Cancer is one of the most well-known groups of diseases that finds its cause in cells having chromosomal aberrations. How and why these aberrations can occur is one of the most important questions asked in modern molecular biology. In the last decades it has become clear that gene regulation in the nucleus, where the chromosomes reside, is strongly correlated with structural organization of nuclear components like the telomeres, centromeres and the chromosomes. With new microscopes, better cameras and new fluorescent labels, the demand for analyses of all the images that can be made is growing. The goal of this thesis is the development of image processing and analyses methods for three dimensional (3D) images acquired by fluorescence microscopy. Several microscopy techniques are described, together with two techniques to visualize the nuclear components (chapter 2). In chapter 3 several deconvolution methods are described. Chapter 4 gives two methods to segment the components in the images. Several analyses can be done after segmentation. In chapter 4 we show, after localization, that telomeres from mouse lymphocytes redistribute into a disk-like structure during G2. In chapter 5 we give a novel method to determine the relative length of telomeres by measuring the integrated intensity in the 3D images. Using this information we can define extremely high signals as telomere aggregates. In chapter 6 we use the segmentation and localization techniques to measure the radial redistributions of components in human mesenchymal stem cells.