Real-time particle concentration measurement from a hologram by deep learning

Although current methods for measuring the concentration of transparent particles in digital holographic technology are effective, they involve complex procedures and require significant time and computational resources. The objective of this study was to accurately measure particle concentration from a single hologram. Deep learning was employed to measure the quantities of the particles of the same size, and we achieved a relative error less than 10% compared to the ground truth values. This indicates the potential to obtain results closely aligned with actual particle quantities without the reconstruction and denoising processes. The time needed for hologram prediction was at millisecond level, which offers a new possibility for real-time processing.