In this paper, we propose a new characteristic measure for relative people density and motion dynamics for the purpose of long-term crowd monitoring. While many related works focus on direct people counting and absolute density estimation, we will show that relative densities provide reliable information on crowd behaviour. Furthermore, we will discuss the derivation of a so-called Congestion Level of local areas in the crowd, which takes the current dynamics and density within a certain image region into account. Our density estimation approach is based on a well-known KLT feature tracking algorithm, combined with a post-processing for motion vector association. The resulting feature tracks (tracklets) represent movements of detected objects in the scene. These trajectories are used as basic features for later estimation of track density and relative inertia (changes in motion dynamics), which together are combined to a joint Congestion Level. We show the results of our approach by comparing the characteristic measures of track density and Congestion Level with manually annotated Ground truth data of both artifical and real scenes.