Received 30 July 2022, accepted 15 August 2022, date of publication 18 August 2022, date of current version 26 August 2022.

Digital Object Identifier 10.1109/ACCESS.2022.3199738

Encrypting Multiple Images With

an Enhanced Chaotic Map
LAIPHRAKPAM DOLENDRO SINGH1 , (Member, IEEE), ROHIT THINGBAIJAM1 ,
KHOIROM MOTILAL SINGH2 , AND MOATSUM AL AWIDA 3
1 Department of Computer Science and Engineering, National Institute of Technology Silchar, Silchar, Assam 788010, India
2 Department of CSE, Koneru Lakshmaiah Education Foundation, Vaddeswaram, Andhra Pradesh 522302, India
3 Department of Computer Sciences and Information Technology, Abu Dhabi University, Abu Dhabi 59911, United Arab Emirates

Corresponding author: Laiphrakpam Dolendro Singh ([email protected])

This research is funded by the Office of Research & Sponsored Programs of Abu Dhabi University, United Arab Emirates and this work
has been partially supported by Abu Dhabi University under Grant No 19300635.

ABSTRACT A multiple-image encryption scheme based on an enhanced chaotic map is proposed. This
scheme combines multiple grayscale images into three planes. An amplified sine map is used to generate
a dynamic permutation table and a chaotic sequence. The initial parameters of the amplified sine map are
obtained from an elliptic curve coordinate, computing using elliptic curve point multiplication between the
input image hash value and the seed value of an elliptic curve. The dynamic permutation table performs
cyclic shift transformation on the input image, diffusing it horizontally and vertically. The diffused image is
converted to a cipher image through an XOR operation with the chaotic sequence generated by the amplified
sine map. Using an amplified sine map provides larger key space, more extensive chaotic range, more
extensive control parameters, better sensitive initial values and enhanced security against cryptanalysis.
The experiment results indicate that the proposed scheme is fast, secure, efficient and can resist certain
cryptographic attacks. Compared to other recent schemes, the proposed scheme is shown to be secure and
efficient.

INDEX TERMS Multiple image encryption, cyclic shift transformation, amplified sine map, elliptic curve
cryptography.

I. INTRODUCTION non-convergence, and high sensitivity to the initial condi-

With the rapid advancement of multimedia technology, devel- tions, encryption schemes that are based on chaos-based [7]
opers and researchers have made it extremely simple to trans- have piqued the attention of a large number of developers and
fer multiple data sets simultaneously. As a result, there is a researchers.
greater need for high-security and faster data communication In chaos-based cryptography, two main issues arise (i) one-
methods. Image, video, and other multimedia data have high dimensional (1D) chaotic maps which do not satisfy the
data redundancy and capacity, as well as a high correla- unpredictability property, and (ii) high-dimensional (HD)
tion between adjacent pixels. Traditional encryption schemes chaotic maps, despite exhibiting complex and chaotic behav-
like Data Encryption Standard (DES), Advanced Encryption ior, it requires more computational complexity [8], [9]. This
Standard (AES), and others are not always efficient enough led to the state-of-the-art 1D chaotic map with a simple
to encrypt these data. Researchers developed various encryp- structure, more extensive chaotic range, better sensitive initial
tion techniques by using concepts on DNA encoding [1], values with low computational complexity.
wavelet transform [2], [3], compressed sensing [4], ElGa- Mansouri et al. [10] proposed a new sine powered chaotic
mal cryptosystem [5], visual secret sharing [6], etc. On the system. The chaotic map is then used in a new encryption
other hand, because of the ergodicity, unpredictability, scheme based on row-by-row and column-by-column con-
fusion and diffusion operations. The chaotic system exhibits
The associate editor coordinating the review of this manuscript and high randomness and sensitivity, solving the correlation of
approving it for publication was Sathish Kumar . adjacent pixels. Later, they also proposed a new amplified

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
87844 VOLUME 10, 2022
L. D. Singh et al.: Encrypting Multiple Images With an Enhanced Chaotic Map

1-D chaotic map generator [11], which amplifies the existing 1D classical chaotic maps suffer when realized on a digital
1-D maps. The amplified chaotic map is then used for the computer is dynamical degradation. The enhanced meth-
generation of initial parameters for the index representation ods can be divided into nine categories, cascading multi-
operation. Finally, a bit-level operation is carried out to gen- ple chaotic maps [21], perturbing the chaotic states [22],
erate the cipher image. switching between multiple chaotic maps [23], error com-
During the process of encryption using Latin squares [12], pensation [24], combining chaotic maps using modular
the three channels of colour images are encrypted one by operation [25], coupling chaotic maps [26], introducing
one, which results in many redundant operations and low delays [27], bit reversal chaotic map [28], and finite state
efficiency. Hua et al. [13] developed a colour image encryp- machine [29].
tion scheme using orthogonal Latin squares and a 2D chaotic These enhanced methods make use of additional exter-
system that overcomes the weakness of the Latin squares. nal sources, including multiple chaotic maps, another
Further, Hua et al. [14], [15] and proposed a 2D parametric chaotic map, and pseudo-random numbers, which makes the
polynomial chaotic system and n-dimensional polynomial chaotic map’s structure complex and boosts computational
chaotic system that possess robust chaos, late chaos degrada- complexity. Our chaotic map can create chaotic points seam-
tion and desired dynamic properties with customized param- lessly without the need for additional mathematical opera-
eters for desired Lyapunov exponents for better dynamic tions. Additionally, our chaotic map can produce high-quality
properties. behaviour like security and randomness without the support
Zhang et al. [16] use DNA encoding and a chaotic sys- of an additional external source like a classical chaotic map.
tem to develop a multiple-image encryption technique. The chaotic image encryption structure’s augmented
Multiple k images are combined and a combined big image is chaotic maps were applied in various recent proposed image
generated and SHA256 is used on the big image to determine encryption algorithms. In a 1D chaotic map, which is used as
the initial parameter of the chaotic map. The big image is fur- a source of entropy, chaotic points are leveraged to encrypt
ther decomposed into k encrypted images after pixel shuffling image pixels using the diffusion and confusion fundamental
with the Piece-wise Linear Chaotic Maps (PWLCM) and processes. Because initial conditions and control parameters
DNA decoding. Concerning the encryption process’s speed, are so sensitive to even tiny changes, the majority of proposed
Enayatifar et al. [17] developed a multiple-image encryp- algorithms employ them as secure keys. A chaotic image
tion method based on Index-based permutation-diffusion and encryption-based chaotic Jaya optimization technique was
DNA sequences. A single big image is created using multi- developed by the authors of [30] to produce S-boxes. The
ple images, and then converted into a one-dimension array. proposed algorithm employed two encryption operations to
DNA sequence and cellular automata are employed to make accomplish Shannon’s confusion/diffusion characteristics.
permutation more secure in a quick and effective combined A new, improved 1D sinusoidal chaotic system (I1DS),
index based permutation and diffusion system. which is the core part of the proposed image encryption, was
Zarebnia et al. [18] created a hybrid chaotic systems-based introduced in [31]. The authors presented new image encryp-
multiple image encryption. In their method, the smaller tion based on dynamic bit-shifting recombination operation
blocks of plain-images are shuffled using the chaotic system, and a non-linear diffusion technique. A novel signcryption
and permutation is performed. Using the Arnold cat map system for medical images was proposed by the authors
and cyclic shift operation, the cipher image is obtained using of [32] based on a new chaotic map and a combination
XoR operation. Another PWLCM based multiple-image strategy of hybrid cryptography techniques. The medical
encryption scheme was proposed by Patro et al. [19]. In their images were encrypted using elliptic curve cryptography.
method, multiple grayscale images were combined to create The authors improved the 1D classical chaotic map (Tent
a bigger 2D-image. Row-wise and column-wise permutation map) using a finite state machine, and they then utilized the
is performed to scramble the big image and produce the enhanced chaotic map to create a novel image encryption
cipher image through the cross-coupled PWLCM technique. algorithm. The proposed chaotic map’s security is a prereq-
By combining the theory of chaotic system and elliptic curve uisite for the proposed algorithm [33].
cryptography (ECC), Singh et al. [20] use hyper-chaotic sys- The proposed method can encrypt twelve secret images
tem with 3D scrambling to develop multiple image encryp- with more than the those given in [16], [18], [19], [20], [34],
tion. They combine n-images to produce a 3D image and and [36] except for [17], [35], and [37] with any i numbers
the hyper-chaotic system’s initial parameters are generated of images. Theoretically, in [17], [35], and [37], i numbers
using SHA512. The permutation and substitution operations of images can be given as input. However, a bottleneck can
in carried out using the hyper-chaotic system. Finally, the occur with respect to resource constraints in the computing
cipher 3D image is formed by conducting a bit-wise XoR device during actual implementation. Unlike other proposed
operation between the scrambled 3D image and the chaotic methods where multiple chaotic systems are used as in [16],
3D image. [18], [19], and [35] to provide the discrete dynamics of
Many methods have been proposed to improve 1D chaos, or that of hyper-chaotic system [20], [37], the proposed
chaotic maps’ security levels so they can be used in method only uses Amplified Sine Map (ASM) to fulfill the
cryptography applications. The fundamental problem that discrete dynamics of chaos. With the use of ECC [38], [39]

VOLUME 10, 2022 87845

 L. D. Singh et al.: Encrypting Multiple Images With an Enhanced Chaotic Map

for key exchange, the proposed method is secure, efficient performed and tabulated in Table 1 to test the randomness
and robust against various statistical attacks. of the pseudo-random sequence (PRS) generated from ASM.
The overall contribution of the proposed encryption
scheme can be summarized as follows: B. CYCLIC SHIFT TRANSFORMATION (CST)
1) A multiple grayscale image encryption based on Cyclic shift transformation (CST) is a method for scrambling
enhanced chaotic map is proposed. image pixels in order to reduce pixel correlation. A dynamic
2) To avoid known plain-text attacks and chosen plain- permutation table is generated using a multi chaotic map
text attacks, SHA384 is adopted to generate the initial which helps in performing random byte circular shifts along
conditions of the amplified chaotic system. horizontal and vertical axes. The step size of the shift is based
3) An amplified Sine map is used, which exhibits complex on the index of the permutation table. The structure of CST
chaotic behaviour with low computational complexity. in the proposed method are as follows:
4) A cyclic shift transformation (CST) is introduced
where random byte shifts along the horizontal and Step 1: An ASM sequence is generated using Equation 2.
vertical axes are carried out. The sequence is sorted in ascending order and dupli-
5) The iteration involved in the computation of random cate elements are removed.
sequence used for encryption is reduced by a factor of Step 2: A permutation table (P_table) is generated using the
four using base conversion operation. values from step 1 whose length is equal (M1 + N1 ),
The rest of the paper is structured as follows. Section II where M1 and N1 are the row and column dimension
describes the preliminaries. Section III describes the pro- of the combined input image.
posed multiple image encryption method. Section IV displays Step 3: Row_P − table and Column_P − table are assigned
the simulation. Section V presents the security analysis of the to the combined input image using the values from
algorithm, followed by the Conclusion in Section VI. step 2.
Step 4: If the particular Row_P − table is i, the correspond-
II. PRELIMINARIES
ing row is circular right shifted to i positions.The
Circular right shift is performed for all the rows
A. CHAOTIC MAP
with respect to its Row_P − table value. Similarly,
Initial values and control parameters are extremely important
if the particular Column_P − table value is j, the
in chaotic maps. Any small variation in the initial conditions
corresponding columns is circular down shifted to
results in a significant difference. One-dimensional (1D) and
j positions.The Circular down shift is performed for
high-dimensional chaotic maps are the two types of chaotic
all the column with respect to its Column_P − table
maps. One-dimensional maps usually feature only one or a
value.
few variables, as well as a limited output range.
The above CST process is illustrated in Fig. 3 and
Fig. 4 for shift row and shift column respectively.
1) SINE MAP
Algorithm 1 represents the pseudo code generated
The 1D Sine Map (SM) is define as: for CST.
xn+1 = γ sin(πxn ) (1)
C. ELLIPTIC CURVE OPERATIONS
where γ ∈ [0, 1] is the control parameter and when γ ∈
Given an elliptic curve Ep : y2 ≡ x 3 + ax + b mod p and the
[0.87, 1], it manifests chaotic behavior.
seed point S(xs , ys ). The elliptic curve point h.S(xhS , yhS ) =
S + S + S + . . . h times, where h is an integer. hS(xhS , yhS )
2) AMPLIFIED SINE MAP
can be computed using Algorithm 2.
Unlike the traditional Sine Map, the Amplified Sine Map
The algorithm to compute point addition (P3 (x3 , y3 ) =
(ASM) [11] possesses a simple structure, larger chaotic range,
P1 (x1 , y1 ) + P2 (x2 , y2 )) when P1 6= P2 or P1 == P2
bigger control parameters and better sensitive initial values.
is shown in Algorithm 3. Point subtraction (P3 (x3 , y3 ) =
The Amplified Sine Map is defined as:
P1 (x1 , y1 ) − P2 (x2 , y2 )) is converted to point addition by
xn+1 = |cos(γ δπsin(πxn ))| /Log(3 − γ sin(πxn )) (2) converting the y2 coordinate to negative as (P3 (x3 , y3 ) =
P1 (x1 , y1 ) + P2 (x2 , −y2 )).
where γ > 0, δ ∈ [0, 1] and xn ∈ [0, 1] is the chaotic system
parameter. III. PROPOSED MULTIPLE IMAGE ENCRYPTION
The bifurcation diagram of SM and ASM is shown in The proposed multiple image encryption is as follows-
Fig. 1. The bifurcation diagram is used to visually analyse
the chaotic behaviour of a dynamic system. Fig. 1 shows that
A. KEY GENERATION
the ASM exhibit chaotic behaviour over the whole control
parameter range, with a wider output range than the SM. The key generation of the proposed method is as follows:
The Lyaponov Exponent in Fig. 2 also shows that the ASM Step 1: Combine all the input grayscale images into three
exhibit good chaotic behaviour. NIST randomness test is planes of equal size and obtain a colour image I1 .

87846 VOLUME 10, 2022

L. D. Singh et al.: Encrypting Multiple Images With an Enhanced Chaotic Map

FIGURE 1. Bifurcation diagram of (a) Sine map (b) Amplified Sine Map.

BaseConvert[hS5 , 2]
δ2 = (7)
2128
BaseConvert[hS6 , 2]
γ2 = 2200 + (8)
2116
where, xn1 , γ1 and δ1 will be used as the keys for
the permutation table of the CST. xn2 , γ2 and δ2 will
be used as the keys for generating a pseudoran-
dom sequence PRS. BaseConvert[list, b] converts
the value of the list with respect to base b.

B. ENCRYPTION PROCESS
The encryption process is carried out as shown in the
algorithm 4 and discussed below.
FIGURE 2. Lyaponov exponent of Sine map and Amplified Sine Map.
Step 1: Combine all the input grayscale images into three
planes of equal size and obtain a colour image I1 .
Step 2: Perform CST on all the three planes of I1 using
Step 2: Compute SHA-348 on I1 to generate a 384-bit hash algorithm 1 and the keys xn1 , γ1 and δ1 .
value (h). Step 3: Using Equation 2 along with the keys xn2 , γ2 and
Step 3: Compute hS(xhS , yhS ) by performing an elliptic δ2 generate a pseudo-random sequence PRS of
curve point multiplication between (h) and elliptic length (M1 × N1 × 3/4), where (M1 × N1 ) is the
curve seed S(xs , ys ). dimension of I1 .
Step 4: Binarize xhS and yhS into 384 bits each and store as Step 4: Compute |ASM [i] × 1016 | and base convert it using
binx and biny respectively. 256 as base upto eight terms and pick the elements
Step 5: Divide the 384-bits in binx and biny into three groups at position three to six and store in PSR.
each of 128 bits and denote as hS1 , hS2 , hS3 , hS4 , Step 5: Perform XoR operation between the values of Step 2
hS5 and hS6 . and PRS.
Step 6: Generate the keys as: Step 6: Generate the cipher image.

BaseConvert[hS1 , 2] C. KEY SHARING

xn1 = (3)
2128 The initial parameters xn1 , γ1 , δ1 , xn2 , γ2 and δ2 are all gener-
BaseConvert[hS2 , 2] ated from h.S(xhS , yhS ). In order to decrypt the images exactly
δ1 = (4) as the original images, the same set of initial parameters are
2128
BaseConvert[hS3 , 2] required in the receiver side. h.S(xhS , yhS ) is secretly shared
γ1 = 2200 + (5) 0 , y0 ) to the communicating party. h.S 0 (x 0 , y0 ) is
as h.S 0 (xhS hS hS hS
2116
computed as:
BaseConvert[hS4 , 2]
xn2 = (6)
2128 h.S 0 = {r.S, (h.S + r.R)} (9)
VOLUME 10, 2022 87847
 L. D. Singh et al.: Encrypting Multiple Images With an Enhanced Chaotic Map

TABLE 1. Randomness Test on sequence generated from ASM.

where, Step 1: Using Equation 2 along with the keys xn2 , γ2 and
r is a random integer δ2 generate a pseudorandom sequence PRS of length
R public of the receiver. (M1 × N1 × 3), where (M1 × N1 ) is the dimension
R = key.S of C1 .
key is the private key of the receiver. Step 2: Perform XoR operation between PRS and the three
The receiver gets (h.S(xhS , yhS ) from h.S 0 (xhS
0 , y )0 by
hS planes of C1 .
using Equation 10 Step 3: Perform reverse CST on the three planes of step
2 using the permutation table the keys xn1 , γ1 , δ1 .
h.S = (h.S + r.R) − key.r.S (10)
Step 4: Separate the three planes of Step 3.
Step 5: Extract all the plain images from Step 4.
D. DECRYPTION PROCESS
Using Equation 10, the receiver computes (h.S(xhS , yhS )
0 , y )0 using his/her private key key. The initial
from h.S 0 (xhS IV. SIMULATIONS
hS
parameters xn1 , γ1 , δ1 , xn2 , γ2 and δ2 are generated from the The proposed multiple grayscale image encryption is simu-
shared key (h.S(xhS , yhS ) using Step 4 to Step 6 given in III-A. lated using Wolfram Mathematica 13 on a Fujitsu Celsius
The decryption process of the proposed method is as follows: Workstation with Intel Xeon(R) W-2133 @ 3.60 GHz and

87848 VOLUME 10, 2022

L. D. Singh et al.: Encrypting Multiple Images With an Enhanced Chaotic Map

FIGURE 3. Shift Row Operation.

FIGURE 4. Shift Column Operation.

1 −1 NX
MX 1 −1
a ram of 32GB. The experimental images were downloaded 1
from the USC-SIPI database [40]. In this paper, due to space Meanserror = [OImg (i, j) − Dimg (i, j)]2
M1 × N1
i=0 j=0
limitations, we used 12 (twelve) images to form a single
combined RGB image (CI ). Each plane of the RGB is com- (12)
posed of 4 (four) grayscale images. Experimental analysis where Maxpmvalue : maximum supported pixel value. Meanserror :
are carried out on four combine images- CI −1 , CI −2 , CI −3 mean squared error. M1 × N1 : dimension of the image.
and CI −4 . The image representation of CI −1 , CI −2 , CI −3 and OImg (i, j): original image pixel value at index (i, j).DImg (i, j):
CI −4 are shown in the Fig. (6a, 6c, 6e and 6g) respectively. decrypted image pixel value at index (i, j).
The images used in combine images- CI −1 , CI −2 , CI −3 , CI −4 y xy
2Mvx Mv + cv 2Cov + c0v



and its respective planes are shown in Table 2. The pro- SSIM(x, y) = y
y

posed method can encrypt and decrypt the multiple grayscale (Mvx )2 + (Mv )2 + cv (Svx )2 + (Sv )2 + c0v
images successfully. (13)
y
where Mvx = mean value of x.Mv = mean value of y.Svx =
V. SECURITY ANALYSIS standard deviation value of x.Svx = standard deviation value
xy
A. PSNR AND SSIM of y.Cov = covariance value of x and y. cv = (kv Lv )2 · c0v =
The peak-signal-to-noise-ratio (PSNR) and structural- 2
kv0 Lv .kv = 0.01 and kv0 = 0.03.Lv = 2count − 1. count=


similarity-index (SSIM) are used to ascertain the accuracy no. of bit per pixel.
of the image after the decryption. The computed values are
shown in Table 3. The PSNR and SSIM are calculated using B. KEY SPACE
the below formula: A key space is the total number of all the possible keys
for a given encryption scheme. The key space in an ideal
Maxpmvalue encryption technique should be large enough to resist a brute
PSNR = 20 × log10 (11)
Meanserror force attack. The keys used in the proposed method are

VOLUME 10, 2022 87849

 L. D. Singh et al.: Encrypting Multiple Images With an Enhanced Chaotic Map

FIGURE 5. The process of encryption using ASM and CST for multiple images.

(xn1 , γ1 , δ1 , xn2 , γ2 , δ2 ). As a result, the suggested enhanced The overall time complexity to execute the encryption
encryption scheme’s total key space is 2332 (approx). The procedure is O(3 × (M1 + N1 )) + O(3 × (M1 + N1 ) + O(3 ×
value is sufficiently big to overcome any type of brute force M1 ×N1 )+ O(8×3×M1 ×N1 ) = O(27M1 N1 )+O(6(M1 +N1 )
attack.
E. INFORMATION ENTROPY ANALYSIS
C. HISTOGRAM ANALYSIS The information entropy analyses how the image has random-
An image histogram is an analysis of the statistical attack ness in it by computing:
which is shown graphically by plotting the frequency. The n
frequency is the plot on every pixel value across all the pixel X
Ep (χ) = − Pb (χi ) logβ Pb (χi ) (14)
values possible. The histogram of a plain image shows uneven
i=1
pixel frequency distribution while the histogram of a well
decrypted image shows equal pixel frequency distribution. where Ep (χ): entropy. Pb (χi ) : probability mass function.
Fig. (6a, 6b, 6c, 6d) and (6e, 6f, 6g, 6h) shows the plane β = 2. The entropy value is shown in Table 4 which is
images and the cipher images. Fig. (7a, 7b, 7c, 7d) and computed using equation 14 on the cipher images.A Com-
(7e, 7f, 7g, 7h) represents the histogram of the plane images parison of the entropy with the other state of the art methods
and histogram of the cipher images. is tabulated in Table 10.

D. COMPLEXITY OF THE PROPOSED METHOD F. AVALANCHE EFFECT

The computational complexity of the proposed method can The Avalanche effect is an idea related to the specific
be summarised as: behaviour of the arithmetic function used in encryption.
1) The time complexity of generating the ASM sequence It states that ‘‘a minor variation in plain text can impact
for the permutation table is O(3 × (M1 + N1 )). a substantial variation in the cipher text’’. As a result, it’s
2) The time complexity of performing the CST operation one of the processes to check for desirable qualities of the
on the three planes is O(3 × (M1 + N1 )). cipher image. An effective encryption technique is predicted
3) The time complexity of generating the chaotic to be good if the value of the avalanche is very near to
sequence PSR is O(3 × M1 × N1 ). 50 percent. For the suggested approach, the avalanche value
4) The time complexity of performing a bit-XOR oper- for the cipher image Fig. (8b, 8e, 8h, 8k) generated by
ation on two n-bit value is O(n) and each values are the plain image Fig. (8a, 8d, 8g, 8j) and the cipher image
represented by 8-bits. So, the time complexity of per- Fig. (8c, 8f, 8i, 8l) derived by some minor variation in the
forming bit-XOR operation in between CST[T1 ] and pixel value of plain image Fig. (8a, 8d, 8g, 8j) is shown in
PSR is O(8 × 3 × M1 × N1 ) the Table 5.

87850 VOLUME 10, 2022

L. D. Singh et al.: Encrypting Multiple Images With an Enhanced Chaotic Map

FIGURE 6. (a, c, e, g) Combined images CI−1 , CI−2 , CI−3 and CI−4 . (b, d, f, h ) Cipher image of CI−1 , CI−2 ,
CI−3 and CI−4 .

Algorithm 1: The Process of CST Algorithm 2: Elliptic Curve Point multiplication(.)

Input: INPUT IMAGE I1 , image size M1 × N1 , ASM Input: a, b, p, S(xs , ys ), h
initial parameters (xn1 , γ1 , δ1 ) Output: h.S(xhS , yhS )
1 Start with n = h, P1 = ∞, P2 = S
Output: CST IMAGE, image size M1 × N1
2 while (n 6= 0) do
1 for i ← 1 to M1 + N1 do 3 if (n mod 2 == 0) then
2 Generate ASM [i] 4 n = n/2
3 i++ 5 P1 = P1
4 end for 6 update P2 = P2 + P2
7 end if
5 Seqx = ASM [[1; ; M1 ]]
8 else
6 Seqy = ASM [[M1 + 1; ; N1 ]] 9 n=n−1
7 Sortx =Sort (Seqx ) 10 update P1 = P1 + P2
8 Sorty =Sort (Seqy ) 11 P2 = P2
9 for i ← 1 to Length[Seqx ] do 12 end if
13 end while
10 Row_P − table[i] = Position of Sortx [i] in Seqx
14 if (n == 0) then
11 i++ 15 P1 = hS(xhS , yhS )
12 end for 16 end if
13 for i ← 1 to Length[Seqy ] do
14 Column_P − table[i] = Position of Sorty [i] in Seqy
15 i++ Algorithm 3: Elliptic Curve Point addition(+)
16 end for
Input: a, b, p, P1 (x1 , y1 ), P2 (x2 , y2 )
17 for i ← 1 to Row_P − table[i] do Output: P3 (x3 , y3 )
18 RotateRight, Row[i] of I1 by Row_P − table[i] 1 if (P1 (x1 , y1 ) 6= P2 (x2 , y2 )) then
position 2 λ = yx2 −y 1 mod p
2 −x1
19 i++ 3 x3 = λ − x1 − x2 mod p
2

20 end for 4 y3 = λ(x1 − x3 ) − y1 mod p

21 for i ← 1 to Column_P − table[i] do
5 end if
6 else if (P1 (x1 , y1 ) == P2 (x2 , y2 )) then
22 RotateDown, Column[i] of I1 by 3x 2 +a
Column_P − table[i] position 7 λ = 2y1 mod p
1
23 i++ 8 x3 = λ2 − 2 x1 mod p
24 end for
9 y3 = λ(x1 − x3 ) − y1 mod p
10 end if

G. OCCLUSION ATTACK receiver. In our analysis the cipher image generated were
Occluded data occurs because of the lack of security in occluded 6.5%,12.5%, 25% and 50%. The occluded ciph-
the channel while transferring the encrypted image to the er images were performed decryption process. The results of

VOLUME 10, 2022 87851

 L. D. Singh et al.: Encrypting Multiple Images With an Enhanced Chaotic Map

FIGURE 7. (a, c, e, g) Histogram of Combined images, CI−1 , CI−2 , CI−3 and CI−4 . (b, d, f, h) Histogram of Cipher
image, CI−1 , CI−2 , CI−3 and CI−4 .

Algorithm 4: The Encryption Process H. KEY SENSITIVITY

Input: INPUT IMAGE I1 , image size M1 × N1 , xn1 , γ 1, An ideal cryptographic scheme should have good key sensi-
δ1, xn2 , γ 2 and δ2 tivity. A slight change in the key should completely alter the
Output: CIPHER IMAGE C1 , image size M1 × N1 output of the cryptographic scheme. Fig. 8 shows the output
1 CST[I1 ] ← I1 , γ1 , δ1 , xn1 . using the original key and by altering one bit of data in the
2 for i ← 1 to M1 × N1 × 3/4 do original key.
3 Generate ASM [i] ← γ2 , δ2 , xn2
4 i++
5 end for I. DIFFERENTIAL ATTACKS
15
6 PSR ← BaseConvert[|ASM [i] × 10 |,256,8][[3;;6]] In order to prevent the differential attack, the cipher image
7 C1 =PSR XoR CST[I1 ] Cimg1 and the cipher image Cimg2 derived by some minor vari-
ation in the pixel value of the plain image. They both should
be random enough that the pixel index should not be similar.
Two methods to estimate the intensity of a differential attack
the occlusion attack are represented in Fig. 8, which is are the Number of Pixel Changes (NPCR) and the Unified
identifiable with some pixels lost. The PSNR value of the Averaged Changed Intensity (UACI). The theoretically ideal
occluded images was calculated with the original image and value of NPCR and UACI should be closed to 100% and
the decrypted image. The values were tabulated in the Table 6. 33.3% [42] respectively. The rate of pixel change of the cipher

87852 VOLUME 10, 2022

L. D. Singh et al.: Encrypting Multiple Images With an Enhanced Chaotic Map

TABLE 2. Image file used in CI−1 , CI−2 , CI−3 , CI−4 and its respective planes.

TABLE 3. PSNR and SSIM values of encrypted and decrypted image. TABLE 4. Entropy values for plane image and encrypted images.

TABLE 5. Avalanche value for cipher image on 1 bit changed in plain

image.

images Cimg1 and Cimg2 is measured by NPCR:

where,NPCR1 represents the NPCR and UACI1 represents
1 X UACI of the 512-grayscale image. The size of the images
NPCR1 = × D1 (i, j) × 100 (15) is represented by M1 × N1 , L1 represents the largest pixel
M1 × N1
i,j
that can be supported and D1 (i, j) is represented by the below
The average variation of the intensity between original expression.
image and cipher image is measured by UACI using the below (
expression. 0 if Cimg1 (i, j) = Cimg2 (i, j)
D1 (i, j) = (17)
 1 if Cimg1 (i, j) 6= Cimg2 (i, j)
 
1 X Cimg1 (i, j) − Cimg2 (i, j)
UACI1 =   (16) The value for NPCR and UACI on four different images i.e in
M1 × N1 L1
i,j the Fig. (6a, 6c, 6e, 6g) is shown in the Table 7. The NPCR and

VOLUME 10, 2022 87853

 L. D. Singh et al.: Encrypting Multiple Images With an Enhanced Chaotic Map

FIGURE 8. (a, d, g, j) plain images with the (b, e, h, k) cipher images and (c, f, i, l) is cipher image obtained with
some minor variation in pixel value of (a, d, g, j) plain images.

TABLE 6. PSNR and SSIM values of occluded image. TABLE 7. Result of the proposed approach for NPCR and UACI taken on
four sets of images.

UACI comparison with other proposed methods are tabulated is considered to be random if the p-value is more than 0.01.
in Table 10. The NIST test is performed on the cipher images shown in
Fig. (8b, 8e, 8h, 8k) using the proposed encryption method
J. CORRELATION ANALYSIS and Table 9 shows the result of the test.
The correlation of an image is computed as:
L. ATTACK ANALYSIS
Covariance [vec1 , vec2 ]
Correlation [vec1 , vec2 ] = (18) The proposed method has got a vast keyspace. So, when
ρ [vec1 ] × ρ [vec2 ] an adversary has only cipher-text information, the proposed
Here, Covariance is the covariance. ρ is the standard deviation. method will withstand the exhaustive key attack. The pro-
vec1 and vec2 are the input vectors. posed method can also withstand Known plain-text attacks.
The orientation of the horizontal, vertical and diagonal The initial keys for the enhanced chaotic map are derived
which is denoted by Htal ,Vcal and Dnal of Plane-1, Plane-2, from the Secure Hash Algorithm’s hash value. So every dif-
Plane-3 and the generated RGB image correlation coefficient ferent image that may differ just by one bit also generates a
is shown in the Table 8. The correlation coefficient along the different hash value leading to different initial values for the
Htal ,Vcal and Dnal direction of RGB generated plain image enhanced chaotic map.
and cipher image is plotted by computing with 10000 random
pixel values as shown in Fig. 10. M. COMPARISON AND DISCUSSION
The proposed multiple image encryption scheme is compared
K. RANDOMNESS ANALYSIS FOR ENCRYPTED IMAGES with some of the recent related works [16], [17], [18], [19],
The randomness of the cipher is checked by perform- [20], [34], [35], [36], [37] that encrypts multiple images
ing the NIST test on the cipher image. The proposed where each image is of image dimension 512×512. The com-
approach uses the NIST test to analyze the randomness of parison is made with respect to correlation, NPCR, UACI,
the generated cipher image. The NIST test is proposed by Entropy, Key-space, Number of images encrypted per execu-
Andrew et al. [41] and it consists of 15 different tests. For tion and execution time of encryption per image as tabulated
each and every test p-value are calculated. A cipher image in Table 10.

87854 VOLUME 10, 2022

L. D. Singh et al.: Encrypting Multiple Images With an Enhanced Chaotic Map

FIGURE 9. (a, e, i, m) Occluded cipher images with 6.25%, 12.5%, 25% and 50% occlusion.
(b, f, j, n) Decrypted Plane-1, (c, g, k, o) Decrypted Plane-2 and (d, h, l, p) Decrypted Plane-3 of occluded
cipher images.

TABLE 8. Correlation coefficient for plane image and encrypted images.

VOLUME 10, 2022 87855

 L. D. Singh et al.: Encrypting Multiple Images With an Enhanced Chaotic Map

TABLE 9. Randomness Test.

With respect to statistical analyses such as correlation (hor- Brute force attack with [18] having the maximum of 2648
izontal, vertical, diagonal), NPCR, UACI and entropy, all the and [17] having the maximum of 2128 . The proposed method
compared related works in [16], [17], [18], [19], [20], [34], has a moderate key space of 2332 . The claimed key space
[35], [36], and [37] show a comparable result. For a good of [34] is 2628 , which consists of eight initial parameters
cipher image, the correlation coefficients are expected to have derived from the hash value of SHA-256 applied on the input
a value close to zero, indicating a weak correlation between images with a precision of 1014 and a set of fixed eight exter-
the pixels in the cipher image. The proposed and compared nal parameters (c1 , c2 , . . . , c8 ). As the external parameters
methods can generate a low correlated cipher image. The are fixed and the initial parameters of the chaotic system are
NPCR and UACI data indicate that the proposed and com- derived from the hash value of SHA-256, the actual key-space
pared methods can resist differential attacks. With eight bits is 2256 . The proposed method can encipher twelve secret
representing each pixel of the cipher image, the maximum images which is more than the those given in [16], [18],
entropy value is eight. The proposed and compared methods [19], [20], [34], and [36] except for [17], [35], and [37]
can generate an entropy value of nearly eight. The key space with any i numbers of images. Theoretically, in [17], [35],
for all the compared related work is large enough to resist and [37], i numbers of images can be given as input. However,

87856 VOLUME 10, 2022

L. D. Singh et al.: Encrypting Multiple Images With an Enhanced Chaotic Map

FIGURE 10. (a, b, c) Htal ,Vcal and Dnal correlation graphs of generated RGB plane image. (d, e, f) Htal ,Vcal
and Dnal correlation graphs of encrypted image.

TABLE 10. Comparison.

a bottleneck can occur with respect to resource constraints in in Section III-C. The keys are safe from the adversary as
the computing device during actual implementation. The pro- the elliptic curve discrete logarithmic problem is one of the
posed method has a decent execution speed, faster than most complicated problems that cannot be solved in a feasible time
of the compared methods. The proposed method can decipher as of now.
the exact original images from the cipher images, while the
method given in [36] provides visually similar images but VI. CONCLUSION
not the exact original images. For the proposed method, the This paper proposed a multiple images encryption scheme
PSNR and SSIM value are ∞ and one, respectively, while based on an amplified Sine map. The ASM exhibits com-
for [36], the PSNR is in the range of 25 to 33 and SSIM plex chaotic behaviour with a more extensive chaotic range
is in the range of 0.89 to 0.97. The proposed method has and low computational cost. Two different key sets of the
the edge over other methods in reducing the iteration of a ASM is generated by using SHA384. A dynamic permutation
chaotic system. If M 1 × N 1 is the size of the image, the table is generated for the CST using the first key set of
proposed method reduces the iteration in the chaotic system ASM. The CST then performs random byte circular shifts
by a factor of four by employing a base conversion operation. along the horizontal and vertical axes of the combined input
In contrast, the compared methods iterate for M 1 × N 1 to image. Again, using the second key set of ASM, a chaotic
generate the chaotic sequence. The keys used for the proposed image is generated whose dimension is equal to that of the
method is securely shared as an elliptic curve point as given input combined image. Bitwise XoR is performed between

VOLUME 10, 2022 87857

 L. D. Singh et al.: Encrypting Multiple Images With an Enhanced Chaotic Map

the CST image and chaotic image to get the cipher image. [20] A. Sahasrabuddhe and D. S. Laiphrakpam, ‘‘Multiple images encryption
The experimental analysis and results show the effective- based on 3D scrambling and hyper-chaotic system,’’ Inf. Sci., vol. 550,
pp. 252–267, Mar. 2021, doi: 10.1016/j.ins.2020.10.031.
ness of the proposed method against various cryptographic [21] Y. Zhou, Z. Hua, C.-M. Pun, and C. L. P. Chen, ‘‘Cascade chaotic system
attacks. The performance analysis shows that the proposed with applications,’’ IEEE Trans. Cybern., vol. 45, no. 9, pp. 2001–2012,
algorithm is superior or on par with the other methods under Sep. 2015, doi: 10.1109/TCYB.2014.2363168.
[22] L. Liu, J. Lin, S. Miao, and B. Liu, ‘‘A double perturbation method for
consideration. reducing dynamical degradation of the digital baker map,’’ Int. J. Bifurca-
tion Chaos, vol. 27, no. 7, Jun. 2017, Art. no. 1750103.
[23] Y. Wu, Y. Zhou, and L. Bao, ‘‘Discrete wheel-switching chaotic system and
REFERENCES
applications,’’ IEEE Trans. Circuits Syst. I, Reg. Papers, vol. 61, no. 12,
[1] Y. Qobbi, A. Jarjar, M. Essaid, and A. Benazzi, ‘‘Image encryp- pp. 3469–3477, Dec. 2014.
tion algorithm based on genetic operations and chaotic DNA encod- [24] Y. Deng, H. Hu, and L. Liu, ‘‘Feedback control of digital chaotic systems
ing,’’ Soft Comput., vol. 26, no. 12, pp. 5823–5832, Jun. 2022, doi: with application to pseudorandom number generator,’’ Int. J. Mod. Phys. C,
10.1007/S00500-021-06567-7. vol. 26, no. 2, Feb. 2015, Art. no. 1550022.
[2] S. Huang, L. Huang, S. Cai, X. Xiong, and Y. Liu, ‘‘Novel and [25] Y. Zhou, L. Bao, and C. P. Chen, ‘‘A new 1D chaotic system for
secure plaintext-related image encryption algorithm based on compres- image encryption,’’ Signal Process., vol. 97, no. 11, pp. 172–182,
sive sensing and tent-sine system,’’ IET Image Process., vol. 16, no. 6, Apr. 2014.
pp. 1544–1557, May 2022, doi: 10.1049/ipr2.12429. [26] S. Behnia, A. Akhshani, H. Mahmodi, and A. Akhavan, ‘‘A novel algorithm
[3] W. H. Alshoura, Z. Zainol, J. S. Teh, M. Alawida, and for image encryption based on mixture of chaotic maps,’’ Chaos, Solitons
A. Alabdulatif, ‘‘Hybrid SVD-based image watermarking schemes: Fractals, vol. 35, no. 2, pp. 408–419, Jan. 2008.
A review,’’ IEEE Access, vol. 9, pp. 32931–32968, 2021, doi: [27] L. Liu and S. Miao, ‘‘Delay-introducing method to improve the dynam-
10.1109/ACCESS.2021.3060861. ical degradation of a digital chaotic map,’’ Inf. Sci., vol. 396, pp. 1–13,
[4] X. Wang and Y. Su, ‘‘Image encryption based on compressed sensing and Aug. 2017.
DNA encoding,’’ Signal Process., Image Commun., vol. 95, Jul. 2021, [28] M. Alawida, A. Samsudin, and J. S. Teh, ‘‘Enhanced digital chaotic maps
Art. no. 116246. based on bit reversal with applications in random bit generators,’’ Inf. Sci.,
[5] K. M. Singh, L. D. Singh, and T. Tuithung, ‘‘Text encryption based on vol. 512, pp. 1155–1169, Feb. 2020.
Huffman coding and ElGamal cryptosystem,’’ Recent Patents Eng., vol. 14, [29] M. Alawida, A. Samsudin, J. S. Teh, and W. H. Alshoura, ‘‘Deter-
pp. 3–8, Sep. 2020. ministic chaotic finite-state automata,’’ Nonlinear Dyn., vol. 98, no. 3,
[6] L. D. Singh and K. M. Singh, ‘‘Visually meaningful multi-image encryp- pp. 2403–2421, Nov. 2019.
tion scheme,’’ Arabian J. Sci. Eng., vol. 43, no. 12, pp. 7397–7407, [30] M. A. B. Farah, A. Farah, and T. Farah, ‘‘An image encryp-
Dec. 2018. tion scheme based on a new hybrid chaotic map and optimized
[7] J. Fridrich, ‘‘Symmetric ciphers based on two-dimensional chaotic maps,’’ substitution box,’’ Nonlinear Dyn., vol. 99, no. 4, pp. 3041–3064,
Int. J. Bifurcation Chaos, vol. 8, no. 6, pp. 1259–1284, 1998. Mar. 2020.
[8] M. Alawida, A. Samsudin, N. Alajarmeh, J. S. Teh, M. Ahmad, and [31] X. Wang, Y. Li, and J. Jin, ‘‘A new one-dimensional chaotic system with
W. H. Alshoura, ‘‘A novel hash function based on a chaotic sponge applications in image encryption,’’ Chaos, Solitons Fractals, vol. 139,
and DNA sequence,’’ IEEE Access, vol. 9, pp. 17882–17897, 2021, doi: Oct. 2020, Art. no. 110102.
10.1109/ACCESS.2021.3049881. [32] T. S. Ali and R. Ali, ‘‘A novel medical image signcryption scheme using
[9] M. Alawida, J. S. Teh, D. P. Oyinloye, M. Ahmad, and R. S. Alkhawaldeh, TLTS and Henon chaotic map,’’ IEEE Access, vol. 8, pp. 71974–71992,
‘‘A new hash function based on chaotic maps and deterministic finite 2020.
state automata,’’ IEEE Access, vol. 8, pp. 113163–113174, 2020, doi: [33] M. Alawida, J. S. Teh, A. Samsudin, and W. H. Alshoura, ‘‘An image
10.1109/ACCESS.2020.3002763. encryption scheme based on hybridizing digital chaos and finite state
[10] A. Mansouri and X. Wang, ‘‘A novel one-dimensional sine powered chaotic machine,’’ Signal Process., vol. 164, pp. 249–266, Nov. 2019.
map and its application in a new image encryption scheme,’’ Inf. Sci., [34] X. Zhang and Y. Hu, ‘‘Multiple-image encryption algorithm
vol. 520, pp. 46–62, May 2020. based on the 3D scrambling model and dynamic DNA coding,’’
Opt. Laser Technol., vol. 141, Sep. 2021, Art. no. 107073, doi:
[11] A. Mansouri and X. Wang, ‘‘A novel one-dimensional chaotic map gener-
10.1016/j.optlastec.2021.107073.
ator and its application in a new index representation-based image encryp-
[35] M. Hanif, R. A. Naqvi, S. Abbas, M. A. Khan, and N. Iqbal, ‘‘A novel and
tion scheme,’’ Inf. Sci., vol. 563, pp. 91–110, Jul. 2021.
efficient 3D multiple images encryption scheme based on chaotic systems
[12] M. Xu and Z. Tian, ‘‘A novel image cipher based on 3D bit matrix and
and swapping operations,’’ IEEE Access, vol. 8, pp. 123536–123555, 2020,
Latin cubes,’’ Inf. Sci., vol. 478, pp. 1–14, Apr. 2018.
doi: 10.1109/ACCESS.2020.3004536.
[13] Z. Hua, Z. Zhu, Y. Chen, and Y. Li, ‘‘Color image encryption using
[36] A. Gaffar, A. B. Joshi, S. Singh, V. N. Mishra, H. G. Rosales, L. Zhou,
orthogonal Latin squares and a new 2D chaotic system,’’ Nonlinear Dyn.,
A. Dhaka, and L. N. Mishra, ‘‘A technique for securing multiple dig-
vol. 104, no. 4, pp. 4505–4522, May 2021.
ital images based on 2D linear congruential generator, silver ratio,
[14] Z. Hua, Y. Chen, H. Bao, and Y. Zhou, ‘‘Two-dimensional parametric poly- and Galois field,’’ IEEE Access, vol. 9, pp. 96125–96150, 2021, doi:
nomial chaotic system,’’ IEEE Trans. Syst., Man, Cybern., Syst., vol. 52, 10.1109/ACCESS.2021.3094129.
no. 7, pp. 4402–4414, Jul. 2022, doi: 10.1109/TSMC.2021.3096967. [37] X. Gao, J. Mou, S. Banerjee, Y. Cao, L. Xiong, and X. Chen, ‘‘An effective
[15] Z. Hua, Y. Zhang, H. Bao, H. Huang, and Y. Zhou, ‘‘n-dimensional multiple-image encryption algorithm based on 3D cube and hyperchaotic
polynomial chaotic system with applications,’’ IEEE Trans. Circuits map,’’ J. King Saud Univ.-Comput. Inf. Sci., vol. 34, no. 4, pp. 1535–1551,
Syst. I, Reg. Papers, vol. 69, no. 2, pp. 784–797, Feb. 2022, doi: Apr. 2022, doi: 10.1016/j.jksuci.2022.01.017.
10.1109/TCSI.2021.3117865. [38] N. Koblitz, ‘‘Elliptic curve cryptosystems,’’ Math. Comput., vol. 48,
[16] X. Zhang and X. Wang, ‘‘Multiple-image encryption algorithm based on no. 177, pp. 203–209, 1987.
DNA encoding and chaotic system,’’ Multimedia Tools Appl., vol. 78, no. 6, [39] M. Miller, ‘‘Uses of elliptic curves in cryptography,’’ in Advances
pp. 7841–7869, 2019. in Cryptology—CRYPTO. Berlin, Germany: Springer-Verlag, 1986,
[17] R. Enayatifar, F. G. Guimarāes, and P. Siarry, ‘‘Index-based permutation- pp. 417–426.
diffusion in multiple-image encryption using DNA sequence,’’ Opt. Lasers [40] The USC-SIPI Image Database. Accessed: May 19, 2020. [Online].
Eng., vol. 115, no. 3, pp. 131–140, 2019. Available: http://sipi.usc.edu/database/
[18] M. Zarebnia, H. Pakmanesh, and R. Parvaz, ‘‘A fast multiple-image [41] R. Andrew, S. Juan, N. James, S. Miles, and B. Elaine, ‘‘A statistical test
encryption algorithm based on hybrid chaotic systems for grayscale suite for random and pseudorandom number generators for cryptographic
images,’’ Optik-Int. J. Light Electron. Opt., vol. 179, pp. 761–773, applications,’’ Nat. Inst. Standards Technol., Gaithersburg, MA, USA,
Jan. 2019. Tech. Rep., 2010.
[19] K. A. K. Patro, A. Soni, P. K. Netam, and B. Acharya, ‘‘Multiple grayscale [42] Y. Wu, J. P. Noonan, and S. Agaian, ‘‘NPCR and UACI randomness
image encryption using cross-coupled chaotic maps,’’ J. Inf. Secur. Appl., tests for image encryption,’’ Cyber J., Multidisciplinary J. Sci. Technol.,
vol. 52, Jun. 2020, Art. no. 102470. J. Select. Areas Telecommu., vol. 1, pp. 31–38, Apr. 2011.

87858 VOLUME 10, 2022

L. D. Singh et al.: Encrypting Multiple Images With an Enhanced Chaotic Map

LAIPHRAKPAM DOLENDRO SINGH (Mem- KHOIROM MOTILAL SINGH received the

ber, IEEE) received the master’s degree from the B.Tech. and M.Tech. degrees from the Depart-
National Institute of Technology Agartala, India, ment of Computer Science and Engineering,
and the Ph.D. degree from the National Institute National Institute of Technology Manipur, India,
of Technology Manipur, India. He is currently an in 2014 and 2016, respectively, and the Ph.D.
Assistant Professor with the Department of Com- degree from the National Institute of Technol-
puter Science and Engineering, National Institute ogy Nagaland, India, in 2022. He is currently
of Technology Silchar, Assam, India. He has pub- an Assistant Professor with the Department of
lished several research papers in reputed journals. CSE, Koneru Lakshmaiah Education Foundation,
He has published two patents. His research inter- Vaddeswaram, Andhra Pradesh, India. He has
ests include cryptography and cryptanalysis. He is a Life Member of Cryp- published one patent.
tography Research Society of India. URL: http://cs.nits.ac.in/dolendro/.
MOATSUM AL AWIDA received the Ph.D. degree
in computer science/cybersecurity (cryptography)
from the School of Computer Sciences, Univer-
siti Sains Malaysia, in 2020. He is an Assis-
tant Professor of cybersecurity with Abu Dhabi
University. He published more than 25 articles
in high impact factor journals. In 2021, he was
ranked among the top 2% of all scientists in
the world. His research interests include chaotic
systems, chaos-based applications, multimedia
security, blockchain, cybersecurity, quantum-based cryptography, and cryp-
ROHIT THINGBAIJAM received the B.Tech. tography. He has served as a Referee for some renowned journals, such
degree from the Department of Computer Science as IEEE TRANSACTIONS ON CYBERNETICS, Signal Processing, Information Sci-
and Engineering, National Institute of Technol- ences, Journal of Information Security and Applications, IEEE ACCESS,
ogy Manipur, India, in 2019, and the M.Tech. Wireless Personal Communications, the International Journal of Bifurca-
degree from the Department of Computer Science tion and Chaos in Applied Sciences and Engineering, Optik, Optics and
and Engineering, National Institute of Technology Laser Technology, IET Information Security, IET Image Processing, Security
Silchar, Assam, India. and Communication Networks, Chaos, Solitons and Fractals, Physica A:
Statistical Mechanics and its Applications, and Signal Processing: Image
Communication.

VOLUME 10, 2022 87859