Image Quality Enhancement in X-Ray Microscopy

The goal of this work is to develop digital image processing tools for improving the quality of the images provided by the Göttingen transmission X-ray microscope at BESSY [1, 2]. We first present a method for correcting the inhomogeneous illumination. Then, we present results of image restoration techniques applied to X-ray microscopy images of test objects and of biological specimens containing specific gold probes. Since this restoration process requires knowledge of the point spread function (PSF), we propose a handy method for the measurement of the line spread function (LSF) of the optical system, which can be used as an estimate of the point spread function. Finally, we show the results of applying noise reduction procedures on images of biological specimens. 1 PSF Estimation The knowledge of the PSF is needed for the restoration process. We developed a method for PSF-estimation based on the measurement of the ESF (Edge Spread Function). A gold edge obtained by cutting a gold covered polyimide foil is used as a test-object. To avoid problems with noise, we fit an empirically found function which models the ESF with sufficient precision. The model function we propose is the following : ( ) ESF a a a a a a a a A B ( ) exp( ( )) ( ) exp( ( )2) ( ) ( ) x x x x x x = + + − − + − − − = + 0 1