Video watermarking is very important in many areas of activity and especially in multimedia applications. Therefore, security of video stream has recently become a major concern and has attracted more and more attention in both the research and industrial domains. In this perspective, several video watermarking approaches are proposed but, based on our knowledge, there is no method which verified the compromise between invisibility and robustness against all usual attacks. In our previous work, we proposed a new video watermarking approach based on feature region generated from mosaic frame and multi-frequential embedding. This approach allowed obtaining a good invisibility and robustness against the maximum of usual attacks. In our future work, we propose to optimize the choice of the region of interest by using crowdsourcing technique. This last one is an emerging field of knowledge management that involves analyzing the behavior of users when they