Wavefront sensorless schemes for correction of aberrations induced by biological specimens require a time invariant property of an image as a measure of fitness. Image intensity cannot be used as a metric for Single Molecule Localization (SML) microscopy because the intensity of blinking fluorophores follows exponential statistics. Therefore a robust intensity-independent metric is required. We previously reported a Fourier Metric (FM) that is relatively intensity independent. The Fourier metric has been successfully tested on two machine learning algorithms, a Genetic Algorithm and Particle Swarm Optimization, for wavefront correction about 50 μm deep inside the Central Nervous System (CNS) of Drosophila. However, since the spatial frequencies that need to be optimized fall into regions of the Optical Transfer Function (OTF) that are more susceptible to noise, adding a level of denoising can improve performance. Here we present wavelet-based approaches to lower the noise level and produce a more consistent metric. We compare performance of different wavelets such as Daubechies, Bi-Orthogonal, and reverse Bi-orthogonal of different degrees and orders for pre-processing of images.
Kayvan Forouhesh Tehrani, Luke J. Mortensen, and Peter Kner, "Wavelet-based denoising of the Fourier metric in real-time wavefront correction for single molecule localization microscopy," Proc. SPIE 9717, Adaptive Optics and Wavefront Control for Biological Systems II, 971707 (Presented at SPIE BiOS: February 13, 2016; Published: 15 March 2016); https://doi.org/10.1117/12.2213531.
Conference Presentations are recordings of oral presentations given at SPIE conferences and published as part of the proceedings. They include the speaker's narration with video of the slides and animations. Most include full-text papers. Interactive, searchable transcripts and closed captioning are now available for most presentations.
Search our growing collection of more than 26,000 conference presentations, including many plenaries and keynotes.