Abstract
In digital holography an object wave is numerically reconstructed from a recorded hologram. Using this technique it is
possible to detect the position and size of particles in a 3D domain. In this work, particular focus is placed on
quantification of particles with non-spherical morphologies. The in-line configuration is chosen due to the simplicity of
the optical setup and minimal distortions of in-plane morphologies. However, this geometry is also characterized by a
large depth-of-focus and high uncertainty in the detected depth. To quantify these uncertainties, this work begins with
the definition of a non-dimensional model of hologram recording and reconstruction applied to single spherical and nonspherical
particles. Typical CCD noise sources are included. Application of this model to two particle detection
methods reveals the relevant merits and limitations of each particle detection method. From the lessons learned, a new
hybrid particle detection method is proposed. Simulations indicate the hybrid method significantly improves upon the
accuracy of the measured depth and particle morphologies. Furthermore, the proposed method automatically determines
the optimum threshold for each particle, and, therefore, requires minimal user inputs. Finally, initial experimental results
for spherical particles confirm the accuracy of the proposed hybrid method.
