Random-subset fitting of digital holograms for fast three-dimensional particle tracking

DSpace/Manakin Repository

Random-subset fitting of digital holograms for fast three-dimensional particle tracking

Citable link to this page

 

 
Title: Random-subset fitting of digital holograms for fast three-dimensional particle tracking
Author: Dimiduk, Thomas Gwilym; Perry, Rebecca Wood; Fung, Jerome; Manoharan, Vinothan N.

Note: Order does not necessarily reflect citation order of authors.

Citation: Dimiduk, Thomas G., Rebecca W. Perry, Jerome Fung, and Vinothan N. Manoharan. 2014. Random-subset fitting of digital holograms for fast three-dimensional particle tracking. Applied Optics 53, no. 27: G177-G183.
Full Text & Related Files:
Abstract: Fitting scattering solutions to time series of digital holograms is a precise way to measure three-dimensional dynamics of microscale objects such as colloidal particles. However, this inverse-problem approach is computationally expensive. We show that the computational time can be reduced by an order of magnitude or more by fitting to a random subset of the pixels in a hologram. We demonstrate our algorithm on experimentally measured holograms of micrometer-scale colloidal particles, and we show that 20-fold increases in speed, relative to fitting full frames, can be attained while introducing errors in the particle positions of 10 nm or less. The method is straightforward to implement and works for any scattering model. It also enables a parallelization strategy wherein random-subset fitting is used to quickly determine initial guesses that are subsequently used to fit full frames in parallel. This approach may prove particularly useful for studying rare events, such as nucleation, that can only be captured with high frame rates over long times.
Published Version: doi:10.1364/AO.53.00G177
Terms of Use: This article is made available under the terms and conditions applicable to Other Posted Material, as set forth at http://nrs.harvard.edu/urn-3:HUL.InstRepos:dash.current.terms-of-use#LAA
Citable link to this page: http://nrs.harvard.edu/urn-3:HUL.InstRepos:12748673
Downloads of this work:

Show full Dublin Core record

This item appears in the following Collection(s)

 
 

Search DASH


Advanced Search
 
 

Submitters