Page 84 - IJEEE-2023-Vol19-ISSUE-1

80 |
shares. The watermark must be embedded before it can be
extracted for matching. Thus, extraction is performed by the
receiver, and OR, or XOR is employed during stacking.
According to the results, XOR rather than OR increases
PSNR in this study[8].
Ayu et al., 2019, the suggested method offered dual-layer
fragile digital watermarking, which involves two watermarks
being incorporated into medical images. They used a method
known as the Advanced Encryption Standard. (AES) to
protect the secrecy and validity of the Electronic Patient
Record (EPR). Digital Imaging and Communication in
Medicine (DICOM) tags were embedded in 2LSB as the first
fragile watermark, and their integrity was then verified using
the Secure Hash Algorithm (SHA256). Finally, they divided
the picture into non-overlapping chunks, gave each one an
id, computed the SHA256 for integrity, and used it as a
tamper detector [9].
Gul & Ozturk., 2019, LSB was proposed as a method for
spatial watermarking. They made a watermark by breaking
an image into four sub-blocks and then processing the three
sub-blocks through the SHA-256 hash method. Checking the
watermark's authenticity and integrity is possible by
extracting and comparing it to a hash value. Due to this, the
authentication procedure will be less secure, particularly
against pixel-based attacks such as salt and paper noise [10].
Sinhal et al., 2020 the color picture was separated into 2*4
non-overlapping blocks using pseudo-random, created a
random integer between 0 and 1 and then converted to
binary. The output of this procedure is a 6-bit sequence. It
would then be a fragile watermark embedded. Block-wise
classification as forgery or originally based on neighbor
blocks is performed in the extraction watermark by
performing a series of operations such as LSB and others to
each RGB (red, green, blue) channel. Utilize the six most
critical bits in recovery [11].
Reyes-Reyes et al., 2021, the author suggests a system for
dealing with RGB color pictures by dividing them into non-
overlapping blocks. After that, use Pseudocode on each
channel to create the recovery and authentication watermarks
for each block. In the second phase, one bit of authentication
data for blocks is generated by performing the XOR
operation on each of the three recovery watermarks [12].
Su et al., 2021. Proposed hidden information using the
concept of the Sudoku game. After producing it using a
pseudo-random number generator, they embedded fragile
watermarking. During extraction, compare it to the original.
They employed a two-layer hidden information basis from
the first layer to extract nine candidate pairings. After that,
extract the block index and values index from each pair. They
chose the final pair with the watermark based on Euclidian
distance [13].
III. ARCHITECTURE OF PROPOSED SYSTEM
The suggested type is built on a three-stage design, the first
stage defining the generated watermark. Step two involves
adding a watermark, which is done in a certain technique.
The last step involves the receiver extracting the watermark
from the image and then verifying the integrity of the
embedded watermark. This method relies on a multi-layer
Hussein, Muhajjar, & Mahdi
approach to accomplish its goal at each stage. Figure 1 shows
the general framework of the suggested approach:
Fig. 1: Architecture of proposed system.
A. Stage one (Watermark Generator)
A unique generator using visual cryptography and hashing is
employed to secure our images from this vulnerability. By
ignoring redundant bits, watermark generators minimize the
number of bits concealed in the cover image document.
1) Visual Cryptography (VC)
The proposed technique employs Visual Cryptography (VC)
to generate two random images from a secret image, one for
private sharing and one for public sharing, in order to
guarantee the best level of security and authenticity for the
owner's watermark. A private sharing image is used in the
creation of the watermark. In addition, a secret one is kept to
be utilized in the watermark extraction process and to ensure
that an attacker cannot uncover any hints about a concealed
picture in the original images. The proposed visual
cryptography method entails creating the following step:
• Distribution phase secret encode table
• Reconstruction phase [14]
This approach assumes that an image or message is
composed of black pixels represented as (0) and white pixels
represented as (1). A combination of share stacking and an
OR operation revealed the secret. In other words, the
probability of obtaining a white (or black) pixel is 50 percent
[15].
2) Hash Function
Hashing is Converting random data into a fixed-length
representation (a hash code or message digest) that is often
represented as a hexadecimal number. The hash function
uses an index to get the information that corresponds to the
original value or key. One-way hash functions are the only
function used to determine the hash value from a given input
without the ability to recover the initial information. The
SHA-2 hash method is implemented. While it is an excellent
method for keeping your data safe, it does take a while to
process. A 256-bit digest is generated using SHA-2[16].