International Journal of Computational and Experimental

Science and ENgineering

(IJCESEN)
Vol. 10-No.3 (2024) pp. 434-460
http://www.ijcesen.com
Copyright © IJCESEN ISSN: 2149-9144
Research Article

A Secure Fusion: Elliptic Curve Encryption Integrated with LSB Steganography

for Hidden Communication
Homam EL-TAJ

Cybersecurity Department, Dar Al-Hekma University, Jeddah, Saudi Arabia

* Corresponding Author Email: [email protected] ORCID: 0000-0001-7565-4760

Article Info: Abstract:

DOI: 10.22399/ijcesen.382 In today's digital age, ensuring secure communication is essential. This article presents a
Received : 16 July 2024 novel approach for hidden communication by integrating Elliptic Curve Encryption
Accepted : 02 September 2024 (ECE) with Least Significant Bit (LSB) Steganography.
Our proposed fusion offers a robust solution, Stegno Curve for concealing sensitive
Keywords : information within innocuous cover media while encrypting it using elliptic curve
cryptography. By leveraging the strengths of both techniques, we achieve enhanced
Elliptic Curve Encryption security and confidentiality in data transmission.
LSB Steganography Through a comprehensive literature review, methodology explanation, security analysis,
Secure Communication and implementation details, we demonstrate the feasibility and effectiveness of the
Hidden Communication Stegno Curve Method. The findings of this study not only contribute to advancing the
Image Steganography field of secure communication but also open avenues for practical applications in various
domains, such as secure cloud transitions, smart home technologies, and data encryption.

1. Introduction

The imperative for secure and confidential data

transmission has become more pressing in an era
marked by the exponential growth of digital
communication and increasing cyber threats. From
sensitive financial transactions to confidential
government communications, the potential
ramifications of data breaches are profound [1, 2].
Moreover, with the emergence of stringent data
protection regulations and the growing awareness of
privacy rights among individuals, organizations are
under increasing pressure to adopt robust security
measures.
Figure 1 shows the number of user accounts exposed Figure 1: Number of User Accounts Exposed Worldwide
from data breaches globally, for each quarter of 2020 From 1st Quarter 2020 To 4th Quarter 2023 (In
through Q4 2023. Data is given in millions, and Millions)
across the period, the numbers fluctuate wildly. The
highest number of exposed accounts was in Q4 2020 This has led to the development and utilization of
at about 125.74 million accounts. A peek at the trend various techniques, such as encryption and
reveals the peak of data breaches near the end of steganography, to ensure the confidentiality and
2020, followed by a general decrease, with spikes in integrity of sensitive information [1,2,12-25].
Q1 2021 and Q2 2023. This chart outlines the Encryption involves converting data into a form that
ongoing threat from data breaches and the critical can only be read by authorized parties, while
need to increase measures for security in digital steganography focuses on concealing the existence
communications. of the communication. However, existing methods
 Homam EL-TAJ / IJCESEN 10-3(2024)434-460

face challenges in providing integrated and robust ensuring secure and covert communication in the
solutions for secure communication [26-32]. digital domain.
The following sections of the paper will be
1.1 Problem Statement structured in this manner:
1. Literature Review: This section presents a
While encryption and steganography offer effective synopsis of the recent literature regarding
ways of securing digital communication Elliptic Curve Encryption (ECE) and Least
independently, they present inherent limitations Significant Bit (LSB) Steganography,
[24]. Although highly secure, traditional encryption highlighting both the weaknesses and
methods, such as RSA and AES, may exhibit strengths of each method when used alone
vulnerabilities to emerging cryptographic attacks [1, and in conjunction.
3, 17]. Similarly, conventional LSB steganography 2. Proposed Stegno Curve Methodology:
techniques often suffer from poor payload capacity The methodology is elaborated on,
and susceptibility to detection. It is effective in explaining the processes involved in key
hiding the message [11] but may not provide generation, encryption, embedding,
sufficient protection for the content [13]. Therefore, extraction, and decryption in detail. This
there is a growing need for a combined and secure section also includes mathematical
solution that integrates encryption and formulations and algorithms used.
steganography to address these limitations. 3. Implementation and Results:
Implementation details with the details of
1.2 Purpose of the Article the security analysis carried out.
Performance metrics, system specifications,
This article aims to introduce a novel approach that and the real-world effectiveness of the
integrates Elliptic Curve Encryption (ECC) with proposed method (table 1).
Least Significant Bit (LSB) Steganography to 4. Discussion: Analysis of results, comparison
provide a comprehensive and secure communication with existing methods, practical
solution [17, 32]. This integration leverages the implications, and possible applications of
strengths of both techniques, offering enhanced the Stegno Curve Method.
security and privacy for digital communication by 5. Conclusion and Future Work: It is an
encrypting the data using elliptic curve cryptography overview of the key findings, contribution to
before embedding it within the LSB layers of cover the field, and future research trends that
media, we not only fortify the confidentiality of the would lead to enhancing secure
concealed information but also mitigate the risk of communication techniques further.
detection or tampering. This integration enhances
the security posture of the communication channel 2. Literature Review
and introduces a novel paradigm for covert data
transmission that transcends conventional The literature review for the article includes an
boundaries [17, 15, 25]. overview of Elliptic Curve Encryption (ECC), LSB
This paper presents an innovative way of combining Steganography, existing integration approaches, and
Elliptic Curve Encryption and Least Significant Bit research on the usage of these algorithms.
Steganography to propose the Stegno Curve Method
in order to contribute to secure communications. The 2.1 Overview of Elliptic Curve Encryption
use of ECE for resource-constrained environments
considers the weaknesses of ECE that allow for Elliptic Curve Cryptography (ECC) has emerged as
robust encryption with shorter key lengths, together a powerful encryption method renowned for its
with the strengths. The same encrypted data is efficiency and security [1]. Unlike traditional
embedded in digital media surreptitiously using LSB methods such as RSA, ECC offers equivalent
steganography. This integration increases the security with shorter key lengths, making it
security and secrecy of covert communication with particularly suitable for resource-constrained
an overall solution well-defended against environments such as mobile devices and IoT
cryptographic attacks and steganalysis. The Stegno devices [3].
Curve Method is potent enough for applications that Elliptic Curve Cryptography (ECC) has garnered
demand high security and stealth, such as secure significant attention due to its ability to provide high
cloud transitions, smart home technologies, and security with shorter key lengths compared to
sensitive data transmission. The article will also traditional methods like RSA. ECC's efficiency
highlight the potential advantages of this fusion, makes it suitable for resource-constrained
demonstrating its applicability and effectiveness in environments such as mobile devices and IoT
435
 Homam EL-TAJ / IJCESEN 10-3(2024)434-460

systems [32-51]. Recent advancements include AI- vulnerabilities. [39] present a distinctive encryption
enhanced Schoof’s algorithm for faster elliptic curve strategy for securing healthcare data within IoT-
point computations and new methods for generating enabled health services, utilizing elliptic curve
ECDSA moduli to improve Barrett’s algorithm, cryptography, Advanced Encryption Standard
thereby enhancing ECC's overall performance [47]. (AES), and Serpent. The primary challenge in this
Siddharth and Saini [33] describe lightweight study is to design an encryption method that robustly
cryptography as a form of encryption with minimal protects sensitive patient information from
CPU requirements, designed to ensure data security unauthorized access and breaches, without
and privacy for IoT systems, which often have introducing unnecessary complexity and
limited capabilities. Traditional encryption methods performance issues due to the combined use of ECC
are unsuitable for mobile devices due to these and AES.
constraints. Beyond addressing security and privacy Xavier and Kesavan [40] propose a hybrid elliptic
concerns, it is crucial to consider the overall curve cryptography technique for IoT-based
efficiency of the verification process. However, the systems, involving the selection of the secret key for
authors do not provide a comprehensive evaluation verification via particle swarm optimization or the
of the security offered by various lightweight cuckoo search algorithm. Additionally, this
cryptographic algorithms. They merely assert that technique incorporates encoding and decoding
these techniques resist cryptanalysis attacks without through the Key-Value Pair (KVP) of Elliptic Curve
providing detailed specifics. Side-channel attacks Cryptographic. Despite its innovative approach, the
represent a significant threat to the security of algorithm's security poses a concern when compared
cryptographic systems, particularly for resource- to similar methods in terms of execution speed and
limited IoT devices. Siddharth and Saini's work does accuracy. Qazi et al. [41] offer a method that secures
not explore the impact of side-channel attacks on the node-to-node communication systems and optimizes
security of lightweight cryptographic methods. Their memory usage on nodes by employing the Elliptic
paper reviews existing schemes without discussing Curve Digital Signature Algorithm (ECDSA). This
potential improvements for these methods. approach provides an effective mechanism for
In contrast, Ahmed et al. [34] propose a robust monitoring key reproduction rates, greeting message
cybersecurity approach that combines encryption counts, and packet sizes. However, the study lacks a
with authentication (ELCA), employing an elliptic discussion on how the proposed protocol maintains
curve Diffie-Hellman (ECDH) protocol for key effectiveness and adaptability over time as new
allocation, effectively addressing concerns security challenges arise. [42] suggests a novel
regarding weak bits in the shared secret key. While technique based on elliptic curve cryptography
this study offers a valuable contribution to IoT (ECC) for verification and encoding, utilizing
security, it does not consider the implementation randomized numbers generated by fuzzy logic to
costs of the proposed mechanism, which could be a enhance authentication and encryption in IoT
significant obstacle for resource-constrained systems. Nonetheless, the paper is limited by
devices. specific application requirements, usability aspects,
Elliptic Curve Encryption (ECE) is a cryptographic and the challenge of adapting FECC to resource-
technique based on the mathematical properties of constrained environments while ensuring security.
elliptic curves over finite fields. It offers a high level The research by [43] explores IoT security, covering
of security while requiring smaller key sizes symmetrical, asymmetrical, and hybrid
compared to traditional encryption algorithms like cryptographic solutions. Asymmetric key
RSA and AES [35, 36]. The security of ECE relies cryptography is highlighted for its ability to secure
on the difficulty of solving the elliptic curve discrete communications among multiple individuals while
logarithm problem (ECDLP), making it resistant to preventing key distribution over insecure channels.
attacks by classical and quantum computers [37]. Elliptic Curve Cryptography (ECC) is found to
ECC offers faster encryption and decryption than outperform other techniques in the study. The
traditional algorithms, contributing to efficient critical challenge remains the suitability of
cryptographic operations. algorithms for resource-constrained IoT devices,
Nyangaresi [38] introduces an anonymity-oriented underscoring the need to select cryptographic
lightweight methodology tailored for IoT-based algorithms compatible with the specific limitations
systems, leveraging elliptic curve point and requirements of IoT systems.
multiplication techniques and one-way hashing. This Pawar and Kalbande [44] emphasize the concept of
study grapples with the inherent complexity and contemporary access control systems based on
error-prone nature of Elliptic Curve Cryptography elliptic curve encryption, highlighting the security of
(ECC), particularly in terms of parameter selection, IoT components and applications. The primary
key management, and addressing ECC-specific challenge identified in this study is the development
436
 Homam EL-TAJ / IJCESEN 10-3(2024)434-460

of efficient cryptographic libraries or methods presents a significant challenge. Lara-Nino et al. [51]
tailored to the constraints of IoT devices. Agrawal conduct a review to determine the requirements that
and Tiwari [45] explore the security and privacy enable an ECC-based system to be lightweight and
measures available at each tier of the IoT model, suitable for use in practical, constrained
along with the limitations and future directions for applications. Nonetheless, the actual deployment
enhancing IoT security infrastructure. A reliable and standardization of elliptic curve cryptography
channel is essential for encrypted transmission; their lightweight cryptographic systems for real-world
study also discusses elliptic curve lightweight applications remain major research concerns.
encryption technology for trust management in IoT Dhivya C, Sharmila G, Keerthanadevi S, and
systems. Although the paper might address attacks, Gangalakshmi S. have found widespread adoption of
protocols, trust management, and ECC individually, ECE in various domains, including secure
a significant challenge is the integration of these communication protocols, digital signatures, and
components into an effective security architecture identity authentication systems [15]. Organizations
for IoT. such as financial institutions, government agencies,
Khan et al. [46] propose a secure framework for IoT- and cloud service providers rely on ECE to
based medical sensors using enhanced elliptic curve safeguard sensitive data and protect against
cryptography. This approach introduces an unauthorized access [32].
additional secret key to bolster system security. However, ECC implementations require careful
However, the study lacks thorough real-world parameter selection and rigorous key management
implementation and validation, which remains a practices to mitigate potential vulnerabilities.
critical issue. Kavitha et al. [47] present a framework Furthermore, concerns have been raised about the
for addressing security issues using a hyperelliptic potential impact of quantum computing on the
curve-based public key cryptosystem incorporating security of ECC, prompting ongoing research into
the Digital Signature technique. The study does not post-quantum cryptographic alternatives.
adequately address potential vulnerabilities, threats,
and attack vectors that the framework might 2.2 Overview of LSB Steganography
encounter in real-world environments.
Mehmood et al. [48] conduct a comprehensive Least Significant Bit (LSB) Steganography is a
literature review of data encryption methods in cloud form of data hiding technique that embeds secret
computing and IoT environments. They assert that information within the least significant bits of digital
traditional encryption techniques are impractical for cover objects, such as images, audio files, or text
IoT devices and suggest developing a lightweight documents [14, 16]. LSB steganography operates on
encryption solution to protect data transfer and the principle of imperceptibility, aiming to conceal
communication within the IoT ecosystem. The hidden data while minimizing perceptible changes to
research also considers various encryption the cover medium [15, 20, 28].
algorithms used in cloud computing and IoT LSB Steganography is widely used for data hiding
contexts. However, adapting these strategies to the because of its simplicity and efficiency. However, it
unique and diverse conditions of the IoT ecosystem faces challenges such as low payload capacity and
poses a significant challenge. Vahdati et al. [49] susceptibility to statistical attacks. To overcome
devise a strategy to maintain secrecy and privacy these limitations, hybrid approaches combining LSB
while ensuring the availability of IoT ecosystem with encryption algorithms like AES and Blowfish
services. Their study compares the efficiency of have been proposed. These approaches enhance
RSA (Rivest Shamir Adleman) and ECC (Elliptic security by increasing the complexity of the
Curve Cryptography) techniques to identify the most encryption and embedding processes, thereby
portable, secure, and effective implementation for improving the robustness of the steganographic
IoT, focusing on securing IoT resources such as method [52-59].
sensors, data, networks, and users. The problem of In [52], the authors aimed to achieve protection by
resource efficiency and scalability is particularly implementing a modified steganography technique.
relevant when examining cryptographic techniques The proposed method employed an altered Least
like ECC and RSA in the context of IoT devices. Significant Bit (LSB) algorithm. In [53], the author
Kumari et al. [50] propose an ECC-based posited that steganography is one of the most
authentication strategy for IoT and cloud servers. effective methods for concealing hidden information
The study employs the widely recognized and used within a cover object. The primary method utilized
Automated Internet Safety Protocols Verification for this purpose was image steganography, with a
and Apps tool to evaluate the security features of the focus on the Least Significant Bit (LSB) technique,
proposed scheme. However, addressing the scheme's a leading approach in the spatial domain of image
resistance to evolving security threats over time steganography.
437
 Homam EL-TAJ / IJCESEN 10-3(2024)434-460

The LSB data embedding method plays a pivotal of a cover image. The proposed PVM technique
role in the field due to its support by numerous data ensures high-quality stego images. Experimental
embedding algorithms. It is regarded as the most results confirm that this approach effectively
effective way to conceal secret information within a enhances the perceptibility of the stego image by
cover image. This technique involves substituting increasing the capacity for embedding additional
the least significant bits of the carrier image pixels secret information.
with the bits of the secret information. The payload In [62], a novel method for concealing and retrieving
capacity of the LSB method increases when multiple a secret image was proposed. This method consists
LSBs are used for message embedding, although this of two phases: encryption and decryption. The
can significantly alter the carrier image. While LSB encryption phase involves hiding the hidden RGB
techniques are straightforward to implement, they color image within a cover image and generating
are vulnerable to various statistical attacks such as shares to be transmitted to the receiver. The
image processing activities and Chi-Square analysis, decryption phase aims to restore the hidden image to
as discussed in [54–55]. its original quality as much as possible from the
In [56], the authors presented a Quantum image received shares.
Least Significant Qubit (LSQ) information hiding In [63], the authors introduced a new procedure for
algorithm. This algorithm enhances message embedding secret information within the green or
security by embedding the message within the blue channel of a carrier image based on secret key
frequency spectrum of an image, utilizing the novel bits and the red channel LSB. This method adds an
Enhanced Quantum Representation (NEQR). extra layer of security to the traditional LSB
In [56], the authors investigated the reliable technique by using a secret key. The process
detection of data concealed within the Least involves deciding whether to replace the LSB of the
Significant Bits (LSB) of normal cut images using green or blue channel based on the red channel LSB
statistical hypothesis testing. According to [57], "the and the secret key bit. The proposed procedure offers
proposed system is applied to conceal the secret improved robustness and security compared to the
message in two bitmap color images with sizes (24 basic LSB method. However, secure key exchange
× 64) and (6 × 165)." The performance of the remains a significant challenge and adds an
proposed hiding system was evaluated through a additional overhead to the proposed technique.
matching process between the Stego JPEG images According to the method proposed by Tahir Ali et al.
produced by the system and the original images. The in paper [64], the message bit is embedded in the
study concluded that the proposed hiding system is LSB of one of the three-color components (RGB) of
efficient, simple, fast, and robust against attacks. a 24-bit color image. This embedding is determined
In [58], Liu et al. introduced a metric for assessing based on the parity of the three LSBs of the R, G,
image quality to aid in the analysis of steganalysis and B components. The technique employs a parity
performance. Additionally, "a scheme for check to hide and recover secret data. In a 24-bit
steganalysis of LSB matching steganography is color image, each RGB component consists of 8 bits,
presented based on feature mining and pattern and the LSBs of these components form a group of
recognition approaches." Compared to other three bits. These bits can generate a sequence with
established steganalysis methods for LSB matching either an even or odd number of 1s. If the three bits
steganography, this method performs optimally. The have an odd number of 1s, odd parity is used; if they
results also suggest that the significance of features have an even number, even parity is employed. The
and the detection performance are influenced not parity and message bits together dictate how the LSB
only by the information-hiding ratio but also by the of each color component is embedded. This method
complexity of the image. allows the concealment of substantial amounts of
Xu et al. [59] proposed a unique methodology for data within a single RGB image with minimal
embedding a series of ternary secret information alteration to the pixel values.
within a plain image using an enhanced LSB Vijaya Bhandari and colleagues [65] discuss an LSB
technique implemented with a modulo-three replacement method for 24-bit color images. This
approach. In a related study, another author [60] technique demonstrates that more data can be hidden
introduced a fast algorithm for identifying the closet in the blue plane compared to the red and green
leader by employing a lookup table method, which planes due to the lower intensity of blue light or
is suitable for matrix embedding steganography objects in human visual perception. This approach is
utilizing both performing code and random linear validated using MATLAB, with results showing that
code. the PSNR (Peak Signal-to-Noise Ratio) value of the
The authors in [61] presented the Pixel Value 24-bit color image is higher. Furthermore, histogram
Modification (PVM) methodology, which enables comparisons indicate that the stego image closely
the insertion of a single secret digit into one segment
438
 Homam EL-TAJ / IJCESEN 10-3(2024)434-460

resembles the original cover image, more so than an The study's findings indicate that this method is
8-bit color image. robust and secure, as the data is embedded in the blue
Turning to Transform Domain approaches, Ahmed component DCT coefficients, which are less
ElSayed et al. [66] demonstrate a method for a noticeable to the human eye.
highly secure data concealment system using a low For further enhancement of steganography
frequency curvelet transform under a cover image. algorithms, Dicky Nofriansyah et al. [69] discussed
This approach employs only the low-frequency a novel image encryption technique combining Hill
component of the curvelet transform to enhance cipher, Morse code, and LSB steganography. The
security, utilizing four secret keys: two shuffle keys, paper explains how these methods collectively
an encryption key, and a key for data concealment. create a secure image encryption system. Similarly,
The low-frequency component of the curvelet Dr. R. Sridevi et al. [70] described a technique that
transform offers several advantages over other integrates image steganography and cryptography
steganography techniques, including reduced using the Least Significant Bit (LSB) technique and
processing time. The method effectively handles the Advanced Encryption Standard (AES)
curve discontinuities without impacting edges, algorithm. In this method, the LSB technique hides
thereby improving the quality of the stego object. a secret message within an image, and AES encrypts
The results indicate that while the Wavelet transform the resulting "stego" image. The encrypted stego
case shows no noticeable differences between the image can then be decrypted to reveal the original
stego and cover images, the Curvelet transform case image and extract the hidden message. The authors
exhibits significant variances. conclude that this technique is effective for secure
In a related study, V. Senthooran et al. [67] propose communication and provides robust security. They
a novel data concealment strategy based on the also suggest that future research could explore
values of a modified quantization table and Discrete combining image encryption and data hiding with
Cosine Transform (DCT) coefficients. The lossy compression.
embedding strength of each coefficient is Baothman, Fatmah Abdulrahman, and Edhah's
determined by comparing the appropriate research into LSB steganography has revealed its
quantization table values with the DCT coefficients applicability in various covert communication
in the correct sequence, as per the mathematical scenarios, including digital watermarking, copyright
formula. The hidden bits are then stored in the protection, and clandestine information exchange.
frequency components of the quantized DCT Law enforcement agencies employ LSB steganalysis
coefficients using the LSB approach. This techniques to detect hidden messages in digital
embedding method is divided into three stages. evidence, while intelligence agencies utilize LSB
Initially, the cover image is segmented into non- steganography for covert communication in
overlapping 8x8 pixel blocks. In the second stage, espionage operations [11, 20, 21, 25].
the DCT coefficient values in each block are However, LSB steganography has limitations such
compared to the corresponding quantization table as low payload capacity, susceptibility to statistical
entries. Adjustments to the conventional attacks, and potential degradation of cover object
quantization table in the middle portion of the quality. The capacity constraints imposed by LSB
segment enhance embedding capacity and maintain embedding techniques limit the secret data
acceptable image quality. By modifying the concealed within a cover object, potentially
quantization table and employing interpolation restricting its applicability in scenarios requiring
techniques, the image size can be expanded to 32x32 high data throughput. Additionally, advancements in
or 16x16 pixel blocks in future implementations, steganalysis algorithms and techniques continue to
with the stego image size evaluated in each instance. pose challenges to the resilience of LSB
Jyoti Gaba and colleagues [68] proposed a steganography against detection [20, 21, 24].
technology called compress encrypt stego (CES) for
secure information sharing. This method involves 2.3 Existing Integration Approaches
pre-processing data before embedding it into a cover
image. The pre-processing step uses a compression A few studies have explored the integration of ECC
factor to reduce the data size, which is then adjusted with LSB steganography, presenting an opportunity
using a key. Compression minimizes the data for innovation in secure communication paradigms.
volume, allowing more information to be embedded By integrating ECC's robust encryption with LSB
in the cover image. Since the data is altered through steganography's covert data embedding capabilities,
compression, attackers cannot retrieve the original the proposed fusion aims to address the limitations
information without the key. Additionally, the secret of existing approaches while offering enhanced
data is hidden within the DCT coefficient values security and stealth in hidden communication
rather than the entire data set, enhancing security. channels [24, 25].
439
 Homam EL-TAJ / IJCESEN 10-3(2024)434-460

The integration of ECC with LSB Steganography is and identification system by leveraging
a promising approach for achieving both encryption biometrics and steganography [26].
and data hiding. Recent studies have demonstrated ● Another method proposed to ensure secure
the effectiveness of using ECC to encrypt biometric message transmission is merging Advanced
data before embedding it in cover images via LSB Encryption Standard (AES) encryption with
Steganography, particularly in secure authentication LSB substitution steganography. AES is a
systems [71-74]. Additionally, integrating ECC with well-established symmetric encryption
LSB in IoT environments, such as secure RFID technique recognized for its strong security
mutual authentication in healthcare, has shown features. By coupling AES with LSB
reduced computational and communication costs substitution steganography, which entails
while maintaining high security standards [75]. altering the least significant bit of pixel
Some existing integration approaches are: values, the method aims to deliver a
● One integration approach involves comprehensive solution for secure
combining steganography and cryptography communication. This integration provides a
with Generative Adversarial Networks two-tiered security approach, where
(GANs). It employs deep learning to create encryption safeguards the message content,
believable cover images for concealing and steganography hides the presence of the
encrypted data. Initially, data undergoes communication. The research illustrates a
AES encryption, then it's hidden within the practical application of combining
least significant bits of the cover image encryption and steganography to bolster the
using LSB steganography [4, 15, 26]. The overall security of message transmission [7,
cover image is crafted by a GAN trained on 12, 17].
a substantial set of natural images. This
method seeks to enhance the steganographic 2.4 Significance of the Proposed Fusion
image's security and fidelity.
● Another existing integration approach The literature review emphasizes integrating
includes utilizing a multi-level advanced cryptographic and steganographic
steganography method, combining two techniques to address data security and privacy
encryption algorithms, AES and Blow-Fish, challenges. The proposed fusion of ECC and LSB
to safeguard the cover image while Steganography aligns with this trend, offering a
embedding encryption keys within the comprehensive approach to hidden communication
steganographic image. Initially, data is that leverages the strengths of both algorithms. By
encrypted via AES and embedded into the integrating ECC's robust encryption with LSB
least significant bits of the cover image steganography's covert data embedding capabilities,
using LSB steganography. Encryption keys the proposed fusion aims to address the limitations
undergo a similar process using Blow-Fish of existing approaches while offering enhanced
encryption before being embedded into security and stealth in hidden communication
another image through LSB steganography. channels [11, 18, 29].
A pixel randomization function hides the
key image within the steganographic image. 3. Methodology
This approach endeavors to heighten the
security and intricacy of the encryption and The methodology for integrating Elliptic Curve
concealment procedures [5, 23, 26]. Encryption (ECE) with Least Significant Bit (LSB)
● Integrating Biometrics and Steganography Steganography involves a systematic approach to
is another existing integration of ECC with ensure the seamless fusion of these two techniques
LSB Steganography. Combining biometrics for hidden communication. The process
and steganography involves concealing encompasses key generation, ECE encryption, and
biometric features like fingerprints, iris embedding the encrypted message within cover
patterns, or facial characteristics within media using LSB steganography. The Flowchart
another medium [6]. The biometric data is method of integrating ECE encryption with LSB
initially encrypted using appropriate RSA, Steganography in Figure 2 below illustrates each
ECC, or AES algorithms. Then, it's inserted step of this intricate process, ensuring the
into the cover medium using steganography confidentiality and integrity of hidden
methods like LSB, DCT, or DWT. The communication.
cover medium may vary from images to
audio or video files [6, 18]. This method
strives to establish a robust authentication
440
 Homam EL-TAJ / IJCESEN 10-3(2024)434-460

3.1 Integration of ECC and LSB encrypted message can be embedded with the use of
Steganography LSB steganography by increasing the number of
masking bits in an image pixel for data to be hidden.
There are few computational steps in the The operation ensures secure and covert
integration of ECC with LSB Steganography. The communications with minimal computation
process of elliptic curve key generation is efficient, overhead, making it feasible for practical
given the small size of the keys in ECC. Further, it applications.
offers a high level of security using relatively less The integration of ECC and LSB Steganography
computation power in comparison to conventional involves a multi-step process to ensure data's secure
methods. In fact, elliptic curve point multiplication and covert communication. The following steps
mainly constitutes the hardness of ECC encryption outline the methodology for integrating the two
and decryption. It can be considered computationally techniques [20, 12, 14, 29]:
intensive, although quite feasible nowadays on
modern processors. This makes it easy: An

Figure 2. Flowchart method of integrating ECE encryption with LSB Steganography.

3.1.1 Generation of keys in Elliptic Curve Cryptography (ECC) as it

establishes the foundation for ensuring the secure
Generating elliptic curve keys requires the choice of transmission of data through encryption and
suitable curves, and polynomial-time complexity decryption.
operations of the order O(n) in the size of the key. a) Generate elliptic curve key.
The process is efficient because of the small size of Here are the steps involved in generating an elliptic
the key in ECC. Security keys may be generated fast curve key pair:
even on slow devices, which increases security in ● Selection of Elliptic Curve: An appropriate
many practical applications. elliptic curve is chosen for cryptographic
According to Budianto, Christofer [8], creating an operations. The selection is crucial as
Elliptic Curve key pair is essential for integrating different curves offer varying levels of
different processes. Studies have illustrated the security and efficiency.
efficacy of ECC in bolstering security across diverse ● Random Private Key Generation: A
fields, such as medical records and image private key is generated randomly. This
steganography [22, 27, 10]. Generating this key pair private key is critical in the encryption and
is crucial for securing communication channels decryption processes and must be kept
facilitating message encryption and decryption confidential to maintain security.
through the ECC algorithm [29]. ● Public Key Derivation: The corresponding
To begin, the process is initiated by creating an public key is derived from the generated
Elliptic Curve key pair consisting of a private key private key and the selected elliptic curve.
and a matching public key [31]. This step is crucial This process ensures the public key is
441
 Homam EL-TAJ / IJCESEN 10-3(2024)434-460

mathematically linked to its associated curves follow a form known as the Weierstrass
private key, enabling secure communication equation (1):
and verification. 𝑦2+𝑎𝑥𝑦+𝑏𝑦=𝑥3+𝑐𝑥2+𝑑𝑥+𝑒 (1)
● Key Pair Output: Both the private and
public keys are returned as a pair. While the where 𝑎,𝑏,𝑐,𝑑, and 𝑒 are real numbers. For
private key remains secret and is used for cryptographic purposes, a simplified form of this
decryption, the public key can be shared equation (2) is used:
openly for encryption and verification 𝑦2=𝑥3+𝑎𝑥+𝑏 (2)
purposes.
In equation (2), the coefficients 𝑎 and 𝑏, as well as
3.1.2 Elliptic curve encryption the variables 𝑥x and 𝑦y, are elements of the field of
real numbers. Figure 2 illustrates examples of
ECC relies on quite complex mathematical elliptic curves generated from Equation 2 with
operations in point addition and doubling, which are different parameters 𝑎 and 𝑏.
highly computational. In this regard, ECC is very Elliptic curves can be classified as singular or non-
efficient for encryption and decryption. The main singular. Figure 3 displays an example of a non-
difficulty, of course, arises from the discrete singular elliptic curve, which is characterized by its
logarithm problem, making the system really secure. smoothness. A smooth curve satisfies the
Real-world implications include strong encryption discriminant condition of a polynomial
that continues to be robust against currently feasible 3
𝑓(𝑥)=𝑥 +𝑎𝑥+𝑏:
computational attacks, and therefore ECC is suitable 4𝑎3+27𝑏2≠0 (3)
for the protection of sensitive information in real-
time applications.
Number theory and algebra are fundamental to
cryptography, serving as the backbone for
cryptographic algorithms to resist various forms of
attack [71]. Unlike other public-key cryptography
algorithms, the principles behind Elliptic Curve
Cryptography (ECC) are notably distinct, presenting
unique comprehension challenges. ECC is grounded
in the mathematical properties of elliptic curves and
the operations involving points on these curves. This
section elucidates the concepts of ECC as applied in
cryptography by first exploring the properties and
operations of elliptic curves over real numbers,
Figure 3: Examples of elliptic curves
leveraging geometric visualization to illustrate key
points. The elliptic curve described in Equation 2 is a cubic
Consider a finite set 𝐸 comprising points on the polynomial, meaning it has three distinct roots,
plane (𝑥𝑖,𝑦𝑖) derived from an elliptic curve equation. denoted as 𝑟1,𝑟2, and 𝑟3. The discriminant is given by
Within this set 𝐸, we define a group addition the equation (4):
operator, denoted by +, which operates on two points 𝐷3=∏𝑖<𝑗(𝑟𝑖−𝑟𝑗)2 (4)
𝑃 and 𝑄 to yield a third point 𝑅∈𝐸 such that 𝑃+𝑄=𝑅.
Given a point 𝐺∈𝐸, the task is to compute the sum If the discriminant is zero, it indicates that two or
𝐺+𝐺+𝐺+…+𝐺 using this group operator. more roots coincide, making the curve non-smooth
Specifically, for any arbitrary integer 𝑘, the notation [72]. Singular curves are not suitable for
𝑘×𝐺 denotes the repeated addition of point 𝐺 to itself cryptographic applications because they are more
𝑘 times (with the + operator applied 𝑘−1 times). vulnerable to attacks. Consequently, we focus
The core concept of ECC lies in the difficulty of exclusively on non-singular curves, ensuring that
deducing 𝑘 from the product 𝑘×𝐺. This requires an curves used in ECC algorithms have a non-zero
attacker to test all possible combinations of repeated discriminant.
additions: 𝐺+𝐺,𝐺+𝐺+𝐺,𝐺+𝐺+𝐺+…+𝐺 [72]. This Elliptic Curve Cryptography (ECC) leverages the
problem is known as the discrete logarithm problem mathematical properties of elliptic curves to create
and is the bedrock of ECC's security. secure and efficient cryptographic systems.
Elliptic curves, despite their name, have no direct Below is a detailed explanation of the mathematical
connection with ellipses [72]. They are defined by formulation behind ECC, including the algorithms
cubic equations, which are also used to calculate the
circumference of an ellipse [73]. Typically, these
442
 Homam EL-TAJ / IJCESEN 10-3(2024)434-460

and steps used in the encryption and decryption difficulty of the elliptic curve discrete logarithm
processes. problem underpin the security of this cryptographic
scheme.
Key Generation C. Mathematical Explanation and Algorithms
Key Pair Generation: Elliptic Curve Point Addition:
Private Key: Select a random integer 𝑑 from the To add two points 𝑃=(𝑥1,𝑦1) and 𝑄=(𝑥2,𝑦2) on an
interval [1,𝑛−1], where 𝑛n is the order of the elliptic elliptic curve, the resulting point (𝑥3,𝑦3)=𝑃+𝑄 is
curve (the number of points on the curve). calculated as following equations (7 & 8):
Public Key: Compute the public key 𝑄=𝑑⋅𝐺, where 𝑦 −𝑦 2
𝑥3 = (𝑥2 −𝑥1 ) − 𝑥1 − 𝑥2 (7)
𝐺 is a predefined point on the curve known as the 2 1
generator point, and d is the private key [76-78].
𝑦 −𝑦
𝑦3 = (𝑥2 −𝑥1 ) (𝑥1 − 𝑥3 ) − 𝑦1 (8)
2 1
A. Encryption
This formula involves finding the slope of the line
Elliptic Curve ElGamal Encryption:
that intersects points 𝑃 and 𝑄, squaring this slope to
Plaintext Representation: Begin by representing
find the new 𝑥-coordinate, and then using this value
the plaintext message 𝑀 as a point 𝑃𝑚 on the elliptic
curve. This step involves mapping the message to a to find the new 𝑦-coordinate. Point addition is
specific point on the curve that can be used in the essential for combining different points on the
encryption process. elliptic curve to produce a new point, which is
crucial for elliptic curve cryptographic operations.
Random Integer: Choose a random integer 𝑘k from
the interval [1,𝑛−1], where 𝑛n is the order of the
Elliptic Curve Point Doubling:
elliptic curve group. This random integer 𝑘k is used
To double a point 𝑃=(𝑥1,𝑦1) on an elliptic curve, the
to ensure the security and randomness of the
resulting point (𝑥3,𝑦3)=2𝑃 is calculated as follows in
encryption process.
(9 & 10):
Ciphertext Calculation: Compute two points: 2
3𝑥1 2 +𝑎
𝐶1=𝑘⋅𝐺C1=k⋅G (5) 𝑥3 = ( ) − 2𝑥1 (9)
2𝑦1
𝐶2=𝑃𝑚+𝑘⋅𝑄 (6) 3𝑥1 2 +𝑎
𝑦3 = ( 2𝑦 ) (𝑥1 − 𝑥3 ) − 𝑦1 (10
1
Here, G is the base point on the elliptic curve, and 𝑄
is the public key. These computations generate two
points that together form the ciphertext. Point doubling is similar to point addition but applies
Ciphertext: The ciphertext is the pair (𝐶1,𝐶2). This when the two points are identical. The formulas
pair of points is what gets transmitted to the calculate the slope of the tangent to the elliptic curve
recipient. at the point 𝑃, which is then used to find the new 𝑥-
and 𝑦- coordinates. This operation is fundamental
B. Decryption for elliptic curve cryptographic algorithms,
particularly for calculating multiples of a point,
Elliptic Curve ElGamal Decryption: which is a key part of elliptic curve multiplication.
Received Ciphertext: When the ciphertext (𝐶1,𝐶2) These operations form the foundation of Elliptic
is received, the recipient uses their private key to Curve Cryptography (ECC). Point addition and
decrypt it. point doubling enable the secure creation and
1. Calculate Shared Secret: Compute 𝑆=𝑑⋅𝐶1 manipulation of cryptographic keys. Through these
, where 𝑑d is the private key. This shared operations, ECC provides robust encryption and
secret 𝑆 is an intermediate value used to decryption methods, ensuring secure data
retrieve the original plaintext point. transmission and storage in various applications.
2. Recover Plaintext: Recover the plaintext READ SECRET MESSAGE (SecMess)
point 𝑃𝑚 by computing 𝑃𝑚=𝐶2−𝑆. This Here are the steps for reading a secret message in
step effectively reverses the encryption the context of elliptic curve encryption:
process, subtracting the shared secret from ● Input Acquisition: The system prompts the
𝐶2 to retrieve the original message point 𝑃𝑚. user to provide the secret message, typically
By following these steps, the Elliptic Curve ElGamal through a text box or similar interface.
Encryption and Decryption process ensures that ● User Input Retrieval: The user's input,
messages can be securely encrypted and decrypted representing the secret message, is obtained
using the properties of elliptic curves. The and stored in a variable.
randomness introduced by the integer 𝑘k and the

443
 Homam EL-TAJ / IJCESEN 10-3(2024)434-460

● Return of Secret Message: The function purposes, such as steganography, image

returns the secret message for further processing, or display.
processing within the encryption process.
SHOW IMAGE IN BINARY
ENCRYPT THE SECRET MESSAGE TO BE
(EncSecMess) BY THE PUBLIC KEY OF The Least Significant Bit (LSB) is the lowest bit in a
RECEIVER binary number sequence. For example, in the binary
● Input Parameters: The function takes two number 10110001, the LSB is the rightmost 1. LSB-
parameters: the secret message to be based steganography is a method used to embed
encrypted (secretMessage) and the public secret data into the least significant bits of pixel
key of the receiver (receiverPublicKey). values within a cover image. For instance, the
● Encryption Process: The secret message is number 240 can be concealed in the first eight bytes
encrypted using the receiver's public key. of three pixels in a 24-bit image.
This process involves utilizing a Consider the following pixel values:
cryptographic algorithm and functions the  Pixel 1: (00100111 11101001 11001000)
chosen cryptographic library provides.  Pixel 2: (00100111 11001000 11101001)
● Public Key Encryption: The  Pixel 3: (11001000 00100111 11101001)
PublicKeyEncrypt function uses the
secretMessage and receiverPublicKey as The binary representation of 240 is 01111000. When
arguments. This function performs the embedding 240 into the first eight bytes of the pixel
actual encryption using the provided public grid, the result is:
key.  Pixel 1: (00100110 11101001 11001001)
● Encrypted Message Output: The  Pixel 2: (00100111 11001001 11101000)
encrypted message (encryptedMessage) is  Pixel 3: (11001000 00100110 11101000)
obtained as the output of the encryption
process. In this example, the number 240 is embedded into
● Return Encrypted Message: The function the first eight bytes of the grid, resulting in only 6
returns the encrypted message bits being altered. This demonstrates the efficiency
(encryptedMessage) for further processing and subtlety of LSB-based steganography in hiding
or transmission. data within an image.
Introduced by Mielikainen in 2006, the LSB
D. Steganography matching revisited technique represents a significant
advancement in steganography, allowing for the
READ IMAGE. embedding of messages within monochrome images
Here are the steps for reading an image from a given [10]. This method discreetly alters pixel pair values
path and loading it: to modify the LSB, ensuring hidden messages are
● Input Path: The function takes a parameter effectively embedded without compromising the
imagePath, which represents the image's file visual integrity of the image. Its primary strength lies
path to be read. in its ability to conceal information while preserving
● Image Loading: The function loads the the original appearance of the image, thereby
image specified by the imagePath from the minimizing the risk of detection.
file system. This involves accessing the file The technique encodes covert messages by adjusting
at the provided path and loading its contents the LSBs of pixel values, minimizing perceptual
into memory. differences between the original and modified
● Image Representation: The loaded image images. This precise alteration of pixels is crucial for
is typically a data structure compatible with hiding the embedded message while maintaining the
the chosen programming environment. This image's authenticity. Message bits 𝑚𝑖mi and 𝑚𝑖+1
representation could be a matrix of pixel are embedded using consecutive pixel values 𝑥𝑖 and
values for grayscale images or a set of 𝑥𝑖+1, with the steganographic image representing
matrices for color images, depending on the these pixels as 𝑦𝑖 and 𝑦𝑖+1. Converting 𝑥𝑖 to 𝑦𝑖
image format and the programming involves encoding the message bit into the LSB,
language's image processing capabilities. allowing it to seamlessly integrate with the pixel's
● Return Loaded Image: The function binary structure.
returns the loaded image, allowing for Central to this technique is the binary function 𝑓,
further processing or manipulation. This outlined in Equation (11), which determines the
returned image data can be used for various necessity of pixel value adjustments to accurately
embed the message:
444
 Homam EL-TAJ / IJCESEN 10-3(2024)434-460

𝑦𝑖+1 is adjusted based on 𝑓(𝑥𝑖−1, 𝑥𝑖+1) in relation to

𝑥 𝑚𝑖+1.
𝑓(𝑥𝑖 , 𝑥𝑖 + 1) = 𝐿𝑆𝐵 ( 2𝑖 + 𝑥𝑖 + 1) (11)
Mielikainen’s approach is further validated by its
reduction in the average number of pixel adjustments
This function 𝑓 guides the embedding process, needed for embedding messages. Experiments
indicating how pixel pairs (𝑥𝑖, xi) and (𝑥𝑖+1, xi+1) are indicate that this method averages just 0.375 changes
modified to encode the message bits (𝑚𝑖, mi) and per pixel, a notable improvement from the previous
𝑚𝑖+1. If the LSB of (𝑥𝑖, xi) aligns with (𝑚𝑖, mi), no standard of 0.500 changes with traditional LSB
modification is made, and 𝑦𝑖 remains as (𝑥𝑖, xi). techniques. This advancement not only demonstrates
Conversely, if 𝑓 necessitates an adjustment, 𝑥𝑖 is the method’s high efficacy but also its alignment
incremented or decremented by one to correctly with theoretical models. Figure 4 elucidates the
capture the message bit, thus changing 𝑦𝑖. Similarly, details of this LSB matching process.

Figure 4: Schematic Diagram of LSB Matching Revisited Method

Here is the schematic diagram of the LSB matching Select Cover Image: Choose a cover image 𝐼 with
method. This diagram outlines the decision process dimensions 𝑊×𝐻. This image will serve as the
for embedding message bits 𝑚𝑖 and 𝑚𝑖+1 into pixel medium for embedding the secret message.
values 𝑥𝑖 and 𝑥𝑖+1: Determine Embedding Capacity: Calculate the
1. Start with LSB(𝑥𝑖)=𝑚𝑖. number of bits available for embedding in the cover
2. Check if 𝑓(𝑥𝑖,𝑥𝑖+1)≠𝑚𝑖+1: image using equation 12:
 If Yes (Y): Adjust 𝑦𝑖+1=𝑥𝑖+1±1. Capacity=𝑊×𝐻×3×8 (12)
 If No (N): Set 𝑦𝑖=𝑥𝑖.
3. Otherwise, check if 𝑓(𝑥𝑖−1,𝑥𝑖+1)=𝑚𝑖+1: Here, 𝑊 and 𝐻 are the width and height of the image,
 If Yes (Y): Adjust 𝑦𝑖=𝑥𝑖−1. 3 represents the RGB color channels, and 8 is the
 If No (N): Set 𝑦𝑖+1=𝑥𝑖+1. number of bits per pixel channel. This formula
LSB steganography operates by modifying the least provides the total number of bits available for
significant bits (LSBs) of pixel values in a cover embedding the binary message.
image to embed a secret message. Because Ensure Message Fits: Ensure that the length of the
alterations in the LSB cause minimal changes to the binary message ∣𝐵𝑀∣ does not exceed the
pixel value, these modifications are generally embedding capacity calculated in the previous step.
imperceptible to the human eye, ensuring that the This check is crucial to ensure that the entire
embedded message remains hidden without message can be embedded within the cover image
noticeably affecting the image's appearance. without exceeding its capacity.
Embedding Process Embed Message: For each bit 𝑏𝑖bi in 𝐵𝑀: Select a
Convert Message to Binary: First, convert the pixel 𝑃 in the cover image. Embed 𝑏𝑖 in the least
secret message 𝑀 into its binary representation 𝐵𝑀. significant bit of the selected pixel's channel (R, G,
This step transforms the message into a sequence of or B). Update the pixel value with the new least
binary digits that can be embedded into the image.
445
 Homam EL-TAJ / IJCESEN 10-3(2024)434-460

significant bit, effectively embedding the message maintaining the cover image's visual fidelity while
bit into the image. securely hiding the secret message.
By following these steps, LSB steganography Extraction Process
ensures that the secret message is discreetly Initialize: Begin by initializing an empty binary
embedded into the cover image. The minimal string 𝐵𝑀′. This string will store the extracted bits
changes to the pixel values preserve the visual from the stego image.
integrity of the image, making the presence of the Extract Bits: For each pixel 𝑃 in the stego image 𝐼′:
hidden message undetectable to casual observers. Extract the least significant bit from each color
This method provides a straightforward yet effective channel (R, G, and B) and append these bits to 𝐵𝑀′.
approach to secure message embedding in digital This step involves iterating through the pixels and
images. channels of the stego image to gather all embedded
Mathematical Formulation: bits.
In LSB steganography, the mathematical Reconstruct Message: Convert the binary string
formulation for embedding a secret message into a 𝐵𝑀′ back to its original text representation. This
cover image involves the following steps and conversion translates the sequence of binary digits
notations: into the readable form of the secret message that was
Let I(𝑥,𝑦,𝑐) represent the pixel value at position initially embedded.
(𝑥,𝑦) for color channel 𝑐 (where 𝑐 can be R, G, or B). Mathematical Formulation:
Let 𝑏𝑖 denote the 𝑖th bit of the binary message 𝐵𝑀. Let 𝐼′(𝑥,𝑦,𝑐) represent the pixel value at position
The modified pixel value I′(𝑥,𝑦,𝑐) after embedding (𝑥,𝑦) for color channel 𝑐 in the stego image.
the bit 𝑏𝑖 is given by: The least significant bit 𝑏𝑖 is extracted as follows:
𝐼′(𝑥,𝑦,𝑐)=(𝐼(𝑥,𝑦,𝑐)&∼1) ∥ 𝑏𝑖 (13) 𝑏𝑖=𝐼′(𝑥,𝑦,𝑐)&1 (14)

Here, the operations are defined as follows: Here, & is the bitwise AND operator, which isolates
 & is the bitwise AND operator, which is the least significant bit of the pixel value.
used to clear the least significant bit of the Extracting the Least Significant Bit:
pixel value. The expression 𝐼′(𝑥,𝑦,𝑐)&1 isolates the least
 ∼ is the bitwise NOT operator, which significant bit of the pixel value 𝐼′(𝑥,𝑦,𝑐). The
inverts all the bits of the operand, effectively bitwise AND operation with 1 retains only the least
turning the least significant bit to 0. significant bit, discarding all other bits.
∥ is the bitwise OR operator, which sets the least Appending Extracted Bits:
significant bit to 𝑏𝑖. The extracted bits from the red, green, and blue
Clearing the Least Significant Bit: channels of each pixel are appended sequentially to
The expression 𝐼(𝑥,𝑦,𝑐)&∼1 clears the least the binary string 𝐵𝑀′. This step ensures that all bits
significant bit of the pixel value 𝐼(𝑥,𝑦,𝑐). This is embedded during the embedding process are
done by performing a bitwise AND operation correctly retrieved.
between the pixel value and the bitwise NOT of 1. Reconstructing the Message:
The bitwise NOT of 1 is a number where all bits are Once all bits are extracted and appended to 𝐵𝑀′, the
1 except the least significant bit, which is 0. The binary string is converted back to its original text
AND operation with this number clears the least representation. This conversion reverses the initial
significant bit of the pixel value. process of converting the message to binary, thereby
Embedding the Message Bit: revealing the hidden message in its original form.
The expression (𝐼(𝑥,𝑦,𝑐)&∼1)∥𝑏𝑖 sets the least By following these steps, the LSB steganography
significant bit of the cleared pixel value to the extraction process effectively retrieves the hidden
message bit 𝑏𝑖. The bitwise OR operation ensures message from the stego image. The mathematical
that the least significant bit of the modified pixel formulation provides a clear and precise method for
value becomes 𝑏𝑖, while the other bits remain isolating and gathering the embedded bits, ensuring
unchanged. that the secret message can be accurately
By applying this formula, each bit of the binary reconstructed without loss or distortion. This
message 𝐵𝑀 is embedded into the least significant extraction process underscores the efficiency and
bit of the corresponding pixel values in the cover reliability of LSB steganography for secure message
image. This process is repeated for all bits of the embedding and retrieval.
message, ensuring that the entire message is Here are the steps for displaying the binary
concealed within the image with minimal perceptual representation of an image:
changes. This mathematical approach guarantees ● Image Input: The function takes an image
that the embedding is precise and efficient, (image) as input, which is the image to be
displayed in binary representation.
446
 Homam EL-TAJ / IJCESEN 10-3(2024)434-460

● Conversion to Binary: The function ●Comparison: Compare the size of the

converts the input image into its binary encrypted message with the capacity of the
representation. This process involves image.
transforming the pixel values of the image ● Check Fit: If the size of the encrypted
into binary form. Each pixel's color message exceeds the image capacity, return
components (e.g., red, green, and blue for False indicating that the message does not fit
RGB images) are converted into binary into the image.
digits. ● Return Result: If the size of the encrypted
● Binary Image Representation: The message is within the image capacity, return
resulting binary image (binaryImage) True indicating that the message fits into the
represents the input image's pixel values in image.
binary format. A sequence of binary digits LOCATE THE BINARY PIXELS THAT WILL BE
represents each pixel's color information. USED FOR HIDING THE SECRET MESSAGE
● Display Binary Image: The binary image is
displayed using a suitable output Here are the steps for locating the binary pixels that
mechanism. In the provided code, it is will be used for hiding the secret message:
printed to the console or output stream. This ● Find Suitable Pixels: Determine the pixels
allows users to visualize the binary in the binary image where the message will
representation of the input image. be hidden. This involves identifying
READ ENCSECMESS appropriate locations within the image to
Here are the steps for reading the encrypted secret accommodate the message without
message: significantly altering the image's
● Message Retrieval: The function appearance.
ReadEncryptedSecretMessage() retrieves ● Search Process: Search within the binary
the encrypted secret message, assumed to be image to identify suitable pixels for message
stored or inputted somehow. This could hiding. This process may involve analyzing
involve accessing a file, receiving input pixel values, patterns, or other
from a user, or obtaining data from another characteristics to select appropriate
source. locations.
● Encrypted Message Acquisition: The ● Pixel Selection: Identify and select pixels
encrypted secret message (encSecMess) is that meet the criteria for hiding the message.
obtained from the source where it is stored These pixels should provide sufficient
or inputted. This message represents the capacity to embed the message while
confidential information that has been minimizing the risk of detection.
encrypted using cryptographic techniques. ● Return Result: Return the located pixels
● Return Encrypted Message: The function suitable for hiding the message. These pixels
returns the encrypted secret message will be used in the subsequent steps of the
(encSecMess) for further processing or steganography process.
decryption. This allows the encrypted
message to be passed to other functions or HIDE THE MESSAGE
modules to decrypt and extract the original
content. Here are the steps for hiding the message within the
CHECK MESSAGE FITTING INTO THE IMAGE image:
● Embed Message: Modify the binary image
Here are the steps for checking if the encrypted by embedding the message in the located
message fits into the image: pixels. This involves altering the selected
● Calculate Image Capacity: Determine the pixels to encode the message while
capacity of the image to hold data. This minimizing any perceptible changes to the
involves assessing the available space image.
within the binary image for embedding ● Message Embedding Process: Utilize a
additional information. suitable algorithm or technique to embed the
● Calculate Message Size: Determine the message within the image. This process
size of the encrypted message. This involves ensures that the message is hidden
calculating the number of bits or bytes effectively while maintaining the image's
required to represent the encrypted message. visual integrity.
● Modification of Pixels: Modify the selected
pixels in the binary image to incorporate the
447
 Homam EL-TAJ / IJCESEN 10-3(2024)434-460

message. This modification typically ● Hide Message in Image: Embed the

involves adjusting pixel values or attributes encrypted secret message into the image
to encode the message data. using the located pixels.
● Return Modified Image: Return the ● Show Modified Image: Display the
modified image with the embedded modified image with the hidden message.
message. This image will contain a
concealed message and can be used for 3.1.4 Transmission and decryption
further processing or transmission.
SHOW IMAGE AFTER HIDING THE MESSAGE According to Varghese, F., Sasikala [9], the
● Display Modified Image: Utilize the steganographically modified cover object,
provided ShowModifiedImage function to containing the encrypted message can then be
display the modified image that contains the transmitted through public communication channels
hidden message. without arousing suspicion [20, 17]. The hidden
● Print Image Information: Print a ciphertext is extracted from the steganographic
descriptive message indicating that the carrier using LSB steganalysis techniques at the
image being displayed is the one after the recipient's end. Bitwise operations are applied to
message has been hidden. extract the hidden ciphertext from the
● Image Display: Use the DisplayImage steganographic carrier. These are computationally
function (not defined in the provided code simple and efficient in both encryption and
snippet) to visualize the modified image decryption operations. More importantly, the
containing the concealed message. decrypted extracted ciphertext by ECC requires
elliptic curve point multiplication while maintaining
3.1.3 Main workflow of encryption and polynomial time complexity. This indeed infers
steganography practical implications that ensure the decrypted
messages and the secure restoration of their original
The foremost workflow integrates ECC encryption plain text with efficient attainment and no significant
with LSB steganography, which in turn distributes computational stress. Subsequently, the extracted
computational complexity between operations ciphertext is decrypted using the recipient's private
performed on elliptic curves and bitwise embedding. key, recovering the original plaintext message [26].
ECC encryption is polynomial in complexity,
whereas linear time-related to the number of pixels 3.2 Security Considerations
in the image for embedding through LSB. It is a
hybrid approach that distributes computational load This kind of key management also ensures the
in such a manner as to ensure safe and efficient security of the cryptographic keys; it uses robust
message transmission without considerable encryption algorithms that come with computational
performance loss, even on standard computing complexities important to security but can be
resources. handled by modern hardware. Its practical
Here are the steps for the main workflow of application is that secure communication can be
encryption and steganography: preserved even in environments that are likely to
● Get Image Path: Obtain the file path of the suffer from advanced cryptographic attacks, thus
image from the user. utilizing ECC efficiency and LSB steganography
● Read Image: Read the image from the subtlety to effectively protect sensitive information.
specified file path. Various security considerations must be considered
● Display Image in Binary: Convert the throughout the integration process to mitigate
image to its binary representation and potential vulnerabilities and attacks. These include
display it. ensuring the confidentiality and integrity of
● Convert Image to Binary: Convert the cryptographic keys, employing robust encryption
image to its binary representation. algorithms with appropriate key sizes, and
● Read Encrypted Secret Message: Read the implementing steganographic techniques resistant to
encrypted secret message. statistical analysis and detection [26, 29].
● Check Message Fit: Check if the message
fits within the capacity of the image. If not, 4. Integration of Elliptic Curve Encryption
display an error message and terminate the Integrated With LSB Steganography in
process. Stegno Curve Method
● Locate Pixels for Message: Determine
suitable pixels in the binary image for hiding The Stegno Curve method is a combination of ECC
the message. and LSB steganography, where polynomial time
448
 Homam EL-TAJ / IJCESEN 10-3(2024)434-460

complexity for encryption/decryption is taken from Thus, it is a workable solution when encrypting
ECC and linear complexity in data embedding is sensitive data for many applications.
provided by LSB. This also ensures that the Represent Plaintext: Convert the plaintext message
computational requirements are balanced, thereby 𝑀M into a point 𝑃𝑚Pm on the elliptic curve.
ensuring covert secure data transmission, which is fit Choose Random Integer: Select a random integer
for practice in many applications, especially where 𝑘 from the interval [1,𝑛−1].
the computational power is limited. Overall efficient Calculate Ciphertext Points:
integration is suitable for operation in real-time Compute 𝐶1=𝑘⋅𝐺.
secure communication scenarios. Compute 𝐶2=𝑃𝑚+𝑘⋅𝑄.
The integration of Elliptic Curve Encryption (ECC) Ciphertext: The ciphertext is the pair (𝐶1,𝐶2).
with Least Significant Bit (LSB) Steganography in 3. Embedding Encrypted Message using LSB
the Stegno Curve method significantly advances Steganography
secure communication. Stegno Curve, the proposed The ECC binary form is embedded in the least
method, leverages the combined strength of ECC significant bits of a cover image through the LSB
and LSB steganography to ensure the steganography approach. This process is linear in
confidentiality, integrity, and covert transmission of complexity relative to the number of pixels used in
sensitive information. This integration allows for the image. On the other hand, it is computationally
data encryption using ECC and concealing the light; hence, it allows for quick embedding without
encrypted data within digital media using LSB much processing power. What this really means in
steganography. practice is that encrypted messages could be hidden
Compound Method: ECC and LSB Steganography nicely in digital images, allowing for secretive and
Combining Elliptic Curve Cryptography (ECC) with secure data communication.
Least Significant Bit (LSB) Steganography involves Convert Ciphertext to Binary: Convert the
embedding an encrypted message into the least coordinates of 𝐶1 and 𝐶2 to binary form.
significant bits of pixel values in a cover image. Here Select Cover Image: Choose a cover image 𝐼 with
is a step-by-step formulation of the combined dimensions 𝑊×𝐻.
process: Determine Embedding Capacity: Calculate the
1. Key Generation number of bits available for embedding in the cover
The key generation is carried out for ECC using the image:
selected elliptic curve parameters. Because elliptic Capacity=𝑊×𝐻×3×8
curve operations have polynomial time complexity, Where 𝑊 and 𝐻 are the width and height of the
key generations can be computed within reasonable image, 3 represents the RGB channels, and 8 is the
time and with much more computational efficiency number of bits per pixel channel.
to adopt ECC widely in practice for secure Ensure Message Fits: Ensure that the length of the
communication even by machines with less binary message ∣𝐵𝐶1,𝐶2∣ does not exceed the
computational power. embedding capacity.
Elliptic Curve Key Pair Generation: Embed Binary Data: For each bit 𝑏𝑖bi in 𝐵𝐶1,𝐶2:
Select Elliptic Curve Parameters: Choose an Select a pixel 𝑃 in the cover image. Embed 𝑏𝑖 in the
elliptic curve 𝐸 defined by the equation 𝑦2=𝑥3+𝑎𝑥+𝑏 least significant bit of the selected pixel channel (R,
mod 𝑝, where 𝑎 and 𝑏 are constants and 𝑝 is a prime G, or B).Update the pixel value with the new least
number. significant bit.
Ensure the curve parameters satisfy 4𝑎3+27𝑏2≠0 to Mathematical Formulation:
avoid singularities. Let 𝐼(𝑥,𝑦,𝑐) be the pixel value at position (𝑥,𝑦) for
Generate Private Key: Select a random integer 𝑑d color channel 𝑐 (where 𝑐 can be R, G, or B). Let 𝑏𝑖
from the interval [1,𝑛−1], where 𝑛n is the order of be the 𝑖-th bit of the binary message 𝐵𝐶1,𝐶2. The
the curve. modified pixel value 𝐼′(𝑥,𝑦,𝑐) after embedding 𝑏𝑖 is
Generate Public Key: Compute the public key given by:
𝑄=𝑑⋅𝐺, where 𝐺 is the generator point on the elliptic 𝐼′(𝑥,𝑦,𝑐)=(𝐼(𝑥,𝑦,𝑐)&∼1)∥𝑏𝑖
curve. Here, & is the bitwise AND operator, ∼ is the bitwise
2. Encryption with ECC NOT operator, and ∥ is the bitwise OR operator.
ECC consists in converting plaintext to points of 4. Extraction and Decryption
elliptic curves and making point multiplications The encryption data extraction process from the
during encryption. The computational complexity in stego image includes simple bitwise operations.
this is polynomial; while the operations are Hence, this process is computationally light and
intensive, they are feasible for the modern efficient. Hardware requirements allow for ECC
processors, and hence secure encryption can be decryption to be made manageable through curve
effected in a reasonable amount of time with ECC.
449
 Homam EL-TAJ / IJCESEN 10-3(2024)434-460

point multiplications, where original plain text 4.1 Performance and System Specifications
messages are recovered securely and efficiently to
provide a robust secure communication method. Comprehensive performance analysis and system
Extract Encrypted Data from Stego Image: specifications are imperative to evaluate the efficacy
Extract the binary data from the LSBs of the stego and practicality of the integration of Elliptic Curve
image pixels. Encryption with LSB Steganography in the Stegno
Reconstruct Ciphertext: Reconstruct the ciphertext Curve method. The success of any cryptographic and
points 𝐶1 and 𝐶2 from the extracted binary data. steganographic system heavily relies on its ability to
Calculate Shared Secret: Compute 𝑆=𝑑⋅𝐶1, where maintain adequate performance while ensuring
𝑑 is the private key. robust security measures.
Decrypt to Obtain Plaintext: Specification of the high-performance PC for
Recover the plaintext point 𝑃𝑚 by computing evaluation:
𝑃𝑚=𝐶2−𝑆. Convert 𝑃𝑚 back to the original Processor: Intel Core i7-11370H (11th Gen)
plaintext message 𝑀. RAM: 16 GB DDR4
Compound Pseudocode Storage: 512gb NVMe SSD
1. Encryption and Embedding: Graphics Card: NVIDIA GeForce RTX 3050
Initialize: Define ECC parameters and select the Operating System: Windows 11 Pro
curve. Generate ECC key pair (private key 𝑑, public This powerful hardware configuration ensures
key 𝑄). smooth execution of complex cryptographic and
Input: steganographic algorithms to conduct rigorous
 Secret message 𝑀. testing and analysis of the integrated Stegno Curve
 Cover image 𝐼. method.
Encryption:
 Convert 𝑀 to elliptic curve point 𝑃𝑚. 4.2 Performance Analysis
 Choose random integer 𝑘. Various performance tests are conducted to assess
 Compute 𝐶1=𝑘⋅𝐺. the efficiency and speed of the integrated system.
 Compute 𝐶2=𝑃𝑚+𝑘⋅𝑄. These tests included:
 Convert 𝐶1 and 𝐶2 to binary. 1. Encryption and decryption speed
Steganography: Measuring the time taken to encrypt and decrypt
 Embed binary data of 𝐶1 and 𝐶2 into LSBs messages using Elliptic Curve Encryption in Stegno
of 𝐼. Curve. This evaluation helps determine the
 Generate stego image 𝐼𝑠. computational overhead of encryption and
Output: decryption processes.
 Stego image 𝐼𝑠. 2. Steganographic embedding and
2. Extraction and Decryption: extraction speed
Input: Analysing the time required to embed encrypted
 Stego image 𝐼𝑠. data within images using LSB Steganography and
 Private key 𝑑d. extracting the hidden data. This assessment is crucial
Extraction: for understanding the performance impact of
 Extract binary data from LSBs of 𝐼𝑠. steganographic operations.
 Reconstruct 𝐶1 and 𝐶2. 3. Resource utilization
Decryption: Monitoring CPU, RAM, and GPU usage during
 Compute 𝑆=𝑑⋅𝐶1. encryption, decryption, embedding, and extraction.
 Compute 𝑃𝑚=𝐶2−𝑆. This analysis provides insights into the resource
requirements of the integrated system and helps
 Converts 𝑃𝑚 to plaintext message 𝑀.
identify potential bottlenecks.
Output:
4. Scalability
 Original message 𝑀.
The system can scale with increasing message sizes
By combining ECC and LSB steganography, this
and image dimensions. This test helps determine the
method ensures both the security of the message
system's suitability for handling large volumes of
through encryption and the stealthy embedding of
data while maintaining acceptable performance
the encrypted message into a cover image, providing
levels.
a robust approach to secure communication.
Thoroughly evaluating the performance of the
integrated Elliptic Curve Encryption with LSB
Steganography in the Stengno Curve method

450
 Homam EL-TAJ / IJCESEN 10-3(2024)434-460

provides valuable insights into its practicality and The method generates cryptographic keys used in the
efficiency for real-world applications. steganographic process. These keys can be used for
encryption or as part of the steganographic
4.3 Key Generation with Stegno Curve algorithm.
The secret message and the generated keys are
The key generation process within the Stegno Curve embedded into the Image. This process can
method represents a fundamental aspect of manipulate the pixels of the image, using LSB
establishing secure communication channels. Stegno replacement or another steganographic technique.
Curve employs a robust key generation algorithm After the steganographic process, this method
based on Elliptic Curve Cryptography (ECC) to performs a histogram analysis on the resulting
generate secure key pairs. ECC offers several image. A histogram is a graphical representation of
advantages over traditional encryption algorithms, the distribution of pixel Intensities in an image. This
including smaller key sizes and enhanced security analysis could help identify any Irregularities
(Table 2). The key generation algorithm follows introduced by the steganographic process, as
established cryptographic standards to produce changes in pixel values might be visible in the
random and unique key pairs for encryption and histogram.
decryption purposes [30]. Method will also provide a binary representation of
Here’s a practical example: the steganographic Image. This can help users
Stegno Curve generates the sender's public and visually inspect the changes made to the image at the
private keys for encryption as shown in Figure 5. binary level.
Then the receiver’s public and private keys can also The steganography image is the result, containing
be generated as shown in Figure 6. the embedded message and keys. This image appears
unchanged to the human eye but carries hidden
4.4 Encryption Image Steganography information [31]. (Figure 7a & 7b)

The Stegno Curve also allows the encryption of an

image, generating public and private keys with
encrypted messages and binary values. Here's a
breakdown of the steps you might be taking:

Figure 7a: Encryption Steganography

Figure 5: Sender key generation

Figure 6: Receiver keys generation Figure 7b: Encryption Steganography

451
 Homam EL-TAJ / IJCESEN 10-3(2024)434-460

4.5 Decryption Steganography 4.6 Method Testing and Results

Here are the steps for the steganography decryption To validate the functionality and effectiveness of
process: the Stegno Curve method, extensive testing was
Providing the steganography image containing the conducted to assess its performance in real-world
hidden message. scenarios. This section outlines the testing
The decryption process begins with extracting the procedures employed, presents the results obtained,
hidden data from the steganography image. This and compares them with existing solutions in the
involves analyzing the image to locate the embedded field.
message using steganography detection techniques.
Once the hidden data is extracted, the next step is to 4.6.1 Testing methodology
decode it. The decoding process depends on the
encoding method used during the encryption and The Stegno Curve method was executed on the
embedding process. designated high-performance PC with the specified
If encryption was applied to the hidden message hardware configuration. Various test scenarios were
before embedding it, the user may need to provide a devised to evaluate the method's performance,
decryption key. This key is essential for decrypting including encryption and decryption speed,
the encoded message to its original form. steganographic embedding and extraction
If encryption was used, decrypt the extracted efficiency, and resource utilization.
message using the provided decryption key. This
step ensures that the original plaintext message is 4.6.2 Results
retrieved from the encoded and encrypted data.
Once the decryption is complete, display the The testing results indicate the appropriateness of
decrypted message to the user. This is the original the Stegno Curve model to secure data transmission
plaintext message that was hidden within the because it fuses Elliptic Curve Encryption and LSB
steganography image, as shown in Figure 8a & 8b. Steganography. Some of the main testing results are
outlined below:
1. Encryption / Decryption Speed: The
model indicated great speed in the
encryption and decryption of messages
using Elliptic Curve Encryption, increasing
the speed from 2.07 GHz to 3.50 GHz.
2. Steganographic Embedding and
Extraction Speed: The embedded and
extracted secret message or encrypted data
from the images' LSB substitution was quick
even with larger files, as indicated in the
Stegno Curve.
3. High-speed Performance: The method has
shown efficient and effective system
Figure 8a: Decryption Steganography
resource utilization, with minimal usage of
CPU, RAM, and GPU during the
encryption, decryption, embedding, and
extraction of data.
4. Confidentiality and Integrity of
Cryptographic Keys: Application and
removal processes of encryption using
Elliptic Curve Encryption were carried out
in a way that allows only an authorized
recipient to access and read the messages.
5. Strong Encryption Algorithms with Good
Key Sizes: The Stegno Curve method used
the very strong Elliptic Curve Encryption to
withstand cryptanalytic attacks.
The testing results demonstrate the efficacy of the
Figure 8b: Decryption Steganography Stegno Curve method in securing data transmission
through integrating Elliptic Curve Encryption with
452
 Homam EL-TAJ / IJCESEN 10-3(2024)434-460

LSB Steganography. Key findings from the testing resistance to statistical analysis and
include: detection. Despite embedding and
extracting encrypted data within images, the
Table 1: Performance Metrics
Metric Value Before Value After
method maintained high-speed performance
Stegno Curve Stegno and efficiency, ensuring that the hidden
Curve messages remained covert and difficult to
Encryption Speed 2.07 3.50 detect by unauthorized parties.
(GHz) ● Encryption and Decryption Speed: The
Decryption Speed 2.07 3.50 method showcased remarkable speed in
(GHz) encrypting and decrypting messages using
CPU Usage (%) 5 17 Elliptic Curve Encryption, the speed went
RAM Usage (MB) 42.4 42.4 from 2.07 GHz to 3.50GHz. Figure 9 (a &
GPU Usage (%) 1 25 b) is the visual speed demonstration before
Detection Moderate High and after using the Stegno Curve Method.
Resistance ● Steganographic Embedding and
(Statistical Attacks)
Extraction Speed: Stegno Curve exhibited
efficient embedding and extracting
6. Statistical Analysis and Detection encrypted data within images using LSB
Resistance: It provides a solution that
Steganography, maintaining high-speed
resists statistical analysis and detection
performance even with large image files.
through the integration of the LSB ● Resource Utilization: The method
steganography method incorporated in the
demonstrated optimal utilization of system
Stegno Curve method. With this process of resources, with minimal CPU, RAM, and
embedding and extraction of encrypted data
GPU usage during encryption, decryption,
in images, the process maintains its speed
embedding, and extraction processes. The
performance and efficiency high enough to
CPU usage went from 5% to 17%. The
hide the messages conveyed and make them
memory usage by the method was around
difficult for unauthorized parties to access. 42.4 MB, while the GPU usage during the
● Resistant to Statistical Analysis and process went to 25% from 1% as can be seen
Detection: LSB Steganography, integrated
in Figure 10 (a, b, c).
into the Stegno Curve method, exhibited

Table 2: Enhanced Security Table

Method Key Security Encryption Decryption Payload Detection
Length Level Time (ms) Time (ms) Capacity Resistance
(bits)
RSA Encryption 3072 High 150 140 Medium Medium
AES Encryption 256 High 100 90 Medium Medium
ECC Encryption + LSB 256 Very 50 45 High High
Steganography High

Figure 9.a: Encryption Speed

Figure 9.b: Decryption speed
453
 Homam EL-TAJ / IJCESEN 10-3(2024)434-460

encrypted messages, thereby preserving

confidentiality.
● Robust Encryption Algorithms with
Appropriate Key Sizes: The Stegno Curve
method utilized Elliptic Curve Encryption,
known for its robustness against various
cryptographic attacks. The testing
demonstrated the efficacy of this encryption
algorithm in securing data transmission,
with messages encrypted using appropriate
key sizes. This choice of algorithm and key
size contributed to the overall security of the
method.

Figure 10.a: Resource utilized 4.7 Comparison with Other Solutions

Compared with existing cryptographic and

steganographic applications, the Stegno Curve
stands out for its superior performance and robust
security features. The seamless integration of
Elliptic Curve Encryption with LSB Steganography
sets it apart from traditional methods, offering
enhanced data protection while ensuring covert
communication channels.

ECC-LSB vs. RSA/AES: ECC provides equivalent

security to RSA with much shorter key lengths,
making it more efficient. For example, a 256-bit
ECC key is comparable in security to a 3072-bit
RSA key. This efficiency makes ECC particularly
suitable for environments with limited
Figure 10.b: Resource utilized computational resources. AES, a symmetric
encryption method, provides high security but does
not offer the same public key infrastructure
advantages as ECC [76]. Robustness Against
Attacks: ECC is more resistant to quantum attacks
compared to RSA. The addition of LSB
Steganography further enhances security by hiding
the existence of encrypted data, making it harder for
attackers to detect and intercept [23].
Security Strength
The Stegno Curve method, which integrates ECC
and LSB Steganography, offers significant
advantages over traditional cryptographic
integrations such as RSA and AES. ECC provides
equivalent security to RSA with much shorter key
lengths—for example, a 256-bit ECC key offers
comparable security to a 3072-bit RSA key. This
results in more efficient use of computational
Figure 10.c: Resource utilized. resources while maintaining high security levels.
Additionally, ECC is more resistant to quantum
● Confidentiality and Integrity of attacks, particularly against Shor's algorithm,
Cryptographic Keys: The encryption and whereas RSA is vulnerable to such quantum attacks.
decryption processes using Elliptic Curve While AES provides robust symmetric encryption, it
Encryption ensured that only authorized lacks the key exchange capabilities of ECC.
parties could access and decrypt the Integrating LSB Steganography with ECC enhances
security further by concealing the existence of the
454
 Homam EL-TAJ / IJCESEN 10-3(2024)434-460

encrypted message, making it more challenging for computational power and expertise in machine
attackers to detect the communication. learning.
Computational Efficiency Even though ECE and LSB Steganography
When comparing the computational efficiency of the integratedly enhance the security of the system, LSB
Stegno Curve method with AES and Blowfish Steganography is still lagging in terms of payload
integrations, ECC operations demonstrate faster capacity and can be vulnerable to some kind of
encryption and decryption speeds due to shorter key detection. Further research could focus on
lengths and more efficient computations. Hybrid integrating advanced AI and ML techniques towards
approaches like AES-Blowfish are efficient for optimum adjustments to the embedding process for
encrypting bulk data but add complexity in key more secure communication. As an example,
management. The simplicity of LSB Steganography Generative Adversarial Networks can be used to
ensures minimal impact on computational efficiency build cover images with a higher degree of
during the embedding and extraction processes. complexity than ever before, such that it becomes
Although integrating ECC introduces some statistically impossible to detect them as synthetic.
computational overhead, the combined approach This would also mean that GANs should be trained
remains practical for real-time applications. to generate certain types of images for hiding data of
Robustness interest with the highest payload capacity and lowest
In terms of robustness, the Stegno Curve method detection rate. These adaptive algorithms, driven by
provides dual-layer security, making it resilient AI, might use the content of an image and features
against both cryptographic and steganographic of hidden data to choose which pixels to embed the
attacks. This dual-layer protection is achieved by best, thus further increasing robustness and stealth in
combining the strengths of ECC and LSB steganographic processing.
Steganography. While Multi-Level Steganography The fast-paced development and sophistication in
(MLS) approaches, which employ multiple the field of quantum computing will result in it being
encryption algorithms (e.g., AES, Blowfish) and able to break even the strongest cryptographic
steganographic methods, also offer robust security, algorithms, including the Elliptic Curve Encryption
they tend to be more complex to implement and technique. Hence, research in the near future must
manage. Although LSB Steganography's payload have a strong emphasis on the integration of post-
capacity is limited by the cover image, ECC's quantum cryptographic algorithms with LSB
efficient encryption with smaller key sizes optimizes Steganography. Post-Quantum Cryptography:
the available payload space. Design of quantum-secure cryptographic
Practical Applicability algorithms. Direction for research is in the
The balance between secure and covert identification of suitable post-quantum
communication can be kept in the integration of cryptographic techniques and their incorporation
ECC with LSB steganography under the Stegno with lattice-based, hash-based, or multivariate
Curve method. Since ECC is computationally polynomial-based cryptography, combined with
complex enough for high security with a relatively LSB Steganography. This would ensure that a robust
small key size, whereas LSB steganography is very form of the secure communication methodology can
simple and provides the most effective way for be adapted to deal with the emerging threats due to
embedding data, this combination makes the method quantum computers. In addition to this, post-
practical and useful for real-time applications within quantum methods will have to be scrutinized for
an environment with limited computational viability and real-world applications in order for
resources. The efficiency and robustness of this them to be taken as effective and thus useful in
technique establish it as a salient solution in various developments, including the transition to
safeguarding sensitive information in a host of secure cloud methods, smart home use, and data
different digital communication scenarios. encryption.
Regarding practical applicability, the Stegno Curve Investigation along such research directions in
method is well-suited for scenarios requiring high secure communication allows the field to mature
security and covert communication, such as secure further not only by addressing present limitations but
messaging, military intelligence, and confidential also by considering future technological
data transfer. On the other hand, GAN-based development. Such efforts will result in the
steganography excels in creating highly availability of more robust and effective ways to
undetectable stego images but demands substantial secure sensitive information in an increasingly
computational resources for training GAN models. digital and networked world.
The implementation complexity of the ECC-LSB Encryption and Decryption Times: ECC
approach is significantly lower compared to GAN- operations are generally faster than RSA due to
based methods, which require extensive shorter key lengths and more efficient computations.
455
 Homam EL-TAJ / IJCESEN 10-3(2024)434-460

Recent advancements have further optimized ECC The ability to conceal sensitive information within
algorithms, significantly improving encryption and innocuous cover media while maintaining
decryption speeds [77]. cryptographic security opens new avenues for secure
Steganographic Embedding and Extraction: The communication in environments where privacy and
time required for embedding and extracting data secrecy are paramount.
using LSB Steganography is relatively low. When The Stegno Curve method demonstrates remarkable
combined with ECC, the overall computational load performance in encryption and steganographic
increases slightly due to the encryption process, but operations. During testing, remarkable speed in
remains efficient enough for practical applications, encrypting and decrypting messages using ECE,
especially with modern optimizations [5]. with encryption speeds increasing from 2.07 GHz to
Resistance to Attacks: The combination of ECC and 3.50 GHz with minimal memory usage of around
LSB Steganography provides a dual layer of 42.4MB was observed. Additionally, the method
security. ECC ensures that data is securely exhibits efficient embedding and extraction of
encrypted, while LSB Steganography hides the encrypted data within images using LSB
presence of this data within cover media. This dual steganography, maintaining high-speed performance
approach significantly reduces the likelihood of even with large image files. This efficient resource
successful attacks compared to using either method utilization ensures optimal performance without
alone [7]. compromising security.
The feature comparison table of the Stegno Curve Furthermore, integrating ECE and LSB
and other traditional methods weighs the steganography offers practical implications for
capabilities of the Stegno Curve method and other domains requiring covert data transmission. The
traditional approaches to shed light on the ability to conceal sensitive information within
significance of each method and compare the innocuous cover media while maintaining
features such as their encryption algorithm, cryptographic security opens new avenues for secure
steganography technique, integration of encryption communication in environments where privacy and
and steganography, performance, and more for a secrecy are paramount. This makes the Stegno Curve
better understanding. Table 3 shows the superiority method particularly suitable for military,
of the Stegno Curve method in the integration of intelligence, and cybersecurity applications, where
encryption and steganography over other traditional covert communication channels are essential.
methods, while it also offers top-notch ECC Overall, this research contributes to advancing the
encryption, better resistance to quantum attacks, and field of secure communication by introducing a
encryption combined with steganography to ensure novel approach that addresses the limitations of
better security. existing techniques while offering practical insights
and methods. By harnessing the synergistic potential
5. Conclusions of ECE and LSB steganography, the proposed fusion
system represents a significant step forward in the
This research has presented a novel fusion of Elliptic quest for robust and covert communication channels
Curve Encryption (ECE) with the Least Significant in the digital world.
Bit (LSB) Steganography for hidden communication
in the Stegno Curve method, offering a Author Statements:
comprehensive and robust framework for secure
data transmission. Several key findings have  Ethical approval: The conducted research
emerged through meticulously exploring theoretical is not related to either human or animal use.
foundations, practical implementation, and detailed  Conflict of interest: The authors declare
methodology. that they have no known competing
Firstly, integrating ECE and LSB steganography in financial interests or personal relationships
the Stegno Curve provides a multi-layered defence that could have appeared to influence the
mechanism that ensures confidentiality, integrity, work reported in this paper
and stealth in communication channels. By  Acknowledgement: This project was
encrypting the message before embedding, the funded by Dar Al-hekma University.
fusion system mitigates the risk of unauthorized  Data availability statement: The data that
access or detection, enhancing the security posture support the findings of this study are
of hidden communication. available on request from the corresponding
Secondly, the fusion of ECE and LSB steganography author. The data are not publicly available
in the Stegno Curve offers practical implications for due to privacy or ethical restrictions.
various domains requiring covert data transmission,
including military, intelligence, and cybersecurity.
456
 Homam EL-TAJ / IJCESEN 10-3(2024)434-460

Table 3: Feature comparison of Stegno Curve and other traditional methods

Features Stegno Curve Traditional RSA- Multi-Level GAN-based
Comparison (ECC-LSB) AES Integration Steganography Steganography
(MLS)
Encryption Elliptic Curve RSA for key AES and Blowfish Varies (typically
Algorithm Encryption exchange, AES for for encryption GAN models for
encryption embedding)
Steganography LSB (Least LSB, DCT, Spread Multi-level GAN-generated
Technique Significant Bit) Spectrum techniques (LSB, cover images
DWT, etc.)
Integration of Yes No Yes Yes
Encryption and
Steganography
Performance Superior speed in Moderate to high Moderate to high High computational
encryption and computational load computational load load (GAN training)
steganography
operations
Security Robust ECC Strong (RSA and Strong due to Very strong (GAN-
encryption, resistant AES), but RSA is multiple encryption based
to quantum attacks vulnerable to layers undetectability)
quantum attacks
Covert Yes Yes Yes Yes
Communication
Key Size Smaller key sizes Larger key sizes for Varies based on Varies based on the
due to ECC (e.g., RSA (e.g., 3072-bit) encryption steganography
256-bit ECC) algorithms model
Resistance to High resistance to Vulnerable to High resistance due High resistance due
Attacks brute force and quantum attacks to multiple to GAN-based
mathematical (RSA) encryption layers models
attacks
Usability User-friendly Depends on specific Complex, requires Very complex,
interface, seamless implementation careful requires machine
integration implementation learning expertise
Application Range Suitable for covert General-purpose Secure Highly secure
communication, encryption and communication, communication,
secure messaging, steganography data protection data hiding
data protection

[5] Alanzy M, Alomrani R, Alqarni B, Almutairi S.

References (2023). Image Steganography Using LSB and Hybrid
Encryption Algorithms. Applied Sciences [Internet].
[1] Jan, A., Parah, S.A., Hussan, M. et al. (2022). Double 1;13(21):11771. Available from:
layer security using crypto-stego techniques: a https://www.mdpi.com/2076-3417/13/21/11771
comprehensive review. Health Technol. 12; 9–31. [6] McAteer I, Ibrahim A, Zheng G, Yang W, Valli C.
https://doi.org/10.1007/s12553-021-00602-1 (2019). Integration of Biometrics and
[2] Evaluating the Merits and Constraints of Steganography: A Comprehensive Review.
Cryptography-Steganography Fusion: A Systematic Technologies [Internet]. 8;7(2):34. Available from:
Analysis [Internet]. www.researchsquare.com. 2023 https://www.mdpi.com/2227-7080/7/2/34
[cited 2024 Feb 11]. Available from: [7] G. G, C. Ashwin, B. V. P, A. A and A. Hiremath,
https://www.researchsquare.com/article/rs- (2021). Enhanced Data Encryption in IOT using ECC
3167378/v1 Cryptography and LSB Steganography. International
[3] Menezes, A.J., Vanstone, S.A. (1993). Elliptic curve Conference on Design Innovations for 3Cs Compute
cryptosystems and their implementation. J. Communicate Control (ICDI3C), Bangalore, India,
Cryptology 6; 209–224. pp. 173-177, doi:
https://doi.org/10.1007/BF00203817 10.1109/ICDI3C53598.2021.00043.
[4] Kakunuri Sandya and Subhadra Kompella, (2022). A [8] Budianto, Christofer & Wicaksana, Arya & Hansun,
Combined Approach of Steganography and Seng, (2020). Elliptic Curve Cryptography and LSB
Cryptography with Generative Adversarial Steganography for Securing Identity Data. Available:
Networks: Survey. Lecture notes in networks and https://www.researchgate.net/publication/33532907
systems, pp. 187–196, doi: 7_Elliptic_Curve_Cryptography_and_LSB_Stegano
https://doi.org/10.1007/978-981-19-4863-3_18. graphy_for_Securing_Identity_Data

457
 Homam EL-TAJ / IJCESEN 10-3(2024)434-460

[9] Varghese, Fredy and P. Sasikala. (2023). Secure Data and Innovative Research 10.
Transmission Using Optimized Cryptography and https://www.jetir.org/papers/JETIR2307591.pdf
Steganography Using Syndrome-Trellis Coding. [21] Y.-K. Lee, G. Bell, S.-Y. Huang, R.-Z. Wang, and S.-
Wireless Personal Communications 130;551-578. J. Shyu, (2009). An Advanced Least-Significant-Bit
[10] G. Shilpa, Mohan CM, Chaurasia N. (2023). Image Embedding Scheme for Steganographic Encoding.
steganography using LSB embedding and edge Lecture Notes in Computer Science, pp. 349–360,
adaptive approach. AIP Conference Proceedings doi: https://doi.org/10.1007/978-3-540-92957-4_31.
[11] R. Bansal and N. Badal, (2022). A novel approach for [22] Aslam, Muhammad & Rashid, Muhammad & Azam,
dual layer security of message using Steganography Farooque & Abbas, Muhammad & Rasheed, Yawar
and Cryptography. Multimedia Tools and & Alotaibi, Saud & Anwar, Muhammad, (2022).
Applications, 81(15);20669–20684, doi: Image Steganography using Least Significant Bit
https://doi.org/10.1007/s11042-022-12084-y. (LSB) - A Systematic Literature 2-38,
[12] Zachariah, Babangida & Yabuwat, Patience & [23] S. Roy and Md. M. Islam, (2022). A Hybrid Secured
Bernard, Ephraim, (2016). Application of Approach Combining LSB Steganography and AES
Steganography and Cryptography for Secured Data Using Mosaic Images for Ensuring Data Security. SN
Communication – A Review,” International Journal Computer Science, 3(2), doi:
of Engineering Research & Technology (IJERT), doi: https://doi.org/10.1007/s42979-022-01046-8.
https://www.researchgate.net/publication/30625410 [24] [1]S. Wendzel et al., (2021). A Revised Taxonomy of
7_Application_of_Steganography_and_Cryptograph Steganography Embedding Patterns. doi:
y_for_Secured_Data_Communication_-_A_Review https://doi.org/10.1145/3465481.3470069.
[13] Dhivya C, Sharmila G, Keerthanadevi S, [25] A Robust Security Scheme Based on Novel Encoding
Gangalakshmi S. Steganography (2018). A with LSB Steganography | IEEE Conference
Technique to Hide the Information using LSB Publication | IEEE Xplore ieeexplore.ieee.org.
Algorithm [Internet]. SSRG International Journal of https://ieeexplore.ieee.org/document/9513904
Ceramic Technology (SSRG-IJCT) -Special Issue. [26] Improving Security with Efficient Key Management
2018 [cited 2024 Feb 10]. Available from: in Public Cloud using Hybrid AES, ECC, and LSB
https://www.internationaljournalssrg.org/uploads/sp Steganography comparing with Novel hybrid Cube
ecialissuepdf/ICRMIT/2018/CSE/IJCSE-ICRMIT- Base Obfuscation | IEEE Conference Publication |
P103.pdf IEEE Xplore [Internet]. ieeexplore.ieee.org. [cited
[14] Ahmed DEM, Khalifa OO. (2014). Robust and 2024 Feb 10]. Available at:
Secure Image Steganography Based on Elliptic https://ieeexplore.ieee.org/document/9780661
Curve Cryptography [Internet]. IEEE Xplore. 288– [27] E. Bin Hureib, P. Adnan, and A. Gutub, (2020).
91. https://ieeexplore.ieee.org/document/7031659 Enhancing Medical Data Security via Combining
[15] Kheddar H, Hemis M, Himeur Y, Megías D, Amira Elliptic Curve Cryptography and Image
A. (2023). Deep Learning for Diverse Data Types Steganography. IJCSNS International Journal of
Steganalysis: A Review arXiv.org. Computer Science and Network Security, 20(8):
https://arxiv.org/abs/2308.04522 http://paper.ijcsns.org/07_book/202008/20200801.p
[16] Dhivya C, Sharmila G, Keerthanadevi S, df
Gangalakshmi S. Steganography (2024). A [28] Chandrasekaran, Vanmathi & Sevugan, Prabu,
Technique to Hide the Information using LSB (2016). Steganography: A comparative study,
Algorithm. SSRG International Journal of Ceramic analysis of key issues and current trends. 8;4696-
Technology (SSRG-IJCT) -Special Issue. 4715,
htps://www.internationaljournalssrg.org/uploads/spe https://www.researchgate.net/publication/30967070
cialissuepdf/ICRMIT/2018/CSE/IJCSE-ICRMIT- 0_Steganography_A_comparative_studyanalysis_of
P103.pdf _key_issues_and_current_trends
[17] J. Bhadra, M. K. Banga and M. V. Murthy, (2017). [29] Al-Shaaby, Ahmed & Al-Kharobi, Talal, (2017).
Securing data using elliptic curve cryptography and Cryptography and Steganography: New Approach.
least significant bit steganography. International Transactions on Networks and Communications. 5.
Conference On Smart Technologies For Smart 10.14738/tnc.56.3914,
Nation (SmartTechCon), Bengaluru, India, pp. 1460- [30] D. Nashat and L. Mamdouh, (2019). An efficient
1466, doi: 10.1109/SmartTechCon.2017.8358607. steganographic technique for hiding data. Journal of
[18] Kaur, Ravneet & Singh, Tanupreet, “Hiding Data in the Egyptian Mathematical Society, 27(1), doi:
video Sequences using LSB with Elliptic Curve https://doi.org/10.1186/s42787-019-0061-6.
Cryptography,” International Journal of Computer [31] Provably Secure Public-Key Steganography Based
Applications. 117;36-40. 10.5120/20658-3296. on Elliptic Curve Cryptography | IEEE Journals &
[19] Christofer Derian Budianto, Arya Wicaksana, and Magazine | IEEE Xplore [Internet].
Seng Hansun, “Elliptic Curve Cryptography and LSB ieeexplore.ieee.org. [cited 2024 Feb 10].
Steganography for Securing Identity Data (2019). https://ieeexplore.ieee.org/abstract/document/10418
Studies in computational intelligence, pp. 111–127, 202
doi: https://doi.org/10.1007/978-3-030-25217-5_9. [32] R. Denis and P. Madhubala, (2020). Evolutionary
[20] Nandan P, Raghav M, 2, Amit K, Chaturvedi, Scholar Computing Assisted Visually- Imperceptible Hybrid
P. (2023) Issue 7 Journal of Emerging Technologies Cryptography and Steganography Model for Secure
Data Communication over Cloud Environment.
458
 Homam EL-TAJ / IJCESEN 10-3(2024)434-460

International Journal of Computer Networks and based WSN. Innovative Data Communication
Applications, 7(6);208, doi: Technologies and Application: ICIDCA 2019 (pp.
https://doi.org/10.22247/ijcna/2020/205321. 742-749). https://doi.org/10.1007/978-3-030-38040-
[33] Siddharth, D., & Saini, D. K. J. (2022). IoT Based 3_85
Lightweight Cryptographic Schemes in Smart [45] Agrawal, L., & Tiwari, N. (2020). A review on IoT
Healthcare. In Advancing Computational Intelligence security architecture: attacks, protocols, trust
Techniques for Security Systems Design (pp. 169- management issues, and elliptic curve cryptography.
178). CRC Press. Social Networking and Computational Intelligence:
[34] Ahmed, A. A., Malebary, S. J., Ali, W., & Alzahrani, Proceedings of SCI-2018, 457-465.
A. A. (2023). A provable secure cybersecurity https://doi.org/10.1007/978-981-15-2071-6_36
mechanism based on combination of lightweight [46] Khan, M. A., Quasim, M. T., Alghamdi, N. S., &
cryptography and authentication for internet of Khan, M. Y. (2020). A secure framework for
things. Mathematics, 11(1);220. authentication and encryption using improved ECC
https://doi.org/10.3390/math11010220 for IoT-based medical sensor data. IEEE Access, 8,
[35] Salih, K. O. M., Rashid, T. A., Radovanovic, D., & 52018-52027.
Bacanin, N. (2022). A comprehensive survey on the https://doi.org/10.1109/ACCESS.2020.2980739
Internet of Things with the industrial marketplace. [47] Kavitha, S., Alphonse, P. J. A., & Reddy, Y. V.
Sensors, 22(3), 730. (2019). An improved authentication and security on
https://doi.org/10.3390/s22030730 efficient generalized group key agreement using
[36] Harn, L., & Lin, C. (2014). Efficient group Diffie– hyper elliptic curve based public key cryptography
Hellman key agreement protocols. Computers & for IoT health care system. Journal of medical
Electrical Engineering, 40(6), 1972-1980. systems, 43(8), 260. https://doi.org/10.1007/s10916-
https://doi.org/10.1016/j.compeleceng.2013.12.018 019-1378-2
[37] Rahmani, A.M., Bayramov, S. & Kiani Kalejahi, B. [48] Mehmood, M. S., Shahid, M. R., Jamil, A., Ashraf,
(2022). Internet of Things Applications: R., Mahmood, T., & Mehmood, A. (2019,
Opportunities and Threats. Wireless Pers Commun November). A comprehensive literature review of
122, 451–476. https://doi.org/10.1007/s11277-021- data encryption techniques in cloud computing and
08907-0 IoT environment. 2019 8th International Conference
[38] Nyangaresi, V. O. (2022). Lightweight anonymous on Information and Communication Technologies
authentication protocol for resource-constrained (ICICT) (pp. 54-59). IEEE.
smart home devices based on elliptic curve https://doi.org/10.1109/ICICT47744.2019.9001945
cryptography. Journal of Systems Architecture, 133, [49] Vahdati, Z., Yasin, S., Ghasempour, A., & Salehi, M.
102763. (2019). Comparison of ECC and RSA algorithms in
https://doi.org/10.1016/j.sysarc.2022.102763 IoT devices. Journal of Theoretical and Applied
[39] Das, S., & Namasudra, S. (2022). A novel hybrid Information Technology, 97(16), 4293.
encryption method to secure healthcare data in IoT- [50] Kumari, S., Karuppiah, M., Das, A. K., Li, X., Wu,
enabled healthcare infrastructure. Computers and F., & Kumar, N. (2018). A secure authentication
Electrical Engineering, 101, 107991. scheme based on elliptic curve cryptography for IoT
https://doi.org/10.1016/j.compeleceng.2022.107991 and cloud servers. The Journal of Supercomputing,
[40] Xavier, A. P., & Kesavan, R. (2023). Hybrid Elliptic 74(12);6428-6453. https://doi.org/10.1007/s11227-
Curve Cryptographic Approach for Data Privacy and 017-2048-0
Authentication in Secured Map Reduce Layer (SMR) [51] Lara-Nino, C. A., Diaz-Perez, A., & Morales-
for Optimized CPU Utilization. IETE Journal of Sandoval, M. (2018). Elliptic curve lightweight
Research, 69(2), 670-683. cryptography: A survey. IEEE Access, 6, 72514-
https://doi.org/10.1080/03772063.2022.2027285 72550.
[41] Qazi, R., Qureshi, K. N., Bashir, F., Islam, N. U., https://doi.org/10.1109/ACCESS.2018.2881444
Iqbal, S., & Arshad, A. (2021). Security protocol [52] Odat, A. M., & Otair, M. A. (2016). Image
using elliptic curve cryptography algorithm for steganography using modified least significant bit.
wireless sensor networks. Journal of Ambient Indian Journal of Science and Technology.
Intelligence and Humanized Computing, 12, 547- https://ischolar.sscldl.in/index.php/indjst/article/vie
566. https://doi.org/10.1007/s12652-020-02020-z w/125794
[42] A. Abdaoui, A. Erbad, A. K. Al-Ali, A. Mohamed [53] Hamid, N., Yahya, A., Ahmad, R. B., & Al-Qershi,
and M. Guizani, (2022). Fuzzy Elliptic Curve O. M. (2012). Image steganography techniques: an
Cryptography for Authentication in Internet of overview. International Journal of Computer Science
Things. IEEE Internet of Things Journal, 9(12); and Security (IJCSS), 6(3);168-187.
9987-9998, doi: 10.1109/JIOT.2021.3121350 [54] Akhtar, N., Johri, P., & Khan, S. (2013). Enhancing
[43] Mousavi, S. K., Ghaffari, A., Besharat, S., & Afshari, the security and quality of LSB based image
H. (2021). Security of internet of things based on steganography. 5th International Conference and
cryptographic algorithms: a survey. Wireless Computational Intelligence and Communication
Networks, 27(2), 1515-1555. Networks (pp. 385-390). IEEE.
https://doi.org/10.1007/s11276-020-02535-5 https://doi.org/10.1109/CICN.2013.85
[44] Pawar, R., & Kalbande, D. R. (2020). Elliptical curve [55] Muhammad, K., Ahmad, J., Farman, H., & Zubair,
cryptography based access control solution for IoT M. (2015). A novel image steganographic approach
459
 Homam EL-TAJ / IJCESEN 10-3(2024)434-460

for hiding text in color images using HSI color model. (ICNSC2014).
arXiv preprint arXiv:1503.00388. https://doi.org/10.1109/cnsc.2014.6906644
https://doi.org/10.48550/arXiv.1503.00388 [68] Gaba, J., & Kumar, M. (2013). Implementation of
[56] Wang, S., Sang, J., Song, X., Niu, X. (2015). Least steganography using CES technique. IEEE Second
significant qubit (LSQb) information hiding International Conference on Image Information
algorithm for quantum image. Measurement 73;352- Processing (ICIIP-2013), Shimla, India (pp. 395-
359. 399). https://doi.org/10.1109/iciip.2013.6707622
https://doi.org/10.1016/j.measurement.2015.05.038 [69] Nofriansyah, D., Defit, S., Nurcahyo, G. W., Ganefri,
[57] Al-Shatnawi, A. M., & AlFawwaz, B. M. (2013). An G., Ridwan, R., Ahmar, A. S., & Rahim, R. (2018).
integrated image steganography system with A new image encryption technique combining Hill
improved image quality. Applied Mathematical cipher method, Morse code and least significant bit
Sciences, 7(71), 3545-3553. algorithm. Journal of Physics: Conference Series,
[58] Liu, Q., Sung, A. H., Ribeiro, B., Wei, M., Chen, Z., 954, 012003. https://doi.org/10.1088/1742-
& Xu, J. (2008). Image complexity and feature 6596/954/1/012003
mining for steganalysis of least significant bit [70] Sridevi, D. R., Paruchuri, V. L., & Rao, K. S. S. S.
matching steganography. Information Sciences, (2013). Image steganography combined with
178(1), 21-36. cryptography. international journal of computers &
https://doi.org/10.1016/j.ins.2007.08.007 technology, 9(1); 976–984.
[59] Xu, W. L., Chang, C. C., Chen, T. S., & Wang, L. M. https://doi.org/10.24297/ijct.v9i1.4160
(2016). An improved least-significant-bit [71] Shoup, V. (2009). A computational introduction to
substitution method using the modulo three strategy. number theory and algebra. Cambridge University
Displays, 42, 36-42. Press.
https://doi.org/10.1016/j.displa.2016.03.002 [72] Kak, A. (2015). Lecture notes on computer and
[60] Mao, Q. (2014). A fast algorithm for matrix network security. Purdue University. Available:
embedding steganography. Digital Signal https://engineering.purdue.edu/kak/compsec/NewLe
Processing, 25; 248-254. ctures/Lecture8.pdf
https://doi.org/10.1016/j.dsp.2013.11.001 [73] Stallings, W. (2011). Operating systems: internals
[61] Nagaraj, V., Vijayalakshmi, V., & Zayaraz, G. and design principles. Prentice Hall Press.
(2013). Color image steganography based on pixel Available:
value modification method using modulus function. https://dl.acm.org/doi/abs/10.5555/2012029
IERI Procedia, 4; 17-24. [74] Abd Aziz, A. Z., Mohd Sultan, M. F., & Mohamad
https://doi.org/10.1016/j.ieri.2013.11.004 Zulkufli, N. L. (2024). Image Steganography:
[62] Biswas, D., Biswas, S., Majumder, A., Sarkar, D., Comparative Analysis of their Techniques,
Sinha, D., Chowdhury, A., & Das, S. K. (2012). Complexity and Enhancements. International
Digital image steganography using dithering Journal on Perceptive and Cognitive Computing,
technique. Procedia Technology, 4; 251-255. 10(1); 59–70.
https://doi.org/10.1016/j.protcy.2012.05.038 https://doi.org/10.31436/ijpcc.v10i1.449
[63] Karim, S. M., Rahman, M. S., & Hossain, M. I. [75] Ji, H., Xie, L., Wang, C., Yin, Y., & Lu, S. (2015).
(2011). A new approach for LSB based image CrowdSensing: A crowd-sourcing based indoor
steganography using secret key. 14th international navigation using RFID-based delay tolerant network.
conference on computer and information technology Journal of Network and Computer Applications, 52,
(ICCIT 2011) (pp. 286-291). IEEE. 79-89. https://doi.org/10.1016/j.jnca.2015.02.010
https://doi.org/10.1109/ICCITechn.2011.6164800 [76] Rahman, M. S., Hossain, M. S., Rahat, E. H., Dipta,
[64] Tahir, A., & Amit, D. (2015). A novel approach of D. R., Faruque, H. M. R., & Fattah, F. K. (2019).
LSB based steganography using parity checker. Efficient hardware implementation of 256-bit ECC
International Journal of Advanced Research in processor over prime field. International conference
Computer Science and Software Engineering, 5(1). on electrical, computer and communication
[65] Rawat, D., & Bhandari, V. (2013). A steganography engineering (ECCE) (pp. 1-6). IEEE.
technique for hiding an image in an image using LSB https://doi.org/10.1109/ECACE.2019.8679184
method for 24 bit color image. International Journal [77] Mahto, D., & Yadav, D. K. (2017). RSA and ECC: a
of Computer Applications, 64(20); 15–19. comparative analysis. International Journal of
https://doi.org/10.5120/10749-5625 Applied Engineering Research, 12(19); 9053-9061.
[66] ElSayed, A., Elleithy, A., Thunga, P., & Wu, Z. https://blkcipher.pl/assets/pdfs/ijaerv12n19_140.pdf
(2015). Highly secure image steganography [78] Maimuţ, D., & Matei, A. C. (2022). Speeding-Up
algorithm using curvelet transform and DCT Elliptic Curve Cryptography Algorithms.
encryption. Long Island Systems, Applications and Mathematics, 10(19); 3676.
Technology. https://doi.org/10.3390/math10193676
https://doi.org/10.1109/lisat.2015.7160204
[67] Senthooran, V., & Ranathunga, L. (2014). DCT
coefficient dependent quantization table modification
steganographic algorithm. First International
Conference on Networks & Soft Computing

460