www.nature.

com/scientificreports

OPEN An efficient image encryption

algorithm using 3D-cyclic
chebyshev map and elliptic curve
Shamsa Kanwal1, Saba Inam1, Shaha Al-Otaibi2, Javaria Akbar3, Nasir Siddiqui3 &
Muhammad Ashiq4
With the development of internet technologies, it is now usual to communicate enormous amounts of
text and visual data over networks, which calls for stringent procedures to guarantee confidentiality
and integrity throughout transmission. An important part of safeguarding image communication
over secure channel is cryptography. This study proposes an effective way to secure sensitive data by
creating tamper-proof cryptosystems and authentication mechanisms. By employing elliptic curve
cryptography (ECC) to generate secret encryption key and chaotic maps to successfully disguise and
alter plain image pixel values, the technique strengthens overall system security through the use of
hybrid cryptography. The method has a big enough key space to withstand brute-force attempts and
exhibits robustness against statistical and differential attacks through analysis including evaluations
and assessments of correlation, entropy, and histograms as well as evaluations such as NPCR, UACI,
PSNR and keyspace analysis validate the efficacy of the proposed image encryption algorithm,
affirming its systematic and robust nature and its capacity to offer superior image encryption
protection over some existing ones.

Keywords Image encryption, Chaotic maps, Elliptic curve cryptography, Entropy, Correlation, Unified
average changing intensity

The rapid advancements in internet and information technologies have profoundly impacted all facets of our
companies, businesses, and with greater ease in our everyday lives. Encrypting larger size image files within a
reasonable time frame becomes a challenge. Given the highly sensitive nature of the data contained in images,
ensuring their confidentiality and protection against unauthorized access during storage and sharing has become
imperative.
When dealing with sensitive information, the challenge of secure data transfer emerges. To ensure data
security, cryptography can be employed to safeguard the data being transferred. Image encryption has emerged
as a successful technique for safeguarding private images during communication, leading to the proposal of
numerous methods such as1. The transmission of image information is on the rise, prompting cryptographers
to focus on this area.Image encoding typically exhibit high pixel correlation and possess large data sizes.
Conventional encryption techniques like AES2 are not good enough for multimedia files due to their limited
capacity for handling public data, as well as the high pixel correlation and large amount of redundancy present
in such files3 where by using different statistical attacks the original images will be accessed by attackers.
To meet the requirements of image encryption, various techniques have been introduced for digital images to
ensure robust security and efficiency. In4 the author presented a scheme by utilizing the Pan-Tompkins Algorithm
(PTA) with the Least Significant Bit (LSB) method. In literature, different authors introduced Fractional-order
weighted Spherical Bessel-Fourier Moments (FrSBFMs) technique5,6, and further enhance them for color image
analysis by integrating with quaternion theory to develop Quaternion FrSBFMs (QFrSBFMs), with a detailed
discussion on their rotation invariance and parameter analysis. Fariha et al.7 introduced a novel technique by
combining permutation-based S-box generation with sine series expansion to meet the specific cryptographic
needs.
For addressing communication problems related to images can be solved by Chaos theory. Chaotic maps
exhibit key features8 such as initial conditions sensitivity, controlling variables, unpredictability and these

1Department of Mathematical Sciences, Fatima Jinnah Women University, Rawalpindi, Pakistan. 2Department of
Information Systems, College of Computer and Information Sciences, Princess Nourah bint Abdulrahman University,
P. O. Box 84428, Riyadh 11671, Saudi Arabia. 3Department of Mathematical Sciences, University of Engineering and
Technology, Taxila, Pakistan. 4Department of Humanities and Basic Sciences, MCS Campus, National University of
Sciences and Technology, Islamabad, Pakistan. email: [email protected]

Scientific Reports | (2024) 14:29626 | https://doi.org/10.1038/s41598-024-77955-w 1

www.nature.com/scientificreports/

properties9–13 effectively fulfill the various cryptographic needs like avalanche effect, bit-level balancing,
confusion and diffusion. Multidimensional chaotic systems enhance the security and efficiency of image
encryption by generating pseudorandom sequences for encoding, utilizing diffusion and permutation stages14
to achieve a robust cryptosystem based on chaos theory. As stated in Refs.15,16 the shuffling step aims to change
the locations of the pixels in the image while preserving their original values, effectively there is almost little
association between neighboring pixels in Refs.17,18.
For the improvement of the chaotic system performance and extension in the range of map, Zhu et al.19
suggested novel system utilized the Sine and Tent chaotic maps (STS) which is compound chaotic map. Adeyefa
et al.20 developed a new cryptographic scheme that uses Chebyshev polynomials as a key component of the
encryption function and Laplace transforms as the transformational tool. Integration of 3D cat map and logistic
map with symmetric image based cryptosystem was presented by Chen et al.21 for the purpose of shuffling
picture data. The algorithm add up the chaotic layer between the original and encrypted images every two
rounds of shuffling. Chen’s chaotic system is also used in the encryption to generate keys. Different schemes
are suggested by Kanwal et al. to encrypt the image by employing chaotic maps with orthogonal matrices and
circulant matrices in Refs.22,23. In separate studies24–26, authors proposed different schemes that utilized the
Chebyshev chaotic maps in different technologies in combination with cryptographic hash functions, modular
exponentiation and elliptic curve scalar multiplication, to ensure the data privacy in communications and secret
key generation for secure access.
In this paper, we have employed Elliptic Curve Cryptography (ECC) and Chaos theory as our cryptographic
techniques. Numerous studies27,28 on ECC demonstrated that elliptic curve based Discrete Logarithmic Problem
(DLP) becomes more challenging with increasing key sizes and this characteristic renders it highly favorable for
encoding and decoding processes. Various schemes using the combination of elliptic curves and chaotic maps
have been proposed in literature such as29–34. Also several security challenges have been highlighted in previous
studies, including vulnerabilities to brute-force attacks and the limitations posed by inadequate keyspace
size35. In response to these challenges, our proposed encryption scheme offers a robust solution by leveraging a
significantly large keyspace. With a keyspace size of 2405, our scheme provides exceptional security, substantially
mitigating the risk of brute-force attacks and enhancing overall cryptographic strength.
Outline of the paper is in the following manner. Essential background of our scheme is covered by “Basic
mathematical theories”. The suggested image based hybrid technique is described by “Proposed digital image
algorithm”. In “Experimental results and discussion”, we present our model experiments and their results using
chaotic systems. In “Security evaluation and comparative analysis”, we evaluate the model’s using statistical and
differential assaults. Finally, in “Conclusion” we present the conclusion.

Basic mathematical theories

This section highlights the some basic mathematical preliminaries of proposed scheme.

Applied chaotic maps

Several chaotic systems are being employed in the field of information security. Chaotic maps are mathematical
models characterized by their sensitivity to initial conditions. A modest alteration in starting values can cause a
notable variations in outcomes of the chaotic system, often resulting in vastly different behaviors or trajectories.
This section provides an overview of fundamental chaotic maps, dynamic compound chaotic systems created by
merging tent and cubic chaotic map36 and 3D-cyclic symmetric Chebyshev map (3D-CSCM)37.

Cubic–tent map
By combining the tent and cubic map proposed a novel compound piecewise chaotic system to address the
problems which are present in both maps. Cubic–Tent system is the term given to this newly created system.
The Cubic–Tent system is created by merging the Cubic and Tent maps, resulting in superior chaotic features
      
mod 4 − 34 r un 1 − u2n + 2r un , 1   if un < 0.5
than either map alone. The equation is un+1 =
mod 4 − 34 r un 1 − u2n + 2r (1 − un) , 1 if un ≥ 0.5
where the modulo operation is called mod, the iteration number is n, and r is the control parameter. To make
output data should be inside the range of [0,1], modulo operation is used. Figure 1 demonstrates the bifurcation
visualization of the given map and behavior which is highly chaotic with various pauses across the whole range
[0,4].

3D-cyclic symmetric Chebyshev map (3D-CSCM)

A chaotic map based on traditional Chebyshev map which is a multi-dimensional circularly symmetric map
xn+1 = cos(a1 cos−1(yi))
know as 3D-CSCM. It is defined as yn+1 = cos(a2 cos−1(zi)) here (a1, a2, a3) ∈ (0, 6) are the control parameters
zn+1 = cos(a3 cos−1(xi))
and the seed values are (x0, y0, z0) ∈ (−1, 1). It has several useful characteristics, including the Lyapunov
exponent and behavior randomization. Owing to these qualities, the provided map is highly advised for the use
in cryptographic applications. Displaying 3D-CSCM bifurcation graph in Fig. 2 and easy to determine that this
map has a vast range of chaotic parameters.

Elliptic curve cryptography concept

Suppose Fp an integer field modulo p a prime value which is p > 3. Let E be an elliptic curve which is an
algebraic curve depicted by Fig. 3. Mathematically, it is defined as: y 2 ≡ x3 + qx + r here the parameters

Scientific Reports | (2024) 14:29626 | https://doi.org/10.1038/s41598-024-77955-w 2

www.nature.com/scientificreports/

Fig. 1. Cubic–tent map.

Fig. 2. 3D-CSCM.

q, r ∈ Fp, where p > 3 and fulfill the required condition 4q 3 + 27r2 ̸≡ 0 (mod p) E : Fp comprises of all set of
points (x, y) ∈ Fp × Fp and the point at infinity O which meet the conditions of elliptic curve.The discriminant
requirement shows that the curve has no vertices and self-intersections, indicating that it is non-singular.

Elliptic curve group operations

We take p1 = {(a1, b1)} and p2 = {(a2, b2)} to be two distinct points on the elliptic curve E. The following elliptic
curve E group operations are.

• Point addition: Suppose that two points which are situated on an elliptic curve E are represented by p1 and p2.
After addition we obtained a third point say q = {(a3, b3)}, which should be on the same curve E. Mathemat-
ically, we can define it as: p1 + p2 = q
• Point subtraction: The subtraction of point is defined as: p1 − p2 = q Here p1 = {(a1, b1)} and p2 = {(a1, (−b1)}
.
• Point multiplication: Suppose that the point p = {(a, b)} situates on curve E, then we can calculate the mul-
tiplication of a point p with an integer n by summing p repeatedly n times to itself and resultant value q must
lie on the given curve. Mathematically, it is calculated by: p + p + p + p + p + . . . = np = q
• Point doubling: Point doubling operation is the process of adding point p = {(x, y)} to its own location gives
a resultant value q located on curve E. Point doubling operation define as: p + p = q

Scientific Reports | (2024) 14:29626 | https://doi.org/10.1038/s41598-024-77955-w 3

www.nature.com/scientificreports/

Fig. 3. Elliptic curve.

Elliptic curve Diffie-Hellman Key exchange (ECDHK)

Over insecure channel the secure sharing of the session keys is done by the protocol known as ECDHK. In this
technique both the parties have an access of public and private key pair. The exchange of key is done by using
these steps:

• Step 1: The secret key nA and nB is generated by user A and B. After that user A and B, utilized G a generator
point on the curve and private keys nA, nB to compute the public key (P ub)A and (P ub)(B) and send this
public keys to user B and A respectively.

P ubA =nA · G
P ubB =nB · G

• Step 2: Shared ECDHK key is computed by user A and B by using these two equations:

SAB =nA · PubB

SAB =nB · PubA

Proposed digital image algorithm

This section outlines the stages involved in our proposed hybrid cryptosystem for securing the communication
of confidential images over insecure network. The proposed method involves the generation of sequence using
chaotic map for the permutation phase cipher, utilizing elliptic curve (EC) for the substitution phase cipher,
and employing bit-wise XOR for diffusion. For enhancing the security level, elliptic curve parameters are used
to obtain the self invertible matrix which employed as key with Hill cipher. The input and output of the images
obtained from proposed algorithm are shown in Figs. 4 and 5. The proposed cryptosystem comprises of the
following steps:

Key management
We outline our developed key generation method and selection of control parameters for the suggested
cryptosystem in this subsection. The idea is to create a design that prevents an attacker from computing the
key because it is needed for decryption so, its significance and confidentiality cannot be underestimated.
The suggested scheme is hybrid in nature for securing pictures, by employing both schemes such as private
key and public key. For the expansion of key space and security enhancement we used three distinct secret
keys (k1, k2, k3). For the permutation of image pixel, key k1 = (u0, r) is employed with cubic tent map. Here,
u0 is a seed value and r is the controlling variable of map. In substitution process k2 is generated by using
ECC parameters (a, b, p, nA, nB , G) to develop a self-invertible key matrix of order 4 × 4. Last phase, in pixel
diffusion, we used k3 = (x0, y0, z0, a1, a2, a3) with initial values (x0, y0, z0) and control parameters (a1, a2, a3),
the map is iterated with required iterations.

Image encryption model

For digital image encryption it is recommended to choose a picture with dimension of M × N × 3 pixels, where
M and N represents the rows and columns of image matrix respectively and the “3” denotes the three-color
channels (Red, Green, and Blue). The proposed technique creates a sequence for Shuffling phase using the Cubic

Scientific Reports | (2024) 14:29626 | https://doi.org/10.1038/s41598-024-77955-w 4

www.nature.com/scientificreports/

Fig. 4. Baboon image for experimental results.

Fig. 5. House image for experimental results.

tent map. Self-invertible key matrix is utilized in the substitution phase generated by elliptic curve. Finally, a bit-
XOR is with forward and backward diffusion is used to obtain the jumbled pixels of image for the enhancement
of security by introducing chaotic map name as 3D-cyclic symmetric Chebyshev map. The proposed technique
is intended to encrypt one or more distinct kinds of pictures. The encoded image must have equal dimensions
as input. Our proposed algorithm consists of three basic stages. Suggested algorithm’s flowchart, as depicted in
Fig. 6.

Scrambling algorithm
In the proposed scheme, Algorithm 1 illustrates how this phase shuffles the pixel location of input picture by using
the Cubic-Tent chaotic map. By iterating Cubic tent map, created a chaotic sequence by using key k1 = (u0, r)
repeatedly with required iterations, and then relocate the terms in an increasing manner. By comparing locale of
the generated sequence and ordered sequence to determine the transformation sequence. The original picture’s
pixels are permuted using transformation sequence and this yields a one-dimensional array.

Scientific Reports | (2024) 14:29626 | https://doi.org/10.1038/s41598-024-77955-w 5

www.nature.com/scientificreports/

Fig. 6. Architecture of the proposed scheme.

Scientific Reports | (2024) 14:29626 | https://doi.org/10.1038/s41598-024-77955-w 6

www.nature.com/scientificreports/

Algorithm 1. Pseudo-code of pixel permutation phase.

Substitution algorithm
Substitution process is carried out on permuted image obtained by permutation phase with key k2. A self-
invertible matrix is necessary for the Hill cipher technique in order to permit decryption. The Hill cipher, which
employs a self-inverse matrix with 4 × 4 order based on elliptic curve, is used to carry out the substitution
process, which simplifies and expedites it. During this stage, this matrix produced by employing (ECDHK)
with elliptic curve parameters (a, b, p, nA, nB , G) = (1, 3, 31, 13, 17, (1, 6)) treated as secret key k2, and then
implement the Hill cipher in substitution process depicted by Algorithm 2. It splits the permuted picture into J4
sub-blocks. The produced 4 × 4 self-invertible matrix is multiplied by each of these J4 sub-blocks individually.
The resultant sub-matrices of order 4 × 1 are combined again to obtain the one-dimensional array PS.

Algorithm 2. Pseudo-code of substitution phase.

Diffusion algorithm
The last phase is the diffusion phase which is processed by using the third key which is employed with chaotic map.
A distinct real number sequence is generated by 3D-CSCM with confidential key k3, as a result the sequence is
subsequently transmuted into numerical sequence. Obtained one-dimensional array by the substitution process
is then bit-xoring with this integer sequence. Resultant 1D array is then reorganized in the form of matrix with
J = M × N × 3, which generates the cipher picture. The Algorithm 3 defines the technique.

Scientific Reports | (2024) 14:29626 | https://doi.org/10.1038/s41598-024-77955-w 7

www.nature.com/scientificreports/

Algorithm 3. Pseudo-code of pixels diffusion phase.

Image decryption model

The decryption scheme of a picture aims to retrieve the input image pixels by applying the encoding algorithm
in reverse order. The suggested decryption algorithm comprises three phases, detailed in Algorithm 4. In the
initial phase of an algorithm, the series produced by k3 The key k3 is XORed with the 3D-CSCM map undergoes
bitXOR operation. Hill cipher is employed with the same self invertible matrix generated in encryption algorithm
using k2. The Cubic-tent map generates a random sequence, and through an application of k1, we obtained an
inverse transformation and utilized to revert shuffled pixel. The obtained array is transmuted into picture format
in order to get the input picture.

Algorithm 4. Decryption algorithm.

Scientific Reports | (2024) 14:29626 | https://doi.org/10.1038/s41598-024-77955-w 8

www.nature.com/scientificreports/

Experimental results and discussion

We implemented and evaluated our proposed encryption technique using MATLAB 2016a. The sample pictures
were sourced from the well-known database USC-SIPI Image Database38 to evaluate feasibility of our presented
RGB image based cryptosystem. To decode an encrypted picture, techniques such as pixel permutation, self-
invertible key matrix mixing with Hill cipher, and pixel diffusion are used. Baboon image is selected for the sake
of comparing the outcomes.

Security evaluation and comparative analysis

This section evaluates the proposed algorithm’s statistical and differential parameters. The suggested
algorithm’s strength is investigated using a number of statistical and security tests utilizing proposed Scheme.
For analysis we selected colored image (256 × 256) and performed the encryption using these parameters
k1 = (u0, r) = (0.799, 3.8) for CTM and k3 = (x0, y0, z0, a1, a2, a3) = (0.56, 0.34, 0.78, 3.746, 3.078, 2.379) for
3D-CSCM. The devised cryptosystem was tested for efficiency and resilience against several assaults, ciphered
image uncertainty known as Shannon’s entropy, distribution of pixels, resistance to intruder attacks variance
analysis, chi-square test, possible keys range analysis, differential analysis based on NPCR, UACI, PSNR, and
computation cost comparison.

Histogram statistical evaluation

The analysis of the histogram is an advance method for assessing the pixel distribution within an image.
Frequency of pixel distribution is the visual representation of picture histogram. The image histogram visually
represents the distribution of pixels within an image. Input image histogram with three different components
such as red, green and blue including original image is demonstrated by Fig. 7a–d , illustrating their pixels
distribution.
It aims to distinguish input picture and encoded one. Figure 7e–h shows the uniformity in distribution which
ensure that no data of pixels dispersion will be lost from encoded image in comparison with Fig. 7 depicts
the histogram of input image which exhibits the discrepancy in pixel value. Cipher pictures have a consistent
frequency distribution across histograms, suggesting acceptable quality. This observation reinforces that no
identifiable information regarding the original pixel dispersion is leaked in the cipher image.

Chi-square analysis
More precisely, for the demonstration of encoded image uniformity level we utilized the test known as Chi-
Square test. Objective of this test is to assess homogeneity of pixel distribution and observed coherence in an
encrypted image histogram. High level of consistency in an encoded picture histogram ensure by chi square
 (Ofi −Efi )
minimal value. Mathematical expression for the calculation of chi-square value is given as χ2test = 255
i=0 E fi
recorded and predicted intensity of pixel frequency values are represented by Ofi and Efi, where predicted
value of frequency for i is calculated by Efi = image
256
size
.

Fig. 7. Pixels distribution analysis of original and encoded image.

Scientific Reports | (2024) 14:29626 | https://doi.org/10.1038/s41598-024-77955-w 9

www.nature.com/scientificreports/

Original Cipher
Colored image Size Red Green Blue Red Green Blue Results
Baboon 256 × 256 27904 46231 29066 261 228 218 Pass
House 256 × 256 43861 54631 53167 265 209 233 Pass

Table 1. Original and encrypted colored images Chi-square analysis.

Encryption techniques Entropy value

Presented technique 7.9991
Ref.40 7.9978
Ref.41 7.9976
Ref.42 7.9990
Ref.43 7.9986

Table 2. Comparison of information entropy values.

In our proposed encryption scheme, we calculated chi-square values for some colored images taken from
USC-SIP dataset of different images. Significance level for the introduced technique set at α = 0.05 and α = 0.01
and the hypothesis is accepted at both the levels demonstrated by Table 1. Less value indicates that the dispersion
of pixels is homogeneous in the histogram.

Information entropy results

Within an encoded picture, To assess the randomness or unpredictability in concentration of pixels can be
assessed by utilizing an analytical feature known as Entropy. Shannon’s theory states that in an encrypted image
there is a direct relation between entropy and level of
 unpredictability.
 Formula for the calculation of entropy
N
is given in Ref.39 as H(CI) = 2i=0−1 P (CIi) log2 P (CI 1
i)
Here, cipher image is represented by CI and the
probability of symbols in an encrypted image is P (CIi). Ideally produced encrypted image by an encryption
algorithm should have value of entropy 8. For the secure image based cryptosystem, it must be impossible for an
attacker to anticipate the input image data from encoded picture. The suggested approach calculates the entropy
of ciphered picture H(CI). Table 2 demonstrates the presented scheme entropy value comparison of presented
scheme with other existing algorithms and indicates that there is no data loss because the value of entropy is
near to the optimal level.

Analysis of adjacent pixels correlation

Disrupt a pixel’s association with its adjacent pixels is the main task of any image based cryptosystem. To assess
the quantitative analysis of having ambiguity and dispersion among the adjacent pixels which are adjacent in the
original picture and associated encoded one is done by correlation analysis. Similarity between adjacent pixels
is shown by the coefficients of correlation in three different direction i.e horizontally, vertically, and diagonally.
The mathematical equation [69] that is used to measure correlation is defined below:
    
n np=1 xpyp − np=1 xp np=1 yp
Cr =   2     2 

n np=1 x2p − n
p=1 xp n np=1 yp2 − n
p=1 yp

In above defined equation Cr , xi and yi represents correlation, nearby pixel number and for the calculation of
coefficient, n is selected as a total pixel value. The correlation values are in the range of −1 to 1. Interrelationship
among the nearby pixel is high if correlation number is 1 or near to 1. As a result, picture should be encoded by
our proposed approach with relationship coefficients that are minimal and almost zero.
This approach ensures that any informational patterns are obscured, making it challenging for hackers to
discern meaningful data. Figure 8a–d, represents the correlation of the original image and ciphered image’s pixel
in RGB color bands are depicted by Fig. 8e–h. Table 3 demonstrates the correlation number for input image and
encoded image in all three directions. For encoded picture the value of correlation coefficient near to 0 indicates
that the values of pixels are nearby discordant.

Mean square analysis (MSE)

Mean Square Error (MSE) term is used to distinguish the original and encoded picture. MSE quantifies the
precision and disparity between two images, serving as a measure of their accuracy and divergence. A large
discrepancy among the original and the
cipher image is correlated with high MSE value. MSE is computed by
M −1 N −1
using the given formula MSE = M ×N 1
i=0 j=0 (IP (i, j) − IC (i, j)) . In the formula M and N represents
2

rows and columns respectively. IP , IC are elected for original and encoded images respectively. Table 4
demonstrates the MSE value comparison of suggested technique with literature existing encryption schemes.

Scientific Reports | (2024) 14:29626 | https://doi.org/10.1038/s41598-024-77955-w 10

www.nature.com/scientificreports/

Fig. 8. Pixels distribution analysis of original and encoded image.

Original image Encrypted image

Directions Red Green Blue Red Green Blue
Row 0.8844 0.8392 0.9086 −0.00078 0.0013 −0.0024
Column 0.9159 0.8683 0.9219 −0.0029 −0.0034 0.0059
Diagonal 0.9283 0.8364 0.8799 −0.00048 −0.00087 -0.0043

Table 3. Correlation values between original and encrypted image.

Encryption techniques MSE value

Proposed technique 8722.1
Ref.44 7254.2
Ref.22 8736.9
Ref.23 8783.1

Table 4. Comparison of mean squared error values.

Evaluation of peak signal-to-noise ratio (PSNR)

A tool used to do a quantitative examination of the differences among the input picture and associated encrypted
picture is known as PSNR. This metric provides an information on how much the plain image has deteriorated
after the encryption procedure. For the computation of PSNR  value
 of the picture with the size M × N × 3,
2552
the following equation can be utilized as PSNR = 10 · log MSE (db). For the better quality of image based
cryptosystem, plain and encoded picture PSNR values must be small. The values of PSNR obtained from the
proposed scheme are given in the Table 5.

Differential attack analysis

Sensitivity analysis of plaintext or input picture is known as differential assessment in cryptography. The
Number of Pixel Change Rate (NPCR) and the Unified Average Changing Intensity (UACI) are considered as the
cryptographic strength of an encryption technique against differential attack. Resultant image of an encoding
algorithm affected by a small variation in original picture pixel. To evaluate the impact of small change made to
a plain image on the encrypted cipher is assessed by NPCR. Higher NPCR values ensure that the cryptosystem
is highly secure and that system can withstand various differential attacks. (UACI) criteria is used to assess the

Scientific Reports | (2024) 14:29626 | https://doi.org/10.1038/s41598-024-77955-w 11

www.nature.com/scientificreports/

Encryption techniques PSNR value

Proposed technique 9.4254
Ref.44 9.5248
Ref.23 8.7172
Ref.41 9.1916

Table 5. Comparison of PSNR values.

Encryption techniques NPCR Value UACI Value

Proposed technique 99.63 33.46
Ref45 99.61 33.46
Ref46 99.60 33.43

Table 6. Comparison of NPCR and UACI values.

Fig. 9. (a) Original image (256 × 256), (b) encoded with original key, (c) encoded after slight change in key.

disparity in average intensities among two images. Both of these metrics in cryptography are calculated by given
formulas:

M  N  
1 K(i, j)
NPCR = × 100
M × N i=1 j=1 M × N
M  N  
1 (CI (i, j) − CI′ (i, j))
UACI = × 100
M × N i=1 j=1 255

Consider two input images name as I and I ′ with the same size R × C and their corresponding cipher pictures
are represented by CI and CI′ . If CI (i, j) = CI′ (i, j), then K(i, j) = 0 and K(i, j) = 1, otherwise. 33.33 for UACI
and 99.60 for NPCR are the ideal values for secure cryptosystem. To make the encryption scheme resilience
against differential attack UACI and NPCR should reach the ideal values. Given Table 6 depict the comparison
of suggested technique NPCR and UACI values with the other algorithms.

Key space analysis

Whole collection of keys used during the image algorithm is known as the key space. Two evaluations are
available for crucial space analysis, which takes into account the quantity and sensitivity of the keys.

Sensitivity evaluation of encryption keys

Sensitivity of keys is an essential attribute for secure image based cryptosystem. The cryptosystem is considered
extremely responsive towards the keys if a Infinitesimal alteration in any key results in a remarkable change in an
encoded image. To resilience brute force attack key sensitivity is also important. Cubic-tent and 3D-CSCM map
were employed in this study. The initial seed and control parameters have a significant impact on these chaotic
designs. The produced random sequence will alter entirely if we make a very little alteration to the key or control
parameter. It yields very different encryption/decryption outcomes in return. Baboon as a test image is used to
conduct key sensitivity analysis. Encryption of image is done by these parameters k1 = (u0, r) = (0.799, 3.8)
for CTM and k3 = (x0, y0, z0, a1, a2, a3) = (0.56, 0.34, 0.78, 3.746, 3.078, 2.379) for 3D-CSCM. Figure 9c,

Scientific Reports | (2024) 14:29626 | https://doi.org/10.1038/s41598-024-77955-w 12

www.nature.com/scientificreports/

Encryption techniques Key space

Proposed technique 2405
Ref.47 2249
Ref.42 2282

Table 7. Analysis of Key space between different schemes.

Encryption schemes Computational time

Presented technique 0.0210
Ref.48 0.0735

Table 8. Comparison of time consumption between different algorithms [unit: sec].

Schemes Entropy Correlation NPCR UACI MSE PSNR Keyspace

Proposed 7.9991 −0.0028 99.63 33.46 8722.1 9.4254 10120
Ref.42 7.9991 −0.0004 99.61 33.46 8739.6 8.7172 1072
Ref.41 7.9976 0.0194 99.59 33.43 7833 9.2198 (9 × 10)159
Ref.23 7.9990 0.00112 99.61 33.46 8926.2 9.6253 2282

Table 9. Properties comparison between different algorithms.

shows the result of using a slightly different key than the original key, that is u0 is changed from 0.7990 to
0.799000000000000001. Encoding of picture with original and altered key are depicted by Fig. 9a–c. These
findings demonstrates that the presented scheme is very responsive to secret keys because a minor change in any
key component has an immense impact on encryption results.

Total keys analyzed

A cryptosystem’s key space is regarded as a fundamental component. For ensuring the high resistance against
brute force attack, key needs to have enough size. An extensive range of key can strengthen defense against a
brute force assault and keep the attacker from obtaining the right keys. Every key used in the cryptosystem
is included in the key range, which normally have a value of 2128. The key k = (k1, k2, k3) are employed as a
private key in suggested scheme. The parameters (a, b, p, nA, nB , G) are utilized by elliptic curve Diffie Hellman
key exchange as a secret key k2 for substitution process. Chaotic maps utilized two keys as a starting value and
controlling variable. The initial seed u0 and control parameter r are employed as a secret key k2 with Cubic-tent
chaotic map and k3 = (x0, y0, z0, a1, a2, a3) for the 3D-CSCM map. Observing that the parameters accuracy
is 10−15. For the parameters of the chaotic map, we use a precision level of 10−15, for the suggested scheme
the selecting keys might be (1015)8 = 10120 approximately equal to 2398 will be the key space size. To meet the
condition gcd(t, 256) = 1 has total possibilities of 27. Hence the total key space 2398 · 27 = 2405greater than 2128
. This size drastically increases the vulnerability against brute-force attacks and ensures the security of sensitive
image data. Table 7 demonstrates the contrast among different image encryption schemes.

Computational time assessment

Fast speed and great security performance are also necessary for an effective encryption technique. An essential
factor is the encryption algorithm’s speed of operation. By employing 256 × 256 image in suggested scheme we
measured the encoding/decoding time span using PC with Intel Core i7-4700 at 2.4 GHz CPU with 4 cores, 16
GB RAM and Windows 10 Pro 64. The suggested scheme takes 0.0210 seconds on average to process the Baboon
(256 x 256) picture. Table 8 depicts the comparison of time consumption of proposed scheme with other existing
algorithms.

Comparison
Suggested scheme and different literature presented image encryption methods comparison highlighted by Table
9. It also demonstrates the better performance of suggested algorithm.

Conclusion
We suggested a unique chaotic map-based picture encryption method in the proposed research work. The
proposed scheme comprises on two different stages name as confusion and diffusion. The suggested scheme begins
by generating a permutation phase using a compound chaotic map name as Cubic-tent map. Subsequentially,
substitution phase is done by employing Hill cipher, and its key is produced by elliptic curve based on public key.
In the last, the diffusion phase utilized the bitwise XORed operation with the generated random sequence values,
which is obtained by using a 3D-CSCM chaotic map.

Scientific Reports | (2024) 14:29626 | https://doi.org/10.1038/s41598-024-77955-w 13

www.nature.com/scientificreports/

Present scheme resists many different cryptographic attacks such as distortion, frequency based attacks, brute
force attack, correlation etc have been resisted by proposed encryption algorithm. Also, it exhibits versatility to
deal with different format of images. The security assessment demonstrates the viability and efficiency of the
suggested plan. Because the suggested encryption method is based on several chaotic maps, it could need a lot
of memory, which would raise the computational cost. In the future, the methods for digital image enciphering
and deciphering algorithm can be enhanced by current advancement in encryption algorithms.
As a future work, a combination of chaotic maps and generalized Arnold transformations with the proposed
image encryption can be explored further, particularly in enhancing image watermarking techniques. These
approaches may offer a potential for increased security and robustness in protecting digital image data. Also
our encryption technique could involve the integration of blockchain technology to strengthen security further.
Blockchain’s influence in areas such as secure digital transactions and decentralized data management highlights
its potential to develop more robust and tamper-resistant encryption methods49,50. Utilizing blockchain-based
decentralized key management systems, coupled with smart contracts for real-time access control and validation,
could greatly enhance the security of sensitive data.

Data availability
The datasets analyzed during the current study are available in the USC Signal and Image Processing Institute
(SIPI) Image Database. http://sipi.usc.edu/database.

Received: 9 July 2024; Accepted: 28 October 2024

References
1. Wang, X., Teng, L. & Qin, X. A novel colour image encryption algorithm based on chaos. Signal Process. 92(4), 1101–1108 (2012).
2. Daemen, J. & Rijmen, V. AES Proposal: Rijndael National Institute of Standards and Technology (2003).
3. Abduljabbar, Z. A. et al. Provably secure and fast color image encryption algorithm based on s-boxes and hyperchaotic map. IEEE
Access. 10, 26257–26270 (2022).
4. İhsan, A., & Doğan, N. An innovative image encryption algorithm enhanced with the Pan-Tompkins Algorithm for optimal
security. Multimedia Tools and Applications, 1–31 (2024).
5. Yang, T., Liu, Z., Guo, J., Yu, Y., Ren, F., & Wang, T. Image analysis by fractional-order weighted spherical Bessel-Fourier moments.
Pattern Recognit. 110872 (2024).
6. Liu, Y., Zhang, S., Li, G., Wang, H. & Yang, J. Accurate quaternion radial harmonic Fourier moments for color image reconstruction
and object recognition. Pattern Anal. Appl. 23, 1551–1567 (2020).
7. Nawaz, F., Inam, S., Kanwal, S., Al-Otaibi, S. & Hajjej, F. A resilient image encryption scheme using Laplace transform. Egypt.
Inform. J. 27, 100512 (2024).
8. Alvarez, G. & Li, S. Some basic cryptographic requirements for chaos-based cryptosystems. Int. J. Bifurcation Chaos 16(08), 2129–
2151 (2006).
9. Chai, X. et al. Color image compression and encryption scheme based on compressive sensing and double random encryption
strategy. Signal Process. 176, 107684 (2020).
10. Chai, X. et al. An efficient approach for encrypting double color images into a visually meaningful cipher image using 2D
compressive sensing. Inform. Sci. 556, 305–340 (2021).
11. Khan, M. et al. An efficient image encryption scheme based on double affine substitution box and chaotic system. Integration 81,
108–122 (2021).
12. Hua, Z. & Zhou, Y. Image encryption using 2D Logistic-adjusted-Sine map. Inform. Sci. 339, 237–253 (2016).
13. Yu, J., Xie, W., Zhong, Z. & Wang, H. Image encryption algorithm based on hyperchaotic system and a new DNA sequence
operation. Chaos Solitons Fractals 162, 112456 (2022).
14. Güvenoğlu, E. & Tunalı, V. ZigZag transform with Durstenfeld shuffle for fast and secure image encryption. Connection Sci. 35(1),
2162000 (2023).
15. Jirjees, S. W., Alkalid, F. F. & Shareef, W. F. Image encryption using dynamic image as a key based on multilayers of chaotic
permutation. Symmetry 15(2), 409 (2023).
16. Zahid, A. H., Arshad, M. J., Ahmad, M., Soliman, N. F. & El-Shafai, W. Dynamic S-Box Generation Using Novel Chaotic Map with
Nonlinearity Tweaking. Comput. Mater. Continua. 75(2) (2023).
17. Chu, R., Zhang, S. & Mou, J. A multi-image compression and encryption scheme based on fractional chaotic map. Physica Scripta
98(7), 075213 (2023).
18. Hadj Brahim, A., Ali Pacha, A. & Hadj Said, N. An image encryption scheme based on a modified AES algorithm by using a
variable S-box. J. Optics 53(2), 1170–1185 (2024).
19. Zhu, S., Wang, G. & Zhu, C. A secure and fast image encryption scheme based on double chaotic S-boxes. Entropy 21(8), 790
(2019).
20. Adeyefa, E. O., Akinola, L. S., & Agbolade, O. D. A new cryptographic scheme using the Chebyshev polynomials. in 2020
International Conference in Mathematics, Computer Engineering and Computer Science (ICMCECS) (pp. 1–3). IEEE (2020, March).
21. Chen, G., Mao, Y. & Chui, C. K. A symmetric image encryption scheme based on 3D chaotic cat maps. Chaos Solitons Fractals
21(3), 749–761 (2004).
22. Kanwal, S. et al. An effective color image encryption based on Henon map, tent chaotic map, and orthogonal matrices. Sensors.
22(12), 4359 (2022).
23. Kanwal, S. et al. A new image encryption technique based on sine map, chaotic tent map, and circulant matrices. Security Commun.
Netw. 2022(1), 4152683 (2022).
24. Long, Y., Peng, C., Tan, W., & Chen, Y. (2024). Blockchain-based anonymous authentication and key management for internet of
things with Chebyshev Chaotic Maps. IEEE Trans. Ind. Inform.
25. Chatterjee, S. et al. Secure biometric-based authentication scheme using Chebyshev chaotic map for multi-server environment.
IEEE Trans. Dependable Secure Comput. 15(5), 824–839 (2016).
26. Lee, T. F. & Huang, Y. C. Efficient extended Chaotic Map-based IBE for industrial environment. IEEE Access 10, 71278–71283
(2022).
27. Zhang, X. & Wang, X. Digital image encryption algorithm based on elliptic curve public cryptosystem. IEEE Access 6, 70025–70034
(2018).
28. Dawahdeh, Z. E., Yaakob, S. N. & bin Othman, R. R. A new image encryption technique combining Elliptic Curve Cryptosystem
with Hill Cipher. J. King Saud Univ.-Comput. Inform. Sci. 30(3), 349–355 (2018).

Scientific Reports | (2024) 14:29626 | https://doi.org/10.1038/s41598-024-77955-w 14

www.nature.com/scientificreports/

29. Abd El-Latif, A. A. & Niu, X. A hybrid chaotic system and cyclic elliptic curve for image encryption. AEU-Int. J. Electron. Commun.
67(2), 136–143 (2013).
30. Parida, P., Pradhan, C., Gao, X. Z., Roy, D. S. & Barik, R. K. Image encryption and authentication with elliptic curve cryptography
and multidimensional chaotic maps. IEEE Access 9, 76191–76204 (2021).
31. Abdelfatah, R. I. Secure image transmission using chaotic-enhanced elliptic curve cryptography. IEEE Access 8, 3875–3890 (2019).
32. Laiphrakpam, D. S. & Khumanthem, M. S. Medical image encryption based on improved ElGamal encryption technique. Optik
147, 88–102 (2017).
33. Reyad, O., Kotulski, Z. & Abd-Elhafiez, W. M. Image encryption using chaos-driven elliptic curve pseudo-random number
generators. Appl. Math. Inf. Sci. 10(4), 1283–1292 (2016).
34. Abbas, A. M., Alharbi, A. A. & Ibrahim, S. A novel parallelizable chaotic image encryption scheme based on elliptic curves. IEEE
Access 9, 54978–54991 (2021).
35. Li, C., Shen, X., & Liu, S. Cryptanalyzing an image encryption algorithm underpinned by 2D lag-complex logistic map. IEEE
MultiMedia. (2024).
36. Aouissaoui, I., Bakir, T., Sakly, A. & Femmam, S. Improved one-dimensional piecewise Chaotic Maps for information security. J.
Commun. 17(1), 11–16 (2022).
37. Zhang, Y., Xiang, H., Zhang, S. & Liu, L. Construction of high-dimensional cyclic symmetric chaotic map with one-dimensional
chaotic map and its security application. Multimed. Tools Appl. 82(12), 17715–17740 (2023).
38. USC Signal and Image Processing Institute (SIPI) Image Database. http://sipi.usc.edu/database/.
39. Wu, Y. et al. Lo cal Shannon entropy measure with statistical tests for image randomness. Inform. Sci. 222, 323–342 (2013).
40. Lone, P. N. et al. Cryptanalysis and improved image encryption scheme using elliptic curve and affine hill cipher. Mathematics
10(20), 3878 (2022).
41. Akraam, M., Rashid, T. & Zafar, S. A Chaos-based image encryption scheme is proposed using multiple chaotic maps. Math.
Problems Eng. 2023(1), 2003724 (2023).
42. Kanwal, S. et al. Analytic study of a novel color image encryption method based on the chaos system and color codes. Complexity
2021(1), 5499538 (2021).
43. Jiang, M. & Yang, H. Image encryption using a new hybrid chaotic map and spiral transformation. Entropy 25(11), 1516 (2023).
44. Yasser, I., Khalifa, F., Mohamed, M. A. & Samrah, A. S. A new image encryption scheme based on hybrid chaotic maps. Complexity
2020(1), 9597619 (2020).
45. Hasanzadeh, E. & Yaghoobi, M. A novel color image encryption algorithm based on substitution box and hyper-chaotic system
with fractal keys. Multimed. Tools Appl. 79, 7279–7297 (2020).
46. Patro, K. A. K., Acharya, B. & Nath, V. Various dimensional colour image encryption based on non-overlapping block-level
diffusion operation. Microsyst. Technol. 26(5), 1437–1448 (2020).
47. Zhu, S., Deng, X., Zhang, W. & Zhu, C. Image encryption scheme based on newly designed chaotic map and parallel DNA coding.
Mathematics 11(1), 231 (2023).
48. Güvenoğlu, E. An image encryption algorithm based on multi-layered chaotic maps and its security analysis. Connection Sci. 36(1),
2312108 (2024).
49. Inam, S., Kanwal, S., Firdous, R. & Hajjej, F. Blockchain based medical image encryption using Arnold’s cat map in a cloud
environment. Sci. Rep. 14(1), 5678 (2024).
50. Kanwal, S., Inam, S., Quddus, S. & Hajjej, F. Research on color image encryption approach based on chaotic Duffing map. Physica
Scripta 98(12), 125252 (2023).

Acknowledgements
Princess Nourah bint Abdulrahman University Researchers Supporting Project number (PNURSP2024R136),
Princess Nourah bint Abdulrahman University, Riyadh, Saudi Arabia.

Competing interests
The authors declare no competing interests.

Additional information
Correspondence and requests for materials should be addressed to S.K.
Reprints and permissions information is available at www.nature.com/reprints.
Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives
4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in
any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide
a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have
permission under this licence to share adapted material derived from this article or parts of it. The images or
other third party material in this article are included in the article’s Creative Commons licence, unless indicated
otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence
and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to
obtain permission directly from the copyright holder. To view a copy of this licence, visit ​h​t​t​p​:​/​/​c​r​ea​ ​t​i​v​e​c​o​m​m​o​
n​s​.​o​r​g​/​l​i​ce​ ​n​s​e​s​/​b​y​-​n​c​-​n​d​/​4​.​0​/​​​​.​ ​​

© The Author(s) 2024

Scientific Reports | (2024) 14:29626 | https://doi.org/10.1038/s41598-024-77955-w 15