Font Size: a A A

Based On DWT And SVD Domain Watermarking Algorithm Of Color Image

Posted on:2017-09-25Degree:MasterType:Thesis
Country:ChinaCandidate:M L JinFull Text:PDF
GTID:2348330488970894Subject:Software engineering
Abstract/Summary:PDF Full Text Request
With the development of Internet and the popularity of multimedia technology to facilitate access to information at the same time, the infringement also will be easier. How to prevent the occurrence of this phenomenon is an important issue facing the current to take the most important protection is a digital watermarking technology. In comparison, gray-scale image watermarking technology is relatively mature, while color images related to channel selection and color space issues, so the technology can not be directly applied to a color image. This paper studies a color image watermarking technology, mainly as follows:This paper presents the theoretical basis of wavelet transform color image watermarking algorithm. This method watermark image information trichromatic separation of the RGB components separately Arnold transform, Fibonacci transform,magic transform to make improvements in the scrambling process, getting better scrambling effect. Similarly the three primary colors of color host image to give the components do DWT, based on human visual system of the information systems of different colors on different principles embedded watermark information in place of the original image component information. The algorithm can be color watermark into a color image, and after watermark embedding imperceptible, to common data attacks also has strong robustness.Secondly, this paper also presented an color image watermarking algorithm based on wavelet transform and singular value decomposition principle. For each RGB color component of the original color image process wavelet decomposition, after getting the low and mid-band wavelet coefficients then process singular value decomposition with the original watermark image. Then the singular value decomposition again be embedded into three-channel and low and mid-frequency band singular value, and the correspondence between the singular values and vision critical threshold is used to enhance the embedding strength and coordinate operation, through these operations, it is possible to achieve transparency and ensure watermark robustness. The simulation result shows that this algorithm can achieve coordination between the robustness and transparency of watermark, therefore it is a robust algorithm.In this paper, two different algorithms are applied to color image watermark, analysis and simulation results show that good performance can be achieved on the robustness and invisibility.
Keywords/Search Tags:Digital watermark, Color image, Wavelet transform, Singular value decomposition, Human visual system
PDF Full Text Request
Related items