We present an image analysis approach for quantifying assembly and disassembly of the Golgi complex in fluorescence microscopy images of live cells. The approach comprises nucleus detection, nucleus tracking, and quantification of Golgi complex localization patterns for single cell nuclei over time. In particular, we define and compare five different measures for quantifying changes in compact and diffuse phenotypes. We propose two measures based on mathematical morphology which take into account shape, size, and contrast of image structures. From the experimental results it turned out that these measures have a higher discrimination power to distinguish compact and diffuse phenotypes than a previous frequently used measure.