The current context of biodiversity loss is particularly marked by the Colony Collapse Disorder (CCD) of honeybees. The CCD combines toxicological (phytosanitary treatments), parasitic (Varroa Destructor), hornets, and viral causes. Consequently, the beekeeper has to face these difficulties in order to maintain the population of bees, to save the species but also to make its exploitation profitable. By leveraging the Internet of Things to collect through sensors a lot of measured data on hives, and submitting these data to machine learning based on algorithms, the apiarist is able to get a fine knowledge of his livestock and also decision-making elements to improve the management of his exploitation, in the long term (selection of strains) or short-term (operational actions). One of the field data is the behavior of bees at the hive entrance. Storch and Mosbeux 1 considers that one can understand what is happening inside the hive by observing what is going on outside, near the hive. We propose to individually capture the flight trajectories of bees and then characterize the pace of the global activity in front of the hive to infer observations that will be consolidated and made available to apicultural data scientists. Firstly, using a camera placed above a beehive entrance such that the entire entrance width is covered by the camera bee are detected and tracked. Several techniques exist to detect flying bees ?, ? and compute the trajectory modeling. In this work, the approach to detect and track bees is an improvement of the method developed by Magnier et al. ? This tracking technique, estimating ellipse orientations and displacements, allows to extract positions of the flying bees, even in the presence of shadows created by bees themselves or objects close to the hive.