Total variation based image restoration of three dimensional microscopic objects

The inverse problem involving the determination of a three-dimensional biological structure from images obtained by means of optical-sectioning microscopy is ill-posed. Regularization methods must often be used in order to obtain a reasonable solution. Recently, the total variation (TV) regularization, as proposed by Rudin, Osher and Fatemi (1992), has become very popular for this purpose. An iterative algorithm is used for minimizing a TV-penalized least squares problems. We also employ transform based methods for solving large linear subproblems arising from TV-penalized least squares problems. Preliminary numerical results show that the method performs quite well.