Digital holographic particle volume reconstruction using a deep neural network
This work addresses a domain-specific issue in optical imaging by improving efficiency and resolution for particle analysis, representing an incremental advancement in digital holography.
The paper tackles the problem of slow and low-resolution particle volume reconstruction from in-line holograms by proposing a deep neural network that simultaneously detects particle positions and sizes, achieving faster computation times than conventional diffraction-based methods.
This paper proposes a particle volume reconstruction directly from an in-line hologram using a deep neural network. Digital holographic volume reconstruction conventionally uses multiple diffraction calculations to obtain sectional reconstructed images from an in-line hologram, followed by detection of the lateral and axial positions, and the sizes of particles by using focus metrics. However, the axial resolution is limited by the numerical aperture of the optical system, and the processes are time-consuming. The method proposed here can simultaneously detect the lateral and axial positions, and the particle sizes via a deep neural network (DNN). We numerically investigated the performance of the DNN in terms of the errors in the detected positions and sizes. The calculation time is faster than conventional diffracted-based approaches.