Please use this identifier to cite or link to this item: http://hdl.handle.net/10603/183000
Title: Design and Analysis of Digital Image Watermarking schemes using Machine Learning Algorithms
Researcher: Rajesh Mehta
Guide(s): Navin Rajpal
University: Guru Gobind Singh Indraprastha University
Completed Date: 2015
Abstract: The popularity of internet access makes the communication and distribution of multimedia newlinedata (images, video and audio) very easy. Due to this ease, multimedia data can be newlineduplicated and illegally distributed and thus violating the intellectual property rights. The newlineprotection of multimedia data has become a challenging issue in current scenario. In the last newlinefew years, digital watermarking has become the effective solution for copy right protection, newlinecopy protection and authentication. A number of image watermarking algorithms have newlinealready been developed by various researchers with distinctive applications. newlineThe main objective of this thesis is to design an imperceptible, robust and secure digital newlineimage watermarking schemes for the application of copyright protection, copy protection newlineand ownership assertion. Based on the domain, digital image watermarking schemes are newlinedivided into spatial and frequency domain schemes. In this thesis, the main focus is to build newlinethe frequency domain image watermarking schemes with machine learning algorithms to newlineincrease the imperceptibility and to enhance the robustness against image processing attacks. newlineIn the watermarking applications such as copyright protection, copy protection and newlineownership assertion, high degree of robustness is the basic requirement. The proposed newlineschemes in this thesis are based on fractional discrete Cosine transform (FrDCT) / subband newlineDCT (SB-DCT) / discrete wavelet transform (DWT) / lifting wavelet transform (LWT) newlinefollowed by singular value decomposition (SVD) / QR factorization that can be applied newlinesuccessfully to extract the features of selected image blocks. The blocks are selected using newlinethe statistical property of each image block or the texture analysis (based on fuzzy entropy) newlineof the blocks. The characteristics of human visual system (HVS) model are used to prioritize newlinethe blocks to be altered in order to achieve good visual quality...
Pagination: 
URI: http://hdl.handle.net/10603/183000
Appears in Departments:University School of Information and Communication Technology

Files in This Item:
File Description SizeFormat 
01_coverpage.pdfAttached File97.71 kBAdobe PDFView/Open
02_certificate.pdf32.94 kBAdobe PDFView/Open
03_acknowledgement.pdf20.19 kBAdobe PDFView/Open
04_abstract.pdf27.76 kBAdobe PDFView/Open
05_toc.pdf52.86 kBAdobe PDFView/Open
06_tables.pdf65.32 kBAdobe PDFView/Open
07_figures.pdf71.83 kBAdobe PDFView/Open
08_abbreviations.pdf24.62 kBAdobe PDFView/Open
09_publications.pdf47.88 kBAdobe PDFView/Open
10_chapter_01.pdf206.09 kBAdobe PDFView/Open
11_chapter_02.pdf254.38 kBAdobe PDFView/Open
12_chapter_03.pdf987.33 kBAdobe PDFView/Open
13_chapter_04.pdf1.93 MBAdobe PDFView/Open
14_chapter_05.pdf1.41 MBAdobe PDFView/Open
15_chapter_06.pdf950.57 kBAdobe PDFView/Open
16_chapter_07.pdf109.06 kBAdobe PDFView/Open
17_bibliography.pdf322.7 kBAdobe PDFView/Open
18_biography.pdf4.93 kBAdobe PDFView/Open
Show full item record


Items in Shodhganga are licensed under Creative Commons Licence Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0).

Altmetric Badge: