Optimizing feature selection in video-based recognition using Max-Min Ant System for the online video contextual advertisement user-oriented system

The online-advertising has been grown to focus on multimedia interactive model with through the Internet. Our Online Video Advertisement User-oriented (OVAU) system combined the machine learning model for face recognition from camera, multimedia streaming protocols, and video meta-data storage technology. face recognition (FR) is an importance phase which can to enhance the performance of our system. Feature Selection (FS) problem for FR is solved by MMAS-FS algorithms based-on PZMI and DWT features. The features set are represented by digraph G(E, V). Each node used to show the features, and the ability to choose a combination of features is presented the edges connecting between two adjacent nodes. The heuristic information extracted from the selected feature vector as ant's pheromone. The feature subset optimal is selected by the shortest length features and best presentation of classifier. The best subset used to classify the face recognition used Nearest Neighbor Classifier (NNC). The experiments were analyzed on FS shows that our algorithm can be easily applied without the priori information of features. The execution assessed of our calculation is more effective than previous approaches for Video-based recognition based on FS problem.