In this report, we present the sanitization methodology adopted in the Mitik project to prevent any re-identification of the individual who has generated a given trace. We propose for data sanitization strategy in 3 levels following the frame collection process and traces analysis. In the following, we provide more details on the three levels. In the first level, we limit the number and type of collected frames. In the second level, we propose an on-the-fly obfuscation strategy. We finally mask the spatial and temporal information from collected traces in the third level. In addition, we also present a deep performance analysis of the proposed on-the-fly obfuscation solution. Finally, we propose a storage and backup strategy of both collected data and related developed tools for extended code availability.