UTHM Institutional Repository

Enhanced image watermarking techniques with contour compression methods

Mohammed El Bireki , Majdi Farag (2017) Enhanced image watermarking techniques with contour compression methods. PhD thesis, Universiti Tun Hussein Onn Malaysia.


Download (16MB)


Techniques that hide information have undoubtedly become valuable and useful in commercial, public or private application areas. However, the fast-developing technologies have also caused numerable downsides e.g. illegal transmission, copying, manipulation or duplication of data contents, loss of revenues because of worldwide piracy and copyright problems. Thus, digital watermarking algorithms that provide imperceptibility, robustness and security to combat these problems are highly desirable. Problems that are related to image recognition after watermarking are examined and tested through the attacked watermarked image using a well-defined curve approximation of active contours by compression. The proposed watermarking algorithms embed watermark bits into a cover image using zonal sampling methods in the frequency domains DCT, DWT and joint DWT-DCT. Comparative results show that the proposed watermarking methods are imperceptible, robust and secure against various signal/image processing attacks with PSNR values of 30-35 dB and above, with NCC values of 0.60 and above. Experimental and comparative results for contour compression show that the proposed Adaptive Triangle method has a high compression ratio greater than or equal to 94%. The method's simplicity with accepted level of reconstruction is its main advantage. The combination of contour extraction and contour compression using the Adaptive triangle method from the attacked watermarked image increases value to image recognition ability and security of the proposed method.

Item Type: Thesis (PhD)
Subjects: Q Science > QA Mathematics > QA76 Computer software
Depositing User: Mr. Mohammad Shaifulrip Ithnin
Date Deposited: 02 Jul 2018 03:23
Last Modified: 02 Jul 2018 03:23
URI: http://eprints.uthm.edu.my/id/eprint/10010
Statistic Details: View Download Statistic

Actions (login required)

View Item View Item


Downloads per month over past year