The diffraction patterns of fixed fluorophores are characteristic of the orientation of the molecules' underlying
dipole. Fluorescence localization microscopy techniques such as PALM and STORM achieve super-resolution by
sequentially imaging sparse subsets of fluorophores, which are localized by means of Gaussian-based localization.
This approach is based on the assumption of isotropic emitters, where the diffraction pattern corresponds to a
section of the point spread function. Applied to fixed fluorophores, it can lead to an estimation bias in the range
We introduce a method for the joint estimation of position and orientation of single fluorophores, based on
an accurate image formation model expressed as a 3-D steerable filter. We demonstrate experimental estimation
accuracies of 5 nm for position and 2 degrees for orientation.
We introduce an efficient, image formation model-based algorithm that extends super-resolution fluorescence localization
to include orientation estimation, and report experimental accuracies of 5 nanometers for position estimation and 2 degrees
for dipole orientation estimation.
We propose a method for sub-resolution axial localization of particles in fluorescence microscopy, based on
maximum-likelihood estimation. Given acquisitions of a defocused fluorescent particle, we can estimate its axial position with nanometer range precision.