Semantic Video Segmentation with Using Ensemble of Particular Classifiers and a Deep Neural Network for Systems of Detecting Abnormal Situations

Abstract

A new approach based on the use of a deep neural network and an ensemble of particular classifiers is proposed. This approach is based on use of the novel block of fuzzy generalization for combines classes of objects into semantic groups, each of which corresponds to one or more particular classifiers. As result of processing, the sequence of frames is converted into the annotation of the event occurring in the video for a certain time interval.