Normal view MARC view ISBD view

Image authentication and self-recovery with texture preservation in fragile watermarking / Afrig Aminuddin

By:

Afrig Aminuddin [author.]

Material type: Text

text

Media type:

unmediated

Carrier type:

volume

ISBN:

THE0010059 (Local)

Subject(s):

Dissertation note: Thesis (Doctor of Philosophy) -- Universiti Malaysia Pahang – 2024 Abstract: This research proposes a novel image watermarking scheme with self-recovery capability to authenticate and recover tampered images using the embedded watermark data. The scheme employs a new texture preservation technique to maintain the details of the image texture in the recovered image. The texture set is adaptively computed according to the image content. Each image has its own texture set tailored to the characteristics of the image. The scheme generates the watermark data from each image block, comprising average, range, and texture values. The scheme embeds the watermark data into 2 LSB of the original cover image using a block map and a secret key. The scheme also employs a texture preservation technique to enhance the quality of the recovered image. The scheme utilizes a three-layer authentication process to detect and localize the tampered regions. The scheme solves the issue of tampering coincidence using a new block mapping technique, a tamper coincidence localization, and an image inpainting technique. The experimental results show that the proposed scheme can achieve highquality watermarked and recovered images and high tamper detection and localization accuracy under various attack scenarios. The texture preservation technique produces a high-quality recovered image with a PSNR value of 40.04 dB and SSIM value of 0.9953 under 10% tampering rates on the USC-SIPI dataset. The scheme also outperforms several existing watermarking schemes with self-recovery regarding the imperceptibility of the watermarked and recovered images. Image authentication is crucial for legal validity, ensuring digital evidence is admissible and credible in court. In journalism, it prevents the spread of misinformation, maintaining the integrity of news visuals. Across various sectors, including healthcare and historical archives, it upholds accuracy and trustworthiness in using and preserving digital images.

Tags from this library: No tags from this library for this title. Log in to add tags.

Average rating: 0.0 (0 votes)

Holdings ( 2 )
Title notes ( 4 )

Holdings
Item type	Current library	Call number	Status	Date due	Barcode
Thesis	UMPLIB PEKAN	FKOM .A37 2024 r Thesis (Browse shelf(Opens below))	Not for loan		T000003487
Thesis	UMPLIB PEKAN	CD13728 (Browse shelf(Opens below))	Final Processing		T000003488

Faculty of Computing

Thesis (Doctor of Philosophy) -- Universiti Malaysia Pahang – 2024

Includes bibliographical references

This research proposes a novel image watermarking scheme with self-recovery capability to authenticate and recover tampered images using the embedded watermark data. The scheme employs a new texture preservation technique to maintain the details of the image texture in the recovered image. The texture set is adaptively computed according to the image content. Each image has its own texture set tailored to the characteristics of the image. The scheme generates the watermark data from each image block, comprising average, range, and texture values. The scheme embeds the watermark data into 2 LSB of the original cover image using a block map and a secret key. The scheme also employs a texture preservation technique to enhance the quality of the recovered image. The scheme utilizes a three-layer authentication process to detect and localize the tampered regions. The scheme solves the issue of tampering coincidence using a new block mapping technique, a tamper coincidence localization, and an image inpainting technique. The experimental results show that the proposed scheme can achieve highquality watermarked and recovered images and high tamper detection and localization accuracy under various attack scenarios. The texture preservation technique produces a high-quality recovered image with a PSNR value of 40.04 dB and SSIM value of 0.9953 under 10% tampering rates on the USC-SIPI dataset. The scheme also outperforms several existing watermarking schemes with self-recovery regarding the imperceptibility of the watermarked and recovered images. Image authentication is crucial for legal validity, ensuring digital evidence is admissible and credible in court. In journalism, it prevents the spread of misinformation, maintaining the integrity of news visuals. Across various sectors, including healthcare and historical archives, it upholds accuracy and trustworthiness in using and preserving digital images.