A Pseudo-High Bit Rate HEVC Video Detection Method Based on Convolutional Neural Network

Abstract

A method for detecting pseudo-high bit rate HEVC video based on convolutional neural network. By calculating the average detection score of HEVC video and comparing it with a threshold to determine whether it is a pseudo high bit rate video. The average detection score is obtained by averaging the detection scores of all I-frames of the video. The calculation method of the detection score of the I frame includes: decompressing the video, extracting the block size of all PUs and the intra prediction mode of the PU in the luma component in the I frame; constructing the PU block size information map and the PU prediction mode information of the I frame Figure; Calculate the square area with the largest spatial complexity in the grayscale image of the decompressed frame of the I frame; construct the PU block size information subgraph and the PU prediction mode information subgraph of the I frame, and input the dual-channel convolution based on the attention mechanism Neural network to get detection scores for I frames. The invention combines the encoding information graph with the neural network based on the attention mechanism, which can effectively improve the detection performance and enhance the robustness to different encoding settings and video content.

Description

technical field [0001] The present invention relates to the technical field of multimedia security, in particular to a method for detecting pseudo-high bit rate HEVC video based on a convolutional neural network. Background technique [0002] With the rapid development of digital video processing technology and network transmission technology, digital video has become one of the important ways for people to get in touch with the latest information, and has been widely used in many fields such as entertainment, justice, finance, medical treatment and education. Video bit rate is often regarded as an important indicator of digital video picture quality, and high bit rate video has better picture quality. However, advanced video editing software, such as Adobe Premiere and FFmpeg, can easily convert low-bit-rate videos to high-bit-rate videos. Such up-converted videos are called pseudo-high-bit-rate videos. In contrast, a video that has undergone only one encoding process is c...

AI Technical Summary

Problems solved by technology

However, this method has the following limitations: 1) This method requires multiple re-encoding and decoding during the feature extraction process, and the operation efficiency will be significantly reduced when the input video resolution is high

2) This method constructs a characteristic curve according to the objective quality of the decompressed frame, which is easily affected by different transcoding parameters, such as different video coding standards, etc., and has poor robustness

This algorithm has fast operation efficiency, but has the following disadvantages: 1) The detection feature of this method contains the division mode information of the inter-coded prediction unit, so it cannot detect the pseudo-high bit rate video that only contains I frames

2) This method uses the frequency of occurrence of different encoded information to construct detection features, which cannot reflect the spatial distribution of encoded information

This algorithm performs well in the case of a single video transcoding setting, but it has the following shortcomings: 1) When the encoding parameters of the video to be tested are different from the re-encoding process of extracting re-encoding errors, the detection performance of the method There will be a significant decline, and the reliability is poor, which is not conducive to the application in actual forensics scenarios

2) The composite neural network used in this method uses a simple splicing operation to fuse input features from different sources, which can easily cause too many limitations of network parameters and increase the risk of overfitting training samples

Images