Information hiding method based on H.265 video pixel domain

Abstract

The invention discloses an information hiding method based on an H.265 video pixel domain. The information hiding method comprises the steps of embedding and extracting; in an embedding part, first performing preprocessing of secret information before an analysis of a high layer, namely an NALU unit, for video decoding; after a video is decoded to entropy decoding, embedding according to a pixel value of a time domain layer of a video source and an embedding rule, and after the embedding is finished, continuing to perform the video decoding at a CTU (Code Tree Unit) level; and in an extracting part, obtaining a carrier video which is embedded and is transmitted, and performing configuration of a video code; and before the code is compressed by an encoder, obtaining video time domain pixel data of a current CTU, performing extraction of the secret information according to an extraction rule, and after the secret information of the current CTU is extracted completely, continuing to perform encoding work of the CTU. According to the method provided by the invention, a video steganography is performed at a time domain level in video encoding and decoding, and an LSB bit of the time domain pixel value is directly modified, so the method has the advantages of easy implementation and good video sensory quality; and the increasing rate of embedded information to a code rate of the carrier video is very small.

Description

An Information Hiding Method Based on H.265 Video Pixel Domain technical field The invention belongs to the technical field of information security, and more specifically relates to an information hiding method based on the H.265 video pixel domain. Background technique Video hiding is a technology that hides secret information in audio and video for authorized parties to receive, but unauthorized parties cannot perceive its transmission behavior and content. So far, researchers at home and abroad have proposed many steganographic algorithms that use network data streams (video, text, images, etc.) in existing networks to carry out covert communication. These steganographic algorithms are divided into three categories according to the location of the embedded information. The first category is the pre-embedded video steganographic algorithm that embeds secret information in the original video domain. It focuses on embedding information in the original video stream and inde...

AI Technical Summary

Problems solved by technology

The first type, the pre-embedding algorithm, has the advantage of strong undetectability because it is embedded in the original domain, but it is basically based on the H.264 / AVC video codec standard or the previous technical standard as the background, and has not Fully consider the update of the international official mainstream video codec technology, the performance of the video coding compression rate is poor, so it has insufficient practicality; especially in today's video resolution requirements are getting higher and higher, represented by 1080p of high-definition video has gradually become the mainstream trend of online video

However, the video steganography scheme based on H.264 cannot fully meet the requirements of low compression rate and high resolution.

The second type, the built-in embedding algorithm, emphasizes the combination of the encoder and the embedding process. If it is embedded in the DCT coefficients before quantization, the secret information will be lost due to the quantization process, so this type of method is not robust; Embedded in the final DCT coefficients, although the quantization process can be skipped, the visual effect is not good and its implementation is complicated

The third type of method is the post-embedding algorithm, which is easy to implement, but because the modified carrier is the video code stream, the performance of undetectability is poor, and the concealment is not strong enough.

Images