The tissue microenvironment is a major contributor to cellular functions, such as cell adhesion, migration and invasion. A critical physical parameter for determining the effect of the microenvironment on cellular functions is the average pore-size of filamentous scaffolds, such as 3D collagen fiber matrices, which are assembled by the polymerization of biopolymers. The scaffolds of these matrices can be analyzed easily by using state-of-the-art laser scanning confocal imaging. However, the generation of a quantitative estimate of the pore-size in a 3D microenvironment is not trivial. In this study, we present a reliable and fast analytical method, which relies on a two-step 3D pore-size analysis utilizing several state-of-the-art image analysis methods, such as total variation (TV) denoising and adaptive local thresholds, and another crucial parameter, such as pore-coverage. We propose an iterative approach of pore-size analysis to determine even the smallest and obscure pores in a collagen scaffold. Additionally, we propose a novel parameter, the pseudo-pore-size, which describes a virtual scaffold porosity. In order to validate the advanced two-step pore-size analysis different types of artificial collagens, such as a rat and bovine mixture with two different collagen concentrations have been utilized. Additionally, we compare a traditional approach with our method using an artificially generated network with predefined pore-size distributions. Indeed, our analytical method provides a precise, fast and parameter-free, user-independent and automatic analysis of 3D pore topology, such as pore-sizes and pore-coverage. Additionally, we are able to determine non-physiological network topologies by taking the pore-coverage as a goodness-of-fit parameter.