Multiresolution Watermarking For Digital Images by pptfiles

VIEWS: 14 PAGES: 33

									Multi resolution Watermarking For
Digital Images
        Presented by:
     Mohammed Alnatheer
       Kareem Ammar
          Instructor:
      Dr. Donald Adjeroh
  CS591K Multimedia Systems
Overview
Why multi resolution watermarking?
Introduction
Previous Work
Multi resolution representation
Multi resolution watermark embedding
Watermarking extracting procedures
Discussion
Experimental results
Conclusion
Limitations
Improvements/Suggestions
Why multi resolution
watermarking?

• Image quality degradation, the low
  resolution rendition of the watermark will
  still be reserved within the corresponding
  low resolution components of the image.
Introduction

  Digital Watermark was
  proposed as a way to claim the
  ownership of the source and the
  owner.
• In order for watermarking to
  achieve the maximum
  protection, the watermarking
  should be:
Introduction

• Undeletable
• Perceptually invisible
• Statistically undetectable
• Resist to lossy data
  compression
• Resist to common image
  processing operations
• Unambiguous
Previous work
  They use watermark as a symbol or ID
 number which is comprised of a sequence
 of bits, and can only be detected by
 employing the “Detecting Theory”. The
 original image is subtracted from the
 image in question, and the similarity
 between the difference and specific
 watermark is obtained. The similarity is
 compared with a predefined threshold to
 determine whether the image in question
 contains the specific watermark.
Goal of this paper

• In this paper, the watermark is
  visually recognizable pattern,
  which can be extract instead of
  only detected and characterized
  the owner. So, they can use
  both the extracted pattern and
  the similarity measurement for
  determining whether the image
  is watermarked.
Restrictions for embedding

• Watermark imbedding concentrated
  in low frequencies
• Important Watermark features
  should be not be in high frequencies
• Watermarks high rez copies are
  mapped to high host frequencies
• Watermark needs to be randomly
  placed in the image so it is difficult
  to maliciously remove
1) Pyramid Structure of a binary
watermark

• JBIG is used as a resolution
  reduction function of the
  watermark
• Low rez – high rez = differential
  watermark
• no loss of the binary watermark
  under such a reducing and
  reconstructing process.
JBig
2) Wavelet decomposition of the
host image

• Wavelet transform is used to
  hierarchically decompose the host
  image into a series of successively
  lower resolution reference images
  and their associate detail images.
• At each level, the low resolution
  image and the detail images contain
  the information needed to
  reconstruct the reference image at
  next higher resolution level.
Wavelet Transform
3) Pseudo Permutation

• To mix the inlay of the
  watermark for difficult
  extraction
• 2d pseudo-random number
  traversing
   • Number each pixel in watermark in
     ascending order
   • Then randomly order the numbers
4) Block Based image-
dependent permutation
• To improve perceptional
  invisibility
• Pixels are conformed to a
  pattern of the host image
5) Modification of DWT
Coefficients
• Watermarks are imbedded into a
  neighboring relationship on a per rez
  bases
• A residual mask is used to map
  watermark to transformed host
    • Compute residual polarity between
      neighboring pixels
    • Modifiy corresponding pixels of transformed
      host by adding or subtracting polarity
Residual mask
6) Inverse Discrete
Wavelet Transform
• Simply apply the
  inverse discrete
  wavelet
  transform
Multi resolution watermark
embedding
Watermarking extracting
procedures
• We need both the original and the
  watermark images not only for
  obtaining the permutation mapping
  specified in image-dependent
  permutation, but also in computing
  the correlation with the extracted
  result as a watermark verification
  process.
The extraction steps are the following:
Watermarking extracting
procedures
1. Discrete wavelet transformation:
Both the original image and the watermarked image
     are DWT transformed.

2. Generation of the residual results:
By applying the residual masks during the embedding
    step we can generate the residual results for the
    detail images.

3. Extract the permuted differential watermarks:
Perform the exclusive-or (XOR) operation on the
     residual results to obtain the permuted binary
     data.
Watermarking extracting
procedures
4. Reverse the permutations:
  The image dependent permutation mapping
  can be obtain from the original detailed
  image and the differential watermark.
  Then, reverse the image-dependent
  permutation by the image-dependent
  permutation mapping, and followed with
  the pseudo-random permutation according
  to the predefined pseudo-random order.
 Watermarking extracting
 procedures
5. Reconstruct the watermark:
Reconstruct the higher-resolution
  layers to obtain the extracted
  watermarks.
6. Similarity Measurement:
Similarity measurement of the
  extracted watermark and the
  referenced watermark can be
  defined as:
Discussion
1. User Key.
A user key is used as additional feature that
    can be implemented to serve various
    embedding processes by using the same
    embedding technology.
2. Evaluation of Wavelet filters.
Choice of wavelet filters is critical issue that
    affect the quality of the watermarked
    image and the robustness to
    compression attacks.
Experimental results
Image Processing Operation
Image Processing Operation
JPEG lossy compression
JPEG lossy compression
Conclusion

with the characteristics of
 successive approximation, as a
 higher-resolution images are
 obtained, the higher resolution
 watermark will be extracted
Limitations
• This method can only work if the image is
  to be worked is ½ the size of the image to
  watermark. So, we have to place a serious
  restrictions on the type of watermark
  which can be used. Since the watermark
  can be any size, this method isn’t
  appropriate or reliable for variable size
  image which is used as a watermark.
• We used both the watermarked image and
  the original images. Other methods require
  only the watermark image in order to tell
  whether the image is embedded or not.
Limitations
• The evaluation of suitability of wavelet
  filters for invisible watermarking is under
  exploring.
Improvements/Suggestions

• They should extend their method to work
  for also color images.
• This paper was not giving any solution to
  any general form of image manipulation.
Questions??????

								
To top