Sperm motility analysis is an important part in the fertility analysis and at the same time is a good example of the tracking multiple objects and video surveillance problem from a computational point of view. Currently the most used method for this analysis is the direct inspection which is inaccurate, subjective, a not repeatable procedure, and difficult to teach. This article attempts to overcome these barriers using computer vision techniques and also proposes a heuristic model based on movement direction and Euclidean distance to track sperm in videos obtained from an artificial sperm cells simulator.