Image filtering in structured illumination microscopy using the Lukosz bound

Abstract

Various aspects of image filtering affect the final image quality in Structured Illumination Microscopy, in particular the regularization parameter and type of regularization function, the relative height of the side bands, and the shape of the apodization function. We propose an apodization filter without adjustable parameters based on the application of the Lukosz bound in order to guarantee a non-negative point spread function. Simulations of digital resolution charts and experimental data of chromatin structures and of actin filaments show artefact free reconstructions for a wide range of filter parameters. In general, a trade-off is observed between sharpness and noise suppression.