Self-Recovery in Fragile Image Watermarking Using Integer Wavelet Transform

Ernawan, Ferda and Afrig, Aminuddin and Eh Phon, Danakorn Nincarean and Alsheikh, Emad Addin and Wibowo, Adi (2022) Self-Recovery in Fragile Image Watermarking Using Integer Wavelet Transform. In: 2022 IEEE 8th International Conference on Smart Instrumentation, Measurement and Applications (ICSIMA) , Melaka, Malaysia. pp. 21-25. (183892). ISSN 978-166548800-6

Preview	Pdf Self-Recovery in Fragile Image Watermarking.pdf Download (553kB) | Preview
	Pdf Self-Recovery in Fragile Image Watermarking Using Integer Wavelet Transform.pdf Restricted to Repository staff only Download (304kB) | Request a copy

Abstract

The growth of internet technology makes digital images can be easily distributed and modified to change image information. This research proposes fragile image watermarking based on LSB Adjustment on integer wavelet transform. This study focuses on self-recovery and authenticating the integrity of image contents. The proposed scheme embeds a watermark into the two-LSB by using LSB adjustment. The authentication bits are randomly located in the two-LSB with a secret key. The recovery data is generated based on IWT for the recovery process. The proposed watermarking method uses multi-layer authentication to achieve high precision and accuracy. This research also uses image inpainting techniques to improve the recovery under tamper attacks. The experimental results produce high imperceptibility with PSNR of 45.5 dB and SSIM of 0.997. The proposed scheme produces better self-recovered image quality than the existing scheme with PSNR of 37.98 dB and SSIM of 0.9928 against a tampering rate of 10%

Item Type:	Conference or Workshop Item (Paper)
Subjects:	Q Science > QA Mathematics > QA76 Computer software T Technology > T Technology (General)
Faculty/Division:	Faculty of Computing
Depositing User:	Miss Amelia Binti Hasan
Date Deposited:	18 Oct 2023 07:20
Last Modified:	18 Oct 2023 07:24
URI:	http://umpir.ump.edu.my/id/eprint/38919
Download Statistic:	View Download Statistics

Actions (login required)

View Item