CODEC CLASSIFICATION SYSTEM USING RECURRENT NEURAL NETWORK AND METHODS THEREOF

摘要

The present invention relates to a codec classification system using a recurrent neural network, a feature extraction module for extracting a feature of a header from a codec included in a bitstream, a learning module for performing recurrent neural network-based learning using the extracted features, and the It characterized in that it comprises a control module for classifying the input codec based on the learned learning model of the learning module.