Computational imaging techniques such as X-ray computed tomography (CT) rely on a significant amount of computing. The acquired tomographic projections are digitally processed to reconstruct the final images of interest. This process is generically called reconstruction, and it includes additional steps prior to or at the end of the execution of an actual reconstruction algorithm. Most of these steps aim at improving image quality, mainly in terms of artifacts compensation and noise reduction. The reconstructed images are then digitally analyzed to derive quantitative data and to support the qualitative visual interpretation. This part involves computational approaches that fall within the generic term image segmentation. Pre- and post- segmentation image processing is often required to improve the final quantification and extract reliable data from a CT dataset. This chapter presents an overview of the reconstruction and segmentation fundamentals for the 3D analysis of high-resolution X-ray CT data. Better knowledge about artifacts and reconstruction issues avoid misinterpretation of the images. Similarly, more insights about the limitations of image segmentation and quantification help commenting the reliability of the derived numerical values. A deeper understanding of these elements is therefore beneficial to optimize the whole workflow that starts from sample preparation and leads to CT-based scientific results.
