IMPLEMENTATION & TESTING OF WATERMARKING TECHNIQUE FOR OWNERSHIP LABELLING ON DIGITAL IMAGES
Nowadays, digital images could be easily distributed through internet networks. Nevertheless, this capability could trigger a few issues such as: ownership issue, copy <br /> <br /> <br /> right infringement, and originality issue. With the increase number of digital images publi...
Saved in:
Main Author: | |
---|---|
Format: | Final Project |
Language: | Indonesia |
Online Access: | https://digilib.itb.ac.id/gdl/view/15881 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Institut Teknologi Bandung |
Language: | Indonesia |
Summary: | Nowadays, digital images could be easily distributed through internet networks. Nevertheless, this capability could trigger a few issues such as: ownership issue, copy <br />
<br />
<br />
right infringement, and originality issue. With the increase number of digital images published and the existence of group of people who manipulate, download, and <br />
<br />
<br />
reproduce the respective image easily, these will eventually cause many disadvantages to the image owner itself. In order to overcome these issues, a digital watermarking technique is specially designed to provide ownership labels to the digital images. The design of digital watermarking is accomplished by the use of DCT transformation in frequency/transform domain. The scheme of the design consists of watermark insertion and detection/ extraction scheme. The watermark insertion technique is performed by performing a modification on DCT coefficient followed by the insertion of watermark to each high-frequency components of DCT 8x8 from the respective <br />
<br />
<br />
image inputs. For simulations in range 0.01≤α≤0.1, PSNR of watermarking images give a good point that is about more than 40 dB. The correlation value between original watermark and extracted watermark that is achieved from non attack/modification simulation equals one. To know the robustness level of watermark, some attack/modification performed such as noise addition (salt&pepper and Gaussian), median filtering, motion blurring, histogram equalization, and image rotating. |
---|