International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056
Volume: 03 Issue: 05 | May-2016 www.irjet.net p-ISSN: 2395-0072
© 2016, IRJET | Impact Factor value: 4.45 | ISO 9001:2008 Certified Journal | Page 1714
REVERSIBLE WATERMARKING WITH DENOISING USING SOURCE
CODING TECHNIQUE
M.Yuvaraju
1
,N.Manju Sunthari
2
1
Assistant Professor, Department of EEE, Anna University Regional Campus, Coimbatore, Tamil Nadu, India
2
PG Scholar , Department of EEE, Anna University Regional Campus, Coimbatore, Tamil Nadu, India
---------------------------------------------------------------------------------***------------------------------------------------------------------------------
Abstract -Watermarking is a widely used technique for
image security and authentication. An important
application in watermarking is image protection against
tampering. This paper proposes an efficient watermarking
algorithm that detects the tampered zones and recovers
the information lost in those zones. The original image is
source coded in the watermark embedding scheme and the
output bit stream is protected using appropriate channel
encoder. For image recovery, erasure locations that can be
detected by check bits help channel erasure decoder to
retrieve the original image. The watermarked image
quality gain is achieved through spending less bit-budget
on watermark. The image recovery quality is being
improved well as a consequence of consistent performance
of designed source and channel codes. The design is used
to protect and provide self recovery for image security
applications and can remove the noise in case of noisy
image during the recovery. Experimental results show that
the proposed scheme outperforms other similar methods
in terms of image quality before and after watermarking.
Keywords: Image watermarking, Source coding,
Tampering protection, Noise removal, Self recovery.
1.INTRODUCTION
In the present globalized world, the internet availability
and diverse image processing tools reach a greater
degree, that there are chances of downloading an image
from the internet, and one can manipulate it without the
permission of the rightful owner. For such reasons image
authentication has become an active and also vital
research area. Embedding watermarks in both signals
and images may cause distortion. A successful
watermarking method satisfies some fundamental
requirements. Many number of media specific hash
functions have been proposed for multimedia
authentication. A multimedia hash is nothing but a
content-based digital signature of the media data. Hence
to generate a multimedia hash, a secret key is used to
extract many features from the data. Those extracted
features are further processed to form the hash. The
hash is transmitted along with the media by appending
or simply embedding it to the primary media data [8]. At
the receiver side, the authenticator uses the same secret
key to generate back the hash values, which are
compared to the ones transmitted along with the data for
verifying the authenticity. Also in addition to the content
authentication, multimedia hashes are being used in
content based retrieval from databases. To search for
any multimedia content, some negative methods like
sample by sample comparisons are inefficient. Also,
these methods are used to compare the lowest level of
content representation and they do not offer robustness
in such situations as geometric distortions.
Robust image hash functions can be used for the
problem solving. Usually the hash is computed for each
data entry in the database and is stored with the original
data and they are in the form of a look-up table. Hence
when it is needed to search for a given query in the
database, its computed hash value is compared with the
data entry corresponding to the matched item, in terms
of certain hash-domain distance that often is responsible
for content similarity, is then fetched. The hash has much
smaller size than the original media and the hash value
matching is more efficient. Image hash functions have
also been used in image and video watermarking
applications. The hash functions are widely used as
image-dependent keys for watermarking. Some
techniques called fragile watermarks are used in
authentication and image tampered zone location. Even
it can be used for recovering lost information in the
tempered zones. Also inceptive watermarks can verify
the image integrity with robustness. Recent methods
achieve 100% localization against many attack varieties.
Some watermarking algorithms of error concealment
restore information in tampered parts that are
previously detected. Further techniques can accomplish
the tampering localization [1] and error concealment by
a single watermark. The self recovery watermarking
trend is being in recent interest [2]. Also some
conventional error control coding methods are adopted
to solve the above problem.
In the past techniques discrete cosine transform (DCT)
coefficients of host image [3] is embedded in least
significant bit (LSB) of original image. A binary image
generated by the difference between the host image and
its chaotic pattern , the original image hash and
watermarks derived by the coefficients of wavelet
transform, a quantized vector of original image
[11].Watermark bits are classified into check bits and
reference bits. The check bits localize the tampered bits
and reference bits restore the original image in
tampered area. For content restoration the reference bits
are embedded into other bits. In some blocks when both
the bits are detected to be tampered then the content
recovery may result in failure. The problem is known as