| Author: | Law, Sai Chung |
| Title: | Investigation of digital image forensics techniques for video application |
| Advisors: | Law, Bonnie (EIE) |
| Degree: | Eng.D. |
| Year: | 2022 |
| Subject: | Image analysis Image processing -- Digital techniques Digital forensic science Hong Kong Polytechnic University -- Dissertations |
| Department: | Faculty of Engineering |
| Pages: | ix, 66 pages : color illustrations |
| Language: | English |
| Abstract: | Nowadays, digital video cameras and smartphones with video cameras are easily affordable and very popular. People can also easily access image editing tools to manipulate images in pictures or videos, so source device identification on video applications has become a significant issue. Research works have been done and well documented to address the problem of source camera identification for digital images, mainly concentrated on linking the given image to its device. However, the research works on video source identification are still relatively few and new compared with those on image source identification. In this research, a signal-based detection method using sensor pattern noise is proposed for verifying the source of the received video, or video in custody. Different challenges for video source identification, such as the effects of image content, image compression and infra-red dark/night scene, have been investigated. The signal used for detection is the photo-response non-uniformity (PRNU) sensor pattern noise, which is unique and intrinsic in every digital image taken by a source camera, like the human fingerprints and signatures. Experiments with video sequences taken by our own proprietary cameras, as well as from the public data set, have been conducted to investigate the operating performance of the proposed signal-based source verification (SSV) system using PRNU. The results demonstrate that the SSV system is useful for reliable video source identification in both network- and cloud-based video surveillance. Furthermore, in this IoT era, security video systems have become part of the IoT devices which are typical of limited resources such as low computation, power, storage and memory. To address these problems in the IoT applications, improved approaches for the SSV system are studied and modified to reduce the computational complexities in operation, while maintaining constant accuracy of source camera verification. By comparison with the state-of-the-art technique of a related work that employs spatial domain averaged (SDA) frames for extraction of reference PRNU, it has been shown that the improved SSV system can also provide reliable source camera identification. Besides, the improved version of the proposed system can speed up the execution time of the detection algorithm by about 25% at the optimal case, when compared to the basic scheme of the SSV system. In conclusion, this research project contributes to the implementation of PRNU-based source attribution methods which are robust, but effective and efficient in the operation, especially for the application of smart video surveillance in the cloud and IoT domains. |
| Rights: | All rights reserved |
| Access: | restricted access |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| 6610.pdf | For All Users (off-campus access for PolyU Staff & Students only) | 3.42 MB | Adobe PDF | View/Open |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/12154