Multimedia Systems (2024) 30:247

https://doi.org/10.1007/s00530-024-01449-y

REGULAR PAPER

Integrate encryption of multiple images based on a new hyperchaotic

system and Baker map
Xingbin Liu1

Received: 15 June 2024 / Accepted: 6 August 2024

© The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2024

Abstract
Image encryption serves as a crucial means to safeguard information against unauthorized access during both transmission
and storage phases. This paper introduces an integrated encryption algorithm tailored for multiple images, leveraging a novel
hyperchaotic system and the Baker map to augment the key space and enhance security measures. The methodology encom-
passes a permutation-diffusion framework, employing sequences derived from the hyperchaotic system for both permuta-
tion and diffusion operations. Initially, the multiple images undergo intermixing, consolidating them into a singular image.
Subsequently, the Baker map is employed to further scramble this amalgamated image, thereby extending the scrambling
period. Ultimately, the ciphertext image is generated through forward–backward diffusion applied to the pixel sequence of
the Zigzag scanned image. Experimental findings substantiate the high-security efficacy of the proposed scheme, demon-
strating resilience against diverse threats.

Keywords Multiple image encryption · Baker map · Hyperchaotic system · Zigzag scan · Security analysis

1 Introduction promising encryption outcomes. For instance, methodolo-

gies such as wavelength multiplexing-based encryption [17],
With the rapid evolution of the Internet and multimedia position multiplexing-based encryption [18], interference-
technologies, image data offers a wealth of visual experi- based encryption [19], and diffusion of odd–even interleaved
ences. However, this proliferation of image data also poses points [20] have demonstrated effective multiple-image
significant security challenges during transmission and stor- encryption capabilities. Perez proposed a nonlinear encryp-
age [1–3]. Consequently, there is a pressing need for robust tion approach for multiple images, utilizing a joint trans-
image content protection, leading to extensive research in form correlator and the Gyrator transform [21], while Sabir
the field of image encryption. Over recent years, numerous presented a multiple image encryption algorithm based on
image encryption algorithms have been developed, drawing an affine Hill cipher, discrete fractional Hartley transform,
upon various methodologies such as optical devices [4, 5], and generalized Arnold map [22]. Additionally, a multiple-
chaotic systems [6–8], compressed sensing [9, 10], quantum image encryption algorithm employing a scrambling and dif-
principles [11–13], and DNA coding [14–16]. fusion structure based on defined super pixels and bit planes
Most of the existing methodologies primarily focus on was introduced in [23]. Su proposed an optical encryption
encrypting single images, yet the encryption efficiency method utilizing a complete trinary tree structure to encrypt
could be notably enhanced by extending the encryption multiple color images, employing phase truncated Fresnel
framework to multiple images. Various endeavors have transform to obtain real-value images [24]. Gao designed a
been made towards multiple image encryption, yielding multiple images encryption algorithm based on single-chan-
nel scrambling and chaotic systems [25]. These endeavors
have effectively realized multiple image encryption. How-
Communicated by Yongdong Zhang.
ever, the images used in these approaches are not scrambled
* Xingbin Liu with each other prior to encryption. To further fortify the
[email protected] security of image encryption algorithms, this paper proposes
a mutual scrambling method.
1
College of Electronic and Information Engineering,
Southwest University, Chongqing 400715, China

Vol.:(0123456789)

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

247 Page 2 of 13 X. Liu

Digital chaotic dynamic systems play a pivotal role filter and discrete logarithmic transform, an image encryp-
in image encryption due to their properties conducive to tion scheme is proposed [46]. Some new chaotic with good
encryption, including sensitivity to initial parameters and characteristics are consistently proposed, such as determin-
control parameters, pseudo-randomness, aperiodicity, long- istic chaotic finite-state automata [47], hyperchaotic system
term unpredictability, fractal characteristics, and universal- based on optimization benchmark function [48]. These
ity. Numerous image encryption methodologies have been newly proposed chaotic systems exhibit desirable proper-
developed based on existing and newly devised chaotic sys- ties such as ergodicity and unpredictability, enhancing the
tems. For instance, Erkan et al. proposed a chaotic system security of image encryption algorithms while facilitating
based on the Schaffer function and employed it in image convenient key distribution. Besides the encryption algo-
encryption [26], while Lai et al. designed a 2D Salomon rithms, some cryptanalysis of image encryption is also rais-
map to alter the high and low bits of pixels, thereby pro- ing some concerns. Abba et al. cryptanalyses key space of
ducing a meaningless image [27]. Cavusoglu developed a some chaos-based algorithms [49]. Wen et al. provided sev-
novel chaotic system and incorporated it into an encryp- eral methods to crack image encryption schemes based on
tion algorithm based on an S-Box, ensuring security and chaotic map and DNA coding [50, 51]. Feng et al. cracked
rapid execution [28]. Al-Hazaimeh et al. proposed an image an image encryption method based on Feistel network and
encryption algorithm utilizing the Lorenz chaotic map, dynamic DNA encoding [52].
with initial secret keys generated from the hash value of the Most existing image encryption methods adhere to a
plaintext image [29]. Wang et al. introduced several new permutation-diffusion architecture [53], as proposed by
chaotic systems and integrated them into image encryption Fridrich. In the permutation stage, techniques such as the
frameworks, including chaotic signals with finite-precision Arnold transform [54], Hilbert transform [55], and other
error [30], an improved ant colony walking path [31], and scrambling methods are commonly employed. The Arnold
a newly-designed coupled map lattice [32], among others. transform, in particular, is frequently utilized due to its ease
Alexan et al. proposed several image encryption methods of implementation [56, 57]. However, the periodic nature
based on chaotic maps, such as a color image cryptosystem of the Arnold transform, with the period dependent on the
based on Sine chaotic map and 4D Chen hyperchaotic map image size, poses potential risks, especially with smaller
of fractional-order [33]. In addition, based on hyperchaotic periods. To mitigate this risk, integrating multiple images
Chen system and variable-base modulo operation, a three- is proposed to expand the key space. Additionally, while the
stage image encryption algorithm is proposed [34]. By com- Baker map is also periodic, it offers superior characteristics
bining the Chen and Chua chaotic functions, pseudo-random compared to the Arnold transform, boasting a larger param-
numbers are generated and used for image encryption [35]. eter space for images of the same size. Consequently, the
Three operations including rescale, rotate, and randomize proposed integrated image encryption algorithm is designed
are employed in an image cryptosystem utilizing chaotic utilizing the Baker map.
maps [36]. A five-stage image encryption method adopting This paper introduces an integrated image encryption
confusion-diffusion structure with Langton’s Ant is proposed algorithm based on a novel hyperchaotic system and the
by Alexan [37]. Moreover, the chaotic maps have been used Baker map. The key innovation lies in integrating multiple
for secure communication of military reconnaissance images images and enabling mutual scrambling between any two
[38]. Laiphrakpam et al. proposed a multiple-image encryp- sub-images. Chaotic system-generated sequences are uti-
tion method utilizing an enhanced chaotic map and achieves lized to scramble pixel positions within the deeply integrated
secure and efficient results [39]. Mehmood et al. designed a image, thereby enlarging its size and consequently extending
time-efficient and noise-resistant cryptosystem with chaotic the period of the Baker map. During the diffusion process,
map, which has a huge key space [40]. By combing the DNA pixel values are altered through forward and backward diffu-
tree and a chaotic state machine map, different sizes and sion. The order of pixel diffusion is determined using Zigzag
types images can be effectively encrypted [41]. To ensure scanning, ensuring uniform spread of pixel energy. To thwart
timely and secure data transfer between sensors and edge chosen-plaintext attacks, initial values and control param-
devices, a perturbed logistic map is introduced and used eters are derived from the plaintext image. Experimental
for image encryption [42]. Kocak et al. proposed an image results and analysis affirm the validity and security of the
encryption algorithm utilizing particle swarm optimization proposed scheme against numerous attacks.
and a modular integrated logistic exponential map [43]. In comparison to existing works, this paper presents sev-
Feng et al. realized a multiple image encryption scheme eral key contributions:
based on designed fractional-order 3D Lorenz chaotic map
[44]. Qian et al. proposed a multi-channel image encryp- 1. Introduction of an integrated encryption method for mul-
tion algorithm using newly designed memristor-enhanced tiple images based on the Baker map, enhancing both
polynomial hyper-chaotic map [45]. Using plane-level image security and efficiency.

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Integrate encryption of multiple images based on a new hyperchaotic system and Baker map Page 3 of 13 247

2. Introduction of a mutual permutation method for multi- denoted as n1 + n2 + ⋯ + ni = Ni and N0 = 0. The discrete
ple images utilizing a hyperchaotic system, enabling full Baker map can be expressed as follows.
permutation of pixel positions within original images. ( ) ( ( ) ( ))
3. Adoption of forward and backward diffusion processes x qi u − Ni−1 + mod v, qi
= ( ) (2)
in the encryption scheme, accompanied by the introduc- y v − mod v, qi ∕qi + Ni − 1
tion of Zigzag scanning to shuffle pixel order, thereby /
enhancing overall security. where qi = N ni. The original pixel at position (x, y) is trans-
formed to a new pixel position (x, y) using Baker map and
The remainder of this article is structured as follows: thus the image is fully permutated.
Sect. 2 provides a concise overview of the fundamental
theories behind hyperchaotic systems and the Baker map. 2.2 Hyperchaotic system
Section 3 elaborates on the proposed integrated multiple
image encryption algorithm. Simulation results and security Hyperchaotic systems exhibit a more intricate dynamic
analysis are presented in Sect. 4. Finally, Sect. 5 concludes behavior compared to one-dimensional chaos, as they pos-
the paper and outlines potential avenues for future research. sess two or more Lyapunov exponents and phase trajectories
that diverge in multiple directions. This complexity renders
them particularly well-suited for secret communication and
2 Basic theories image encryption purposes. Li et al. introduced a novel
four-dimensional hyperchaotic system [59], represented as
2.1 Baker map follows:
� �
The Baker map, akin to the Arnold transform, serves as a ⎧ ẋ = 𝛼 x2 − x1 + k1 x4
tool for image permutation [58]. In contrast to the Arnold ⎪ 1
⎪ ẋ 2 = 𝜒x1 − x1 x3 + k2 x4
transform, the Baker map offers an extended period, thereby ⎨ (3)
⎪ ẋ 3 = −𝛽x3 + x1 x2 + k3 x4
enlarging the key space. The Baker map effectuates the
⎪ ẋ
transformation of pixel positions by stretching and folding ⎩ 4 = −𝛿x1
specific areas. The formal definition of the Baker map is
provided in Eq. (1), and the transformation process is visu- where the parameters are set to 𝛼 = 35, 𝛽 = 3, 𝜒 = 35, 𝛿 = 5,
ally depicted in Fig. 1. k1 = 1, k2 = 0.2, k3 = 0.3, the system presents hyperchaotic
� � behavior. The simulation results of chaotic attractors are
⎧ 2x, y 0≤x<
1 depicted in Fig. 2. It is evident from both the 2D and 3D
⎪� 2 � 2 perspectives that the utilized chaotic system exhibits a robust
B(x, y) = ⎨ y+1 (1)
⎪ 2x − 1,
1
≤x<1 chaotic state.
⎩ 2 2

For an image with size of N × N , according to the

3 Proposed integrate image cryptosystem
theory of Baker map, it can be divided into p rectangle
The proposed image cryptosystem adopts a permutation-
blocks.
{ The width } values of blocks can be represented
diffusion architecture. The permutation process consists of
as n1 , n2 , ... np , and the summation of width is N , i.e.,
two procedures: mutual permutation and the Baker map.
n1 + n2 + ⋯ + np = N . The summation of first i width is
Forward and backward diffusion techniques are employed
in the diffusion process to modify pixel values. A flow dia-
gram illustrating the integrated multiple image encryption
scheme is depicted in Fig. 3.
1 1
3.1 Image integration and mutual permutation

0.5 The image integration process is straightforward and easily

implementable, involving the merging of multiple images
into a single composite image. For illustrative purposes,
0 0.5 1 0 1 four grayscale images are utilized as examples. The diagram
depicting the image integration and permutation is presented
Fig. 1  The diagram of Baker mapping process in Fig. 4. The four images are merged directly, following

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

247 Page 4 of 13 X. Liu

Fig. 2  The chaotic phase

diagrams of new 4-demensional
hyperchaotic system

Diffusion
Integration Mutual Baker map
permutation

Original images Integrated image permutated image Further permutated image Final encrypted image

Fig. 3  The flow diagram regarding the integrate multiple image encryption scheme

( )
which our attention is directed towards the process of mutual contains 16 bits ki ki1 , ki2 , ..., ki16 . The ki is converted to deci-
permutation. mal number by
The mutual permutation operation is introduced to thor-
oughly scramble the pixels of multiple images. Initially, four ∑
16

pseudo-random sequences are generated using a novel four-

ki = kiv × 2v−1 (4)
v=1
dimensional hyperchaotic system. To counter chosen-plain-
text attacks, the initial values of the hyperchaotic system Then the initial values of 4-demensional hyperchaotic sys-
must be correlated with the plaintext image. Consequently, tem are obtained as follows.
the SHA-256 algorithm is employed to generate the initial
values of the four-dimensional hyperchaotic system.
The resulting composite image is considered as an input to
compute the SHA-256 value. There are 256 bits in the result
and it is divided into 16 groups k1 , k2 , … , k16 and each group

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Integrate encryption of multiple images based on a new hyperchaotic system and Baker map Page 5 of 13 247

Fig. 4  The diagram of image

integration and permutation

Mutual
permutation
Synthesis

Composite image Permutated image

Plaintext images

�� � � � � Table 1  The mutual permutation operation according to the Sn value

⎧ k1 + k9 ⊕ k5 + k13
⎪ x1(0) = mod ,1 Image Sn=1 Sn= 2 Sn= 3 Sn= 4
⎪ 256
⎪ �� � � � � 1st sub-image 1 ⇄ 1 inner 1 ⇄ 2 outer 1 ⇄ 3 outer 1 ⇄ 4 outer
⎪ (0) k2 + k10 ⊕ k6 + k14
⎪ x2 = mod ,1 2nd sub-image 2 ⇄ 1 outer 2 ⇄ 2 inner 2 ⇄ 3 outer 2 ⇄ 4 outer
⎪ 256
3rd sub-image 3 ⇄ 1 outer 3 ⇄ 2 outer 3 ⇄ 3 inner 3 ⇄ 4 outer
⎨ �� � � � � (5)
⎪ (0) k3 + k11 ⊕ k7 + k15 4th sub-image 4 ⇄ 1 outer 4 ⇄ 2 outer 4 ⇄ 3 outer 4 ⇄ 4 inner
⎪ x3 = mod ,1
⎪ 256
⎪ �� � � � �
⎪ x(0) k4 + k12 ⊕ k8 + k16 otherwise performs outer permutation. The inner permutation
⎪ 4 = mod ,1
⎩ 256 and outer permutation for 1st sub-image is illustrated in Fig. 5.
{ The
} exchange
{ } position is determined
{ }with chaotic sequences
where x1(0) , x2(0) , x3(0) , x4(0) represent the four initial values X2 and X3{ , and
} the sequence X2 determines the X-axis
for 4-demensional position and X3 determines the Y-axis position.
{ hyperchaotic
} { } { }system. { } The generated cha-
otic sequences X1 , X2 , X3 , X4 are used for subse- { {⌊ ⌋ }
quent mutual permutation and forward–backward diffusion Epx = X2 × 1015 mod 256 + 1
{⌊ ⌋ } (7)
processes. Epy = X3 × 1015 mod 256 + 1
The mutual permutation can be classified to inner permu-
tation and{outer where Epx and Epy denote the positions to be exchanged
} permutation through the value of the chaotic
sequence X1 . in X-axis and Y-axis, respectively. That’s to say, the pixel
{⌊ ⌋ } at (x, y) is exchanged with the pixel at position (Epx, Epy).
Sn = X1 × 1015 mod 4 + 1 (6) For 4 sub-images, there are 4 × 4 permutation operations
are performed. After mutual permutation, the 4 sub-images
where Sn represents the selected sub-image serial num-
are fully scrambled and a large-size permutated image Ip is
ber, the symbol ⌊⋅⌋ denotes the round down function. For
obtained.
the 1st sub-image, if the Sn = 1, the inner permutation will
be executed. In other cases, the outer permutation will be
3.2 Baker map permutation
executed. The same procedures will be performed on the
other 3 images. The specific mutual permutation according
The preliminary permutated image Ip is further scrambled
to the Sn value is shown in Table 1. The symbol ⇄ means the
using Baker map (BM). As the size of Ip is enlarged, the space
permutation process between two sub images.
of parameters of Baker map will be increased.
It can be seen form Table 1, if the Sn is equal to the order of
( )
the sub-image, the algorithm performs inner permutation and Ib = BMT Ip (8)

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

247 Page 6 of 13 X. Liu

Fig. 5  The inner and outer permutation for the 1st sub-image

The forward and backward{ } diffusion processes are com-

pleted with the sequence X4 . Firstly, two random matrices
1 2 3 4 1 2 5 9
R1 and R2 are obtained as follows.
{⌊ ⌋ }
5 6 7 8 6 3 4 7 R1 = X4 × 1015 mod 256 1
{⌊ ⌋ } (10)
9 10 11 12 10 13 14 11
R2 = X4 × 1015 mod 256 2
{ }
The front part and latter part of X4 are used to generate
13 14 15 16 8 12 15 16
random matrix R1 and R2 , respectively.
Then, the forward diffusion is implemented with R1 and
Fig. 6  The schematic diagram of Zigzag scan the diffusion result Id is obtained.
� �
⎧ I (1, 1) = I (1, 1)+R (1, 1) mod 256
⎪ d
� z 1
�
⎪ Id (1, j) = Iz (1, j)+R1 (1, j) + Id (1, j − 1) mod 256
where T represent scrambling period of Baker map. ⎨ � �
⎪ Id (i, 1) = � Iz (i, 1)+R1 (i, 1) + Id (i − 1, j) mod 256�
⎪ I (i, j) = I (i, j)+R (i, j) + I (i, j − 1)+I (i − 1, j) mod 256
⎩ d z 1 d d
3.3 Forward and backward diffusion (11)
After forward diffusion, the intermediate result is
The diffusion process employs both forward and back- obtained. Suppose the size of the integrated image is M × N ,
ward diffusion techniques to alter pixel values. Prior to the backward diffusion can be described as follows.
the application of forward and backward diffusion, a Zig-
zag scan (ZZS) is utilized to further disorganize the posi- � �
⎧ I (M, N) = I (M, N)+R (M, N) mod 256
tions of pixels, enhancing the encryption effectiveness. ⎪ e � d 2
�
⎪ Ie (M, j) = Id (M, j)+R2 (M, j) + Ie (M, j + 1) mod 256
The schematic diagram of the Zigzag scan is depicted in ⎨ � �
Fig. 6. As illustrated in the diagram, the position order ⎪ Ie (i, N) =� Id (i, N)+R2 (i, N) + Ie (i + 1, N) mod 256�
undergoes a change following the Zigzag scan. ⎪ I (i, j) = I (i, j)+R (i, j) + I (i, j + 1)+I (i + 1, j) mod 256
⎩ e d 2 e e
( ) (12)
Iz = ZZS Ib (9)
The final encrypted image Ie is obtained after forward
and backward diffusion. The decryption algorithm precisely

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Integrate encryption of multiple images based on a new hyperchaotic system and Baker map Page 7 of 13 247

mirrors the encryption process. By applying the inverse demonstrates a significant encryption effect. To comprehen-
operations of backward diffusion, forward diffusion, the sively assess the performance of the proposed algorithm,
Baker map, and mutual permutation, the original images security analysis is conducted encompassing statistical
can be effectively recovered. attack analysis, correlation analysis, information entropy,
The proposed algorithm is also developed with inde- differential attack, key sensitivity analysis, and other aspects.
pendent secret keys using RSA algorithm. The RSA algo- The security analysis confirms that the proposed scheme
rithm is based on the mathematical problem of factoring exhibits robustness against various attacks.
large integers and is widely used in asymmetric crypto-
graphic systems. Firstly, two large random prime num- 4.1 Distribution of pixel values
bers p and q are chosen, and the product values p × q and
𝜑 = (p − 1) × (q − 1) are obtained. Then choose a value e To resist statistical inference from encrypted images, it is
and satisfy that 1 < e < 𝜑. Finally, the value is d calculated expected that the distribution of pixel values to be uniform.
using e × d = 1mod(𝜑). The public key is (n, e) and the pri- Histogram analysis serves as a visualization tool to assess
vate key is (n, d). For a message M , the encryption process the uniformity of distribution. The histograms of origi-
can be expressed as:C = M e (modn) . The corresponding nal images and their corresponding ciphertext images are
decryption process is M = Cd (modn). depicted in Fig. 8a and b respectively. It is evident from
The keys include the initial values x1(0) , x2(0) , x3(0) , x4(0) of the figures that the proposed algorithm effectively achieves
chaotic system and period of Baker map T . Suppose these sufficient randomization to withstand statistical inferences.
independent parameters are encrypted using public key (n, e) To quantitatively assess the uniformity performance,
and sent to recipient. The receiver can recover the origi- we compute the histogram variance defined as follows and
nal parameters and generate corresponding pseudo-random present the corresponding values in Table 2. As observed
sequences using private key (n, d). Therefore, the plaintext from the table, the histogram variance undergoes a signifi-
images can be exactly restored. Based on NP-hard problem cant reduction, indicating a more uniform pixel frequency
of RSA, the keys are securely shared. distribution.

1 ∑ ∑ 1(
255 255
)2
var(H) = grayi − grayj (13)
4 Simulation results and analysis 256 × 256 i=0 j=0 2

The simulation platform utilized for the proposed image where grayi and grayj denote number of pixels with value
encryption system is Matlab R2018a, and the test images equal to i and j , respectively. The results of compared meth-
are sourced from the Computer Vision Group-University of ods are also listed in Table 2, it can be seen that the values
Granada (CVG-UGR) database [60] and the University of of the proposed method are less than the compared method.
Southern California-Signal and Image Processing Institute Therefore, the value distribution of the proposed method is
(USC-SIPI) database [61]. Four test images are depicted more uniform.
in Fig. 7a, while the final encrypted image is shown in
Fig. 7b, where it is evident that no useful information can 4.2 Information entropy analysis
be discerned from the encrypted image. With the correct
keys, the original images are entirely recovered without Information entropy serves as a metric to gauge the disorder-
errors, as depicted in Fig. 7c. Thus, the proposed algorithm liness and unpredictability inherent in information. Defined

Fig. 7  Encryption and decryp-

tion results of test images. a
Four input original images, b
the encrypted image, c corre-
sponding decrypted images

Lake Bridge

Camera Elaine
(a) (b) (c)

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

247 Page 8 of 13 X. Liu

by information theory, the information entropy is expressed comparison results are also listed in Table 3, it can be seen
as follows: that the entropy values of the proposed method are much
closer to 8. Therefore, the proposed method has a more ran-
∑ ( ) p(Ii )
N−1
dom output.
H(I) = − p Ii log2 (14)
i=0

where the symbol p denotes the probability of random varia- 4.3 Correlation of adjacent pixels analysis
ble Ii and the total number of possible states is N . Hence, the
information entropy attains a value of 8 for an ideally fully Typically, natural images exhibit a strong correlation
random image comprising 256 Gy levels. The information between adjacent pixels. The correlation between adjacent
entropy values of plaintext images and their corresponding pixels is employed to ascertain whether the proposed algo-
ciphertext images are enumerated in Table 3. It is evident rithm can mitigate this correlation. The correlation coef-
from the table that the entropy values of ciphertext images ficient between two adjacent pixels is defined as follows:
closely approximate 8, indicating that the proposed algo-
rithm yields highly random ciphertext images. The entropy

Fig. 8  The histograms of original images and corresponding ciphertext images

Table 2  The histogram Image Size Histogram variance

variance of plaintext images
and corresponding ciphertext Plaintext image Ciphertext Ref. [30] Ref. [56]
images image

Lake 256 × 256 44,651 238 312 351

Bridge 256 × 256 2,95,850 235 289 289
Camera 256 × 256 1,10,973 226 296 312
Elaine 256 × 256 35,302 273 352 363
Integrated image 512 × 512 5,63,978 939 1206 1108

Bold values indicate the best results

Table 3  The information Image Size Information entropy

entropy of plaintext images
and corresponding ciphertext Plaintext image Ciphertext image Ref. [30] Ref. [56]
images
Lake 256 × 256 7.4898 7.9974 7.9965 7.9942
Bridge 256 × 256 5.7062 7.9970 7.9968 7.9941
Camera 256 × 256 7.0097 7.9975 7.9966 7.9938
Elaine 256 × 256 7.5046 7.9974 7.9969 7.9940
Integrated image 512 × 512 7.6020 7.9994 7.9972 7.9945

Bold values indicate the best results

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Integrate encryption of multiple images based on a new hyperchaotic system and Baker map Page 9 of 13 247

∑N � �� �
cov(x, y) i=1
xi − x̄ yi − ȳ
CCx,y =√ =�
D(x)D(y) ∑N � �2 ∑N � �2
x − x̄
i=1
y − ȳ
i i=1 i
(15)
where x and y represent the mean value of adjacent pixels
x and y , respectively. The total number of selected samples
is denoted as N . To quantitatively evaluate the CC value
of adjacent pixels of original image and ciphertext image,
10,000 adjacent pixel pairs of images in each direction are
randomly selected to calculate. The correlation coefficients
in three directions—horizontal, vertical, and diagonal—
are computed and presented in Table 4. It is evident from
the table that the correlation coefficients of the ciphertext
images are close to zero, indicating a significant reduction
in the strong correlation observed in the original images.
For a more illustrative representation, the sampled pairs are
plotted in Fig. 9, revealing minimal correlation in the three
directions of the ciphertext images. These correlation coef-
ficient results substantiate the robustness of the proposed
algorithm against statistical attacks.

4.4 Spectrum analysis
Fig. 9  The correlation scatter diagrams of original images and corre-
To visually illustrate the statistical characteristics, the sponding ciphertext images
spectra of both original and ciphertext images are plotted
in Fig. 10. It is evident from the figure that the spectrum
becomes uniform post-encryption. Consequently, the pro-
4.5 Differential attack analysis
posed algorithm effectively disperses information, render-
Differential attacks are categorized under chosen-plaintext
ing statistical analysis ineffective for attackers in extracting
attacks, wherein adversaries aim to compromise the crypto-
useful information.
system by scrutinizing the relationship between ciphertext and
plaintext. Consequently, an effective image encryption scheme
should exhibit sensitivity to variations in plaintext images. In
Table 4  The correlation coefficient of plaintext images and corre- this study, the SHA-256 algorithm is employed on the plain-
sponding ciphertext images text image to generate keys, resulting in completely distinct
Image Directions Plaintext image Ciphertext image ciphertext images. To assess the scheme’s resilience against
differential attacks, two metrics, namely the number of pixel
Lake Horizontal 0.9260 0.0200
change rate (NPCR) and the unified average changing intensity
Vertical 0.9403 0.0226
(UACI), are utilized. NPCR quantifies the proportion of pixel
Diagonal 0.8947 0.0274
changes in the ciphertext image resulting from minor altera-
Bridge Horizontal 0.8749 −0.0081
tions in the plaintext image, while UACI measures the degree
Vertical 0.8916 −0.0224
of disparity among the generated ciphertext images.
Diagonal 0.8331 0.0005
Camera Horizontal 0.9797 −0.0598
∑W ∑H
D(x, y)
Vertical 0.9762 0.0055 NPCR =
x=1 y=1
× 100% (16)
W ×H
Diagonal 0.9553 0.0754
Elaine Horizontal 0.9622 −0.0211 where W and H denote the width and height of image,
Vertical 0.9556 −0.0158 respectively. If the pixel value is different at same position,
Diagonal 0.9085 0.0083 the D(x, y) = 1. Otherwise, the D(x, y) is set to 0.
Integrated image Horizontal 0.9204 −0.0358 ∑W ∑H � �
Vertical 0.9359 −0.0078 y=1 � 1 (x, y) − C2 (x, y)�
C
UACI =
x=1
× 100% (17)
Diagonal 0.8948 0.0027 W × H × 2L−1
Bold values indicate the best results

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 247 Page 10 of 13 X. Liu

Fig. 10  The spectrums of original and encrypted images

Table 6  The results of NPCR and UACI for independent secret key
Table 5  The modification of pixel and corresponding results of
NPCR and UACI Image Pixel modification NPCR UACI
Image Pixel modification Proposed Ref. [56] Lake (62,113)212 → 211 99.3726 33.5762
NPCR UACI NPCR UACI Bridge (222,108)93 → 94 99.3852 33.3642
Camera (115,78)17 → 16 99.1563 33.3579
Lake (62,113)212 → 211 99.4892 33.4865 99.3625 33.2621
Elaine (125,228)94 → 95 99.2963 33.3028
Bridge (222,108)93 → 94 99.3732 33.4761 99.2568 33.2532
Average – 99.3026 33.4003
Camera (115,78)17 → 16 99.0124 33.4577 99.0226 33.3101
Elaine (125,228)94 → 95 99.4343 33.5459 99.3556 33.2872 Bold values indicate the best results
Average – 99.3273 33.4916 99.2494 33.2782

Bold values indicate the best results 4.6 Key sensitivity and key space analysis

A robust cryptosystem should exhibit sensitivity to its

where C1 and C2 denote the generated ciphertext images with keys; hence, key sensitivity is assessed. Within a set of
one pixel different of plaintext image. The symbol L denotes keys, a minor deviation in one key is employed to decrypt
the gray level. the ciphertext image while keeping the remaining keys
Assuming there exists only one random bit difference unchanged. The keys employed in the system comprise
between two plaintext images, with the randomly selected parameters and initial values of the hyperchaotic system,
pixel positions listed in Table 5. The corresponding NPCR along with the period of the Baker map. Each initial value
and UACI results are also computed. The results dem- x1(0) , x2(0) , x3(0) , x4(0) are added 10−14 deviation respectively
onstrate that both metrics are close to the ideal values and the decryption results are shown in Fig. 11a–d. The
(NPCR = 99.609%, UACI = 33.464%), suggesting that the period of Baker map is changed to T + 1 and correspond-
proposed algorithm exhibits robustness against differential ing decryption result is shown in Fig. 10e. From the figure,
attacks. The UACI and NPCR comparison results are also it is evident that the decryption results yield a noise-like
listed in Table 5, from which can be seen that the proposed image even with minor variations in the keys. This obser-
method has obvious superiority. In addition, the NPCR vation highlights the sensitivity of the proposed algorithm
and UACI are tested for the independent secret key, and to key changes. Based on the sensitivity analysis, it can
the corresponding results are shown in Table 6. It can be be inferred that ( the)4key space of the proposed algorithm
seen that the numerical values are consistent with Table 5. is larger than 1014 > 2100 . Hence, the proposed method
Therefore, the keys transmitted with public key system can demonstrates resilience against brute-force attacks.
also resist against differential attacks.

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Integrate encryption of multiple images based on a new hyperchaotic system and Baker map Page 11 of 13 247

Fig. 11  The deviation keys and

corresponding decryption result

(a) x1
0
10 14
0
(b) x2 10 14 (c) x3
0
10 14
0
(d) x4 10 14 (e) T 1

Fig. 12  The noise attack and

corresponding decrypted images

(a) s 1 (b) s 5 (c) s 10 (d) s 15

The occlusion attack is further evaluated through differ-

ent proportion cropping. The resulting cropped ciphertext
images and their corresponding decrypted counterparts are
depicted in Fig. 13. Despite some blurring in the restored
images, the primary information remains discernible, even
with a 50% occlusion. The visual evidence presented in the
(a) 6.25% (b) 25% (c) 50% figure underscores the effectiveness of the proposed method
in mitigating occlusion attacks.

5 Conclusion

This paper introduces an integrated image encryption algo-

(d) (e) (f)
rithm leveraging the Baker map and a novel hyperchaotic
system. This algorithm is tailored for multiple image encryp-
Fig. 13  The occlusion attack and corresponding decrypted images tion, resulting in enhanced security. A mutual permutation
method is devised to effectively scramble sub-images using
4.7 Noise and occlusion attack analysis sequences generated by the hyperchaotic system. Through
mutual permutation, the integrated image undergoes com-
During transmission and storage, images are inevitably sub- prehensive permutation. By amalgamating multiple images
jected to disturbances such as noise and data loss. Conse- into a single enlarged image, the period of the Baker map is
quently, it is imperative that the proposed scheme exhibits extended. Furthermore, a forward–backward diffusion pro-
robustness to withstand both noise and occlusion attacks. cess is applied to the Zigzag-scanned pixels, further bol-
Different levels of Gaussian noise G ∼ (0, 1) are added to the stering the security of the algorithm. Experimental results
ciphertext image, and the corresponding decrypted image is and security analyses demonstrate the algorithm’s efficacy
shown in Fig. 12. From the figure, it is evident that although in withstanding various attacks and its overall performance.
the image becomes blurred due to noise, the main content The histogram variance values are close to 200. The infor-
remains recognizable. These results indicate that the scheme mation entropy value of integrated ciphertext is close to
effectively withstands a certain degree of noise attack. 7.9990, and the UACI and NPCR are close to ideal values.
The proposed integrate multi-image encryption algorithm
C� = C + sG (18) can significantly improve the efficiency and security of
image encryption, and has broad application potential in
the real world. For example, in remote healthcare, patient

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

247 Page 12 of 13 X. Liu

privacy can be well protected. In video surveillance, the coding. IEEE Signal Process. Lett. 29, 2033–2037 (2022). https://​
high efficiency method can meet the real-time requirements. doi.​org/​10.​1109/​LSP.​2022.​32091​45
10. Chen, J., Zhang, Y., Qi, L., Fu, C., Xu, L.: Exploiting chaos-
However, one drawback of the proposed scheme is that the based compressed sensing and cryptographic algorithm for image
data volume of the final ciphertext remains unchanged. In encryption and compression. Opt. Laser Technol. 99, 238–248
future research, efforts will be directed towards proposing (2018). https://​doi.​org/​10.​1016/j.​optla​stec.​2017.​09.​008
and employing new hyperchaotic systems to bolster security. 11. Perepechaenko, M., Kuang, R.: Quantum encryption of super-
position states with quantum permutation pad in IBM quantum
Additionally, a key objective will be to reduce the data vol- computers. EPJ Quantum Technol. 10, 7 (2023). https://​doi.​org/​
ume of the ciphertext. 10.​1140/​epjqt/​s40507-​023-​00164-3
12. Abd-El-Latif, A.A., Abd-El-Atty, B., Abou-Nassar, E.M., Ven-
Acknowledgements The work was partly funded by the Chongqing egas-Andraca, S.E.: Controlled alternate quantum walks based
Natural Science Foundation (Grant No. cstc2020jcyj-msxmX0767), the privacy preserving healthcare images in Internet of Things. Optics
National Natural Science Foundation of China (Grant No. 61802037), Laser Technol. 124, 105942 (2020). https://d​ oi.o​ rg/1​ 0.1​ 016/j.o​ ptla​
and the Fundamental Research Funds for the Central Universities stec.​2019.​105942
(Grant No. SWU-KQ22002). 13. Yang, Y.-G., Tian, J., Lei, H., Zhou, Y.-H., Shi, W.-M.: Novel
quantum image encryption using one-dimensional quantum cel-
Author contributions The authors X.B. Liu contributed to the concep- lular automata. Inf. Sci. 345, 257–270 (2016). https://​doi.​org/​10.​
tion of the study, performed the experiment the data analyses and wrote 1016/j.​ins.​2016.​01.​078
the manuscript. 14. Wang, X., Su, Y.: Image encryption based on compressed sensing
and DNA encoding. Signal Process. Image Commun. 95, 116246
Data availability No datasets were generated or analysed during the (2021). https://​doi.​org/​10.​1016/j.​image.​2021.​116246
current study. 15. Yao, X., Xie, R., Zan, X., Su, Y., Xu, P., Liu, W.: A novel image
encryption scheme for DNA storage systems based on DNA
Declarations hybridization and gene mutation. Interdiscip Sci Comput Life
Sci. (2023). https://​doi.​org/​10.​1007/​s12539-​023-​00565-z
Conflict of interest The authors declare no competing interests. 16. Wang, Q., Zhang, X., Zhao, X.: Color image encryption algorithm
based on bidirectional spiral transformation and DNA coding.
Physica Scripta. 98, 025211 (2023). https://d​ oi.o​ rg/1​ 0.1​ 088/1​ 402-​
4896/​acb322
References 17. Situ, G., Zhang, J.: Multiple-image encryption by wavelength mul-
tiplexing. Opt. Lett. 30, 1306 (2005). https://​doi.​org/​10.​1364/​OL.​
1. Xu, C., Li, D., Guo, K., Yin, Z., Guo, Z.: Computational ghost 30.​001306
imaging with key-patterns for image encryption. Optics Com- 18. Situ, G., Zhang, J.: Position multiplexing for multiple-image
mun. 537, 129190 (2023). https://​doi.​org/​10.​1016/j.​optcom.​2022.​ encryption. J. Opt. A Pure Appl. Opt. 8, 391–397 (2006). https://​
129190 doi.​org/​10.​1088/​1464-​4258/8/​5/​005
2. Zhang, F., Guo, Y., Pu, M., Chen, L., Xu, M., Liao, M., Li, L., Li, 19. Kim, Y., Song, J., Moon, I., Lee, Y.H.: Interference-based multi-
X., Ma, X., Luo, X.: Meta-optics empowered vector visual cryp- ple-image encryption using binary phase masks. Opt. Lasers Eng.
tography for high security and rapid decryption. Nat. Commun. 107, 281–287 (2018). https://d​ oi.o​ rg/1​ 0.1​ 016/j.o​ ptlas​ eng.2​ 018.0​ 1.​
14, 1946 (2023). https://​doi.​org/​10.​1038/​s41467-​023-​37510-z 012
3. Yan, S., Gu, B., Wang, E., Ren, Y.: Finite-time synchronization 20. Wang, X., Wang, Y.: Multiple medical image encryption algo-
of multi-scroll hyperchaotic system and its application in image rithm based on scrambling of region of interest and diffusion
encryption. Math. Comput. Simul. 206, 391–409 (2023). https://​ of odd-even interleaved points. Expert Syst. Appl. 213, 118924
doi.​org/​10.​1016/j.​matcom.​2022.​11.​016 (2023). https://​doi.​org/​10.​1016/j.​eswa.​2022.​118924
4. Liu, Y., Shen, X., Liu, J., Peng, K.: Optical asymmetric JTC cryp- 21. Perez, R.A., Vilardy, J.M., Pérez-Cabré, E., Millán, M.S., Torres,
tosystem based on multiplication-division operation and RSA C.O.: Nonlinear encryption for multiple images based on a joint
algorithm. Opt. Laser Technol. 160, 109042 (2023). https://​doi.​ transform correlator and the gyrator transform. Sensors. 23, 1679
org/​10.​1016/j.​optla​stec.​2022.​109042 (2023). https://​doi.​org/​10.​3390/​s2303​1679
5. Li, J., Chen, L., Cai, W., Xiao, J., Zhu, J., Hu, Y., Wen, K.: Holo- 22. Sabir, S., Guleria, V.: Multi-layer security based multiple image
graphic encryption algorithm based on bit-plane decomposition encryption technique. Comput. Electr. Eng. 106, 108609 (2023).
and hyperchaotic Lorenz system. Opt. Laser Technol. 152, 108127 https://​doi.​org/​10.​1016/j.​compe​leceng.​2023.​108609
(2022). https://​doi.​org/​10.​1016/j.​optla​stec.​2022.​108127 23. Zhang, X., Gao, T.: Multiple-image encryption algorithm based
6. Zhang, Z., Tang, J., Ni, H., Huang, T.: Image adaptive encryp- on the bit plane and superpixel. Multimed. Tools Appl. (2022).
tion algorithm using a novel 2D chaotic system. Nonlinear Dyn. https://​doi.​org/​10.​1007/​s11042-​022-​14160-9
(2023). https://​doi.​org/​10.​1007/​s11071-​023-​08397-8 24. Su, Y., Tang, C., Gao, G., Gu, F., Lei, Z., Tang, S.: Optical encryp-
7. Lai, Q., Hu, G., Erkan, U., Toktas, A.: High-efficiency medical tion scheme for multiple color images using complete trinary tree
image encryption method based on 2D Logistic-Gaussian hyper- structure. Opt. Lasers Eng. 98, 46–55 (2017). https://​doi.​org/​10.​
chaotic map. Appl. Math. Comput. 442, 127738 (2023). https://​ 1016/j.​optla​seng.​2017.​05.​019
doi.​org/​10.​1016/j.​amc.​2022.​127738 25. Gao, X., Mou, J., Xiong, L., Sha, Y., Yan, H., Cao, Y.: A fast
8. Wang, X., Zhao, M., Feng, S., Chen, X.: An image encryption and efficient multiple images encryption based on single-channel
scheme using bit-plane cross-diffusion and spatiotemporal chaos encryption and chaotic system. Nonlinear Dyn. 108, 613–636
system with nonlinear perturbation. Soft. Comput. 27, 1223–1240 (2022). https://​doi.​org/​10.​1007/​s11071-​021-​07192-7
(2023). https://​doi.​org/​10.​1007/​s00500-​022-​07706-4 26. Erkan, U., Toktas, A., Lai, Q.: 2D hyperchaotic system based on
9. Zhang, B., Xiao, D., Li, Y., Yang, L.: 2D compressed sensing Schaffer function for image encryption. Expert Syst. Appl. 213,
using nonlocal low-rank prior reconstruction for cipher-image 119076 (2023). https://​doi.​org/​10.​1016/j.​eswa.​2022.​119076

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Integrate encryption of multiple images based on a new hyperchaotic system and Baker map Page 13 of 13 247

27. Lai, Q., Hu, G., Erkan, U., Toktas, A.: A novel pixel-split image and Fractional. 7, 887 (2023). https://​doi.​org/​10.​3390/​fract​alfra​ct712​
encryption scheme based on 2D Salomon map. Expert Syst. Appl. 0887
213, 118845 (2023). https://​doi.​org/​10.​1016/j.​eswa.​2022.​118845 45. Qian, K.: A robust memristor-enhanced polynomial hyper-chaotic
28. Çavuşoğlu, Ü., Kaçar, S., Pehlivan, I., Zengin, A.: Secure image map and its multi-channel image encryption application. Microma-
encryption algorithm design using a novel chaos based S-Box. chines. 14, 2090 (2023). https://​doi.​org/​10.​3390/​mi141​12090
Chaos, Solitons Fractals 95, 92–101 (2017). https://​doi.​org/​10.​ 46. Feng, W., Zhao, X., et al.: Image encryption algorithm based on
1016/j.​chaos.​2016.​12.​018 plane-level image filtering and discrete Logarithmic transform.
29. Al-Hazaimeh, O.M., Al-Jamal, M.F., Alhindawi, N., Omari, A.: Mathematics 10, 2751 (2022). https://​doi.​org/​10.​3390/​math1​01527​
Image encryption algorithm based on Lorenz chaotic map with 51
dynamic secret keys. Neural Comput. Appl. 31, 2395–2405 (2019). 47. Alawida, M., Samsudin, A., et al.: Deterministic chaotic finite-state
https://​doi.​org/​10.​1007/​s00521-​017-​3195-1 automata. Nonlinear Dyn. 98, 2403–2421 (2019). https://​doi.​org/​10.​
30. Zhou, S., Wang, X., Zhang, Y.: Novel image encryption scheme 1007/​s11071-​019-​05311-z
based on chaotic signals with finite-precision error. Inf. Sci. 621, 48. Erkan, U., Toktas, A., Lai, Q.: Design of two dimensional hyper-
782–798 (2023). https://​doi.​org/​10.​1016/j.​ins.​2022.​11.​104 chaotic system through optimization benchmark function. Chaos
31. Wang, X., Leng, Z.: A dynamic image encryption algorithm based Solitons & Fractals 167, 113032 (2023). https://​doi.​org/​10.​1016/j.​
on improved ant colony walking path thought. Sens Imaging. 23, 17 chaos.​2022.​113032
(2022). https://​doi.​org/​10.​1007/​s11220-​022-​00387-1 49. Abba, A., Teh, J.S., Alawida, M.: Towards accurate keyspace analy-
32. Wang, X., Wang, X., Teng, L., Jiang, D.: A novel meaningful image sis of chaos-based image ciphers. Multimed. Tools Appl. (2024).
encryption algorithm based on newly-designed coupled map lattice https://​doi.​org/​10.​1007/​s11042-​024-​18628-8
and adaptive embedding. Optik 270, 170073 (2022). https://​doi.​org/​ 50. Wen, H., Lin, Y.: Cryptanalysis of an image encryption algorithm
10.​1016/j.​ijleo.​2022.​170073 using quantum chaotic map and DNA coding. Expert Syst. Appl.
33. Mamdouh, E., et al.: Color image cryptosystem based on Sine cha- 237, 121514 (2024). https://​doi.​org/​10.​1016/j.​eswa.​2023.​121514
otic map, 4D Chen hyperchaotic map of fractional-order and hybrid 51. Wen, H., Lin, Y.: Cryptanalyzing an image cipher using multiple
DNA coding. IEEE Access. 11, 54928–54956 (2023). https://​doi.​ chaos and DNA operations. J King Saud Univ-Comput. Inf. Sci. 35,
org/​10.​1109/​ACCESS.​2023.​32821​60 101612 (2023). https://​doi.​org/​10.​1016/j.​jksuci.​2023.​101612
34. Alexan, W., El-Damak, D., Gabr, M.: Image encryption based on 52. Feng, W., Qin, Z.: Cryptanalysis and improvement of the image
Fourier-DNA coding for hyperchaotic Chen system, Chen-based encryption scheme based on Feistel network and dynamic DNA
binary quantization S-Box, and variable-base modulo operation, encoding. IEEE Access 9, 145459–145470 (2021). https://​doi.​org/​
IEEE. Access 12, 21092–21113 (2024). https://​doi.​org/​10.​1109/​ 10.​1109/​ACCESS.​2021.​31235​71
ACCESS.​2024.​33630​18 53. Fridrich, J.: Symmetric Ciphers Based on Two-Dimensional Chaotic
35. Gabr, M., Elias, R., et al.: Image encryption via base-n PRNGs and Maps. Int. J. Bifurcation Chaos. 08, 1259–1284 (1998). https://​doi.​
parallel base-n S-boxes. IEEE Access. 11, 85002–85030 (2023). org/​10.​1142/​S0218​12749​80009​8X
https://​doi.​org/​10.​1109/​ACCESS.​2023.​33014​60 54. Liu, Z., Xu, L., Liu, T., Chen, H., Li, P., Lin, C., Liu, S.: Color image
36. Gabr, M., Korayem, Y., et al.: R ­ 3-Rescale, Rotate, and Randomize: encryption by using Arnold transform and color-blend operation in
A novel image cryptosystem utilizing chaotic and hyper-chaotic sys- discrete cosine transform domains. Optics Commun. 284, 123–128
tems, IEEE. Access 11, 119284–119312 (2023). https://​doi.​org/​10.​ (2011). https://​doi.​org/​10.​1016/j.​optcom.​2010.​09.​013
1109/​ACCESS.​2023.​33268​48 55. Alhumyani, H., El-Banby, G.M., El-Sayed, H.S., Abd El-Samie,
37. Alexan, W., Korayem, Y., Gabr, M., et al.: AntEater: when Arnold’s F.E., Faragallah, O.S.: Efficient generation of cancelable face tem-
Cat meets Langton’s Ant to encrypt images. IEEE Access. 11, plates based on quantum image hilbert permutation. Electronics 11,
106249–106276 (2023). https://​doi.​org/​10.​1109/​ACCESS.​2023.​ 1040 (2022). https://​doi.​org/​10.​3390/​elect​ronic​s1107​1040
33193​35 56. Zhou, N., Yan, X., Liang, H., Tao, X., Li, G.: Multi-image encryp-
38. Alexan, W., Aly, L., Korayem, Y., et al.: Secure communication of tion scheme based on quantum 3D Arnold transform and scaled
military reconnaissance images over UAV-assisted relay networks. Zhongtang chaotic system. Quantum Inf. Process. 17, 338 (2018).
IEEE Access. 12, 78589–78610 (2024). https://​doi.​org/​10.​1109/​ https://​doi.​org/​10.​1007/​s11128-​018-​2104-6
ACCESS.​2024.​34078​38 57. Thomas, S., Krishna, A.: Securing grayscale image using improved
39. Laiphrakpam, D.S., Thingbaijam, R., et al.: Encrypting multiple Arnold transform and ElGamal encryption. J. Electr. Imaging
images with an enhanced chaotic map. IEEE Access. 10, 87844– (2022). https://​doi.​org/​10.​1117/1.​JEI.​31.6.​063012
87859 (2022). https://​doi.​org/​10.​1109/​ACCESS.​2022.​31997​38 58. Hussain, I., Shah, T., Gondal, M.A., Mahmood, H.: Efficient method
40. Mehmood, A., Shafique, A., et al.: A time-efficient and noise-resist- for designing chaotic S-boxes based on generalized Baker’s map
ant cryptosystem based on discrete wavelet transform and chaos and TDERC chaotic sequence. Nonlinear Dyn. 74, 271–275 (2013).
theory: An application in image encryption. J Inf. Secur. Appl. 78, https://​doi.​org/​10.​1007/​s11071-​013-​0963-z
103590 (2023). https://​doi.​org/​10.​1016/j.​jisa.​2023.​103590 59. Li, C.L., Yu, S.M.: A new hyperchaotic system and its adaptive
41. Alawida, M.: A novel DNA tree-based chaotic image encryption tracking control. Acta Phys. Sin. 6, 040504 (2012). https://​doi.​org/​
algorithm. J. Inf. Secur. Appl. 83, 103791 (2024). https://​doi.​org/​ 10.​7498/​aps.​61.​040504
10.​1016/j.​jisa.​2024.​103791 60. https://​ccia.​ugr.​es/​cvg/​dbima​genes/​g512.​php. Accessed 15 June
42. Alawida, M.: A novel image encryption algorithm based on cyclic 2023
chaotic map in industrial IoT environments. IEEE Trans. Ind. Inf. 61. https://​sipi.​usc.​edu/​datab​ase/​datab​ase.​php. Accessed 15 June 2023
(2024). https://​doi.​org/​10.​1109/​TII.​2024.​33956​31
43. Kocak, O., Erkan, Ugur, et al.: PSO-based image encryption scheme Publisher's Note Springer Nature remains neutral with regard to
using modular integrated logistic exponential map. Expert Syst. jurisdictional claims in published maps and institutional affiliations.
Appl. 237, 121452 (2024). https://​doi.​org/​10.​1016/j.​eswa.​2023.​
121452 Springer Nature or its licensor (e.g. a society or other partner) holds
44. Feng, W., Wang, Q.W., et al.: Exploiting newly designed fractional- exclusive rights to this article under a publishing agreement with the
order 3D Lorenz chaotic system and 2Ddiscrete polynomial hyper- author(s) or other rightsholder(s); author self-archiving of the accepted
chaotic map for high-performance multi-image encryption. Fractal manuscript version of this article is solely governed by the terms of such
publishing agreement and applicable law.

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:

1. use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
2. use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
3. falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
4. use bots or other automated methods to access the content or redirect messages
5. override any security feature or exclusionary protocol; or
6. share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at

[email protected]