Novel iterative optimizing quantization technique and its application to X-ray tomographic microscopy for three-dimensional reconstruction from a limited number of views

The iterative optimizing quantization technique (IOQT) is a novel method in reconstructing three-dimensional (3D) images from a limited number of 2D projections. IOQT can reduce the artifacts and image distortion due to a limited number of projections and limited range of viewing angles. Equivalently, by reducing the number of projections required for reconstruction, the use of IOQT can reduce the dose delivered to the specimen, simplify the complexity of an experimental setup, and consequently support the development of techniques to nondestructively image microstructures of materials. In this article, we will demonstrate the capability of IOQT to reconstruct an accurate 3D image of an object from a limited number of views, using a computer simulation and an actual 3D test pattern experiment with submicrometer features. In addition, we will introduce a promising application of IOQT to X-ray tomographic microscopy to study microbiological specimens by presenting the 3D reconstructions of the two different-conditioned human sperm cells from six projections. (C) 1997 John Wiley & Sons, Inc.