Horror video recognition based on fuzzy evolution

Fuzzy evolution model is implemented to resolve horror video recognition problem in this paper. Low-level video features include video key-light, video color energy and video rhythm are extracted as typical features. Analytic Hierarchy Process (AHP) is adopted to estimate the weights of extracted features in fuzzy evolution model. Membership function is constructed for horror detection. The experiments results show that the proposed approach has a high precision, high recall rate and F1 measure, and this indicates the availability of our proposed method.