Non-degraded HEVC video steganography method capable of resisting deep learning network detection

A deep learning network and video steganography technology, applied in the field of information hiding of digitally encoded video, can solve the problems of low embedding efficiency and increased video bit rate, and achieve the effect of low embedding cost, improved security and strong resistance.

Active Publication Date: 2021-04-30
SOUTH CHINA UNIV OF TECH +1
View PDF8 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] On the other hand, Hu Yongjian and others published a paper "Large-capacity lossless HEVC information hiding method for modifying flag bits" in the "Journal of South China University of Technology (Natural Science Edition)" in 2018, and proposed a modified motion vector candidate list index The steganographic method, although this

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Non-degraded HEVC video steganography method capable of resisting deep learning network detection
  • Non-degraded HEVC video steganography method capable of resisting deep learning network detection
  • Non-degraded HEVC video steganography method capable of resisting deep learning network detection

Examples

Experimental program
Comparison scheme
Effect test

Example Embodiment

[0079] Example 1

[0080] In this embodiment, a piece of HEVC coded video is used as the carrier video to be embedded in the encrypted message, a text file in TXT format is used as the secret information file, and the TXT file is embedded into the HEVC coded video to describe the implementation process of the present invention in detail;

[0081] This embodiment provides a non-degraded HEVC video steganography method that can resist deep learning network detection, including secret letter embedding and extraction;

[0082] like figure 1 As shown, MeSign embedding includes the following steps:

[0083] S1: Convert the secret information file into a binary bit stream S, and calculate the length of the code stream, marked as L;

[0084] S2: Set the embedding rate α. In this embodiment, α=0.2 is set. In this embodiment, the embedding rate α can be set according to the length of the secret letter, so that the secret letter can be embedded in each GOP (group of pictures) in a dist...

Example Embodiment

[0129] Example 2

[0130] This embodiment provides a non-degraded HEVC video steganography system capable of resisting deep learning network detection, which is provided with a secret letter embedding module and a secret letter extraction module;

[0131] The secret letter embedding module includes: secret letter binarization unit, embedding rate first setting unit, current embedding unit construction unit, prediction block number first statistical unit, embedding index sequence construction unit, distortion cost calculation unit, loading A dense index sequence generating unit, an index value judging unit and a motion vector residual updating unit;

[0132] The secret message binarization unit is used to convert the secret information file into a binary bit stream S, and calculate the code stream length L;

[0133] The embedding rate first setting unit is used to set the embedding rate of secret letter embedding;

[0134] The current embedding unit construction unit is used ...

Example Embodiment

[0149] Example 3

[0150] This embodiment provides a storage medium, the storage medium can be a storage medium such as ROM, RAM, magnetic disk, optical disk, etc., and the storage medium stores one or more programs. When the programs are executed by the processor, the performance of the first embodiment can be realized. Non-degradable HEVC video steganography against deep learning network detection.

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a non-degraded HEVC video steganography method capable of resisting deep learning network detection. A secret message embedding step is implemented by the following substeps of: counting the number of prediction blocks of a GOP in a video by using an AMVP technology; constructing a secret embedding index sequence; calculating the distortion cost of carrying out +/-1 modification on prediction indexes of the prediction blocks in sequence according to a decoding sequence; embedding a secret message into the index sequence by utilizing STC coding; comparing index value change conditions before and after secret embedding, and updating a motion vector residual error; and entropy-encoding the modified data again and writing the encoded data back to a video code stream. An extraction step is implemented by the following substeps of: counting the number of prediction blocks of the GOP in the video by using the AMVP technology; constructing and extracting a secret message index sequence; setting an STC code parity check matrix according to an embedding rate, and extracting a secret message; repeatedly extracting until the secret message is completely extracted. The non-degraded HEVC video steganography method has the advantages of unchanged video quality and unchanged motion vector statistical distribution after steganography, and can effectively resist a deep learning network detector taking the reduction of the quality of the secret-embedded video as a detection basis.

Description

technical field [0001] The invention relates to the technical field of information hiding of digitally coded videos, in particular to a non-degraded HEVC video steganography method capable of resisting deep learning network detection. Background technique [0002] As a new-generation video compression standard replacing the H.264 standard, the HEVC standard has better compression performance and is more suitable for the transmission and storage requirements of high-definition and ultra-clear resolution videos, and has been applied on a large scale. Research on steganographic methods for HEVC compressed video is of great significance. The steganography of HEVC video mostly follows the steganography strategy of previous video compression standards (H.264 or MPEG), including spatial domain steganography and compressed domain steganography. Popular steganography in the compressed domain has methods of modifying compression coding parameters such as motion vectors, intra predict...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): H04N1/32H04N19/176H04N19/467H04N19/70H04N19/91G06N3/04G06N3/08
CPCG06N3/04G06N3/08H04N1/32149H04N19/176H04N19/467H04N19/70H04N19/91
Inventor 胡永健刘烁炜刘琲贝王宇飞
Owner SOUTH CHINA UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products