Video authentication in HEVC compressed domain / Tew Yiqi
High Efficiency Video Coding (HEVC) is the latest video compression standard finalized in year 2013. While H.264/Advance Video Coding (AVC) is still the mostly deployed video-coding standard, HEVC is gaining ground, especially for storage and transmission of high-resolution videos such as High De...
| Main Author: | |
|---|---|
| Format: | Thesis |
| Published: |
2016
|
| Subjects: | |
| Online Access: | http://studentsrepo.um.edu.my/6730/ http://studentsrepo.um.edu.my/6730/4/yiqi.pdf |
| Summary: | High Efficiency Video Coding (HEVC) is the latest video compression standard finalized
in year 2013. While H.264/Advance Video Coding (AVC) is still the mostly
deployed video-coding standard, HEVC is gaining ground, especially for storage and
transmission of high-resolution videos such as High Definition (HD), 4K, 8K and beyond.
In this thesis, video authentication based on information hiding technique is studied.
The concept of authentication, layout and implementation are presented under the
latest HEVC video compression standard. One of the unique properties of HEVC standard,
i.e., combination of coding unit size, which is sensitive to video manipulation, is
utilized in the proposed information hiding technique. A video authentication scheme is
then put forward by exploiting this unique property of HEVC to embed authentication
code based on a predefined mapping rule. In addition, temporal dependency is enforced,
where the authentication tag generated in one video slice is embedded into its subsequent
slice. Furthermore, multiple layers of authentication are presented to detect and localize
the tampered regions in a HEVC video, as well as verifying the source / sender of the
video using a shared secret key. Moreover, several encryption techniques are presented
to incorporate with the proposed scheme to achieve video authentication in encrypted
domain without compromising on compression efficiency. Video sequences from various
classes (i.e., resolutions) are considered to verify the performance of the proposed
multi-layer authentication scheme. Results show that, at the expense of slight degradation
in perceptual quality, the proposed scheme is robust against video tampering within and
across video slices. Lastly, a functional comparison is performed between the proposed
authentication scheme and the conventional schemes for both plaintext and ciphertext
(encrypted) videos. |
|---|