PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Sinogram Denoising of Cryo-Electron Microscopy Images
Taneli Mielikäinen and Janne Ravantti
In: ICCSA 2005, 9-12 May 2005, Singapore.

Abstract

Cryo-electron microscopy has recently been recognized as a useful alternative to obtain three-dimensional density maps of macromolecular complexes, especially when crystallography and NMR techniques fail. The three-dimensional model is constructed from large collections of cryo-electron microscopy images of identical particles in random (and unknown) orientations. The major problem with cryo-electron microscopy is that the images are very noisy as the signal-to-noise ratio can be below one. Thus, standard filtering techniques are not directly applicable. Traditionally, the problem of immense noise in the cryo-electron microscopy images has been tackled by clustering the images and computing the class averages. However, then one has to assume that the particles have only few preferred orientations. In this paper we propose a sound method for denoising cryo-electron microscopy images using their Radon transforms. The method assumes only that the images are from identical particles but nothing is assumed about the orientations of the particles. Our preliminary experiments show that the method can be used to improve the image quality even when the signal-to-noise ratio is very low.

PDF - Requires Adobe Acrobat Reader or other PDF viewer.
EPrint Type:Conference or Workshop Item (Paper)
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Machine Vision
Theory & Algorithms
ID Code:985
Deposited By:Taneli Mielikäinen
Deposited On:18 June 2005