The production of audiovisual content continues to increase, and mainly the content intended for children such as the animated video. However, children are unable to judge the quality, relevance and essentially the adequacy of videos to their ages. In this case, the problem is that many videos can have a negative impact on the psychology and the emotional state of children. Our aim in this work is to identify if an animated video can have a negative impact or not in order to develop a recommendation system. The desired system allows to avoid animated movies causing negative impacts for children. For this, we use a chromatic approach. We propose multi-level descriptors, based on color features, in the judging process. The results of the experimental study of the proposed approach prove its performance and consequently the importance of chromatic features as pertinent features for a recommendation system that aims to reduce the emotional and psychological impacts of children movies and avoid them.