IRJET- Efficient Image Encryption with Pixel Scrambling and Genetic Algorithm

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 06 Issue: 04 | Apr 2019 www.irjet.net p-ISSN: 2395-0072
© 2019, IRJET | Impact Factor value: 7.211 | ISO 9001:2008 Certified Journal | Page 3241
Efficient Image Encryption with Pixel Scrambling and Genetic
Algorithm
Surya R1, Premkumar R2
1PG scholar, Department of ECE Mount zion College of Engineering and Technology, Tamilnadu, India
2Assistant professor, Department of ECE, Mount zion College of Engineering and Technology, Tamilnadu, India
---------------------------------------------------------------------***----------------------------------------------------------------------
Abstract - Advanced image managing, as a PC based
innovation, visual data plays an undeniably essential job in
numerous parts of our day by day life, to them from
unapproved parties are increasingly vital. In this article an
effective picture encryption with GA is proposed. The
proposed strategy utilizes a picture encryption/decoding
calculation dependent on random shuffling of pixels and
permutation, substitution process. Earlier stage picture is
partitioned and pixels are revised to make partiallyencoded
picture. At last Permutation and substitution task are
performed. Prior stage picture is divided and pixels are
rearranged to make in part encoded picture. At last
Permutation and substitution task are performed.
Examinations are performed to watch the security of the
framework.
Key Words: Block crossover, bitplane, mutation, genetic
algorithm, scrambling
1. INTRODUCTION
Nowadays, picture encryption is among rapidly creating
headways. It shapes focus inquire about zone inside
structuring and programming designing controls also.
Encryption is the way of transforming the information to
make sure its security. With the rapid development of web
and computer network, a huge amount of computerized
information is being exchanged over different kinds of
systems. The protection of digital information including
pictures has involved more attention and various picture
encryption techniques have been planned to improve the
security of these pictures. Any picture encryptiondependent
on pixels framework is partitioned into two strategies: pixel
replacement techniquesand pixel scramblingstrategy.In the
pixel replacement technique, everypixel inthe pictureneeds
to change its values and this method also called us
substitution. But, in the pixel modification scheme, the pixel
needs to change its position and this technique is called
confusion.
The encryption process and the decryption process
explained in [1] exactly same. They similarly compriseofthe
comparable activities of plaintext-related scrambling once,
dispersion twice and grid pivoting of 180 degrees multiple
times [1]. Adaptive wavelet transform in frequency domain
used for data hiding and genetic algorithm also used to
encrypt the image in paper [2]. Usingthetravellingsalesman
algorithm the s box process is changed, and this s box is
supported on ga. S box is an essential constituent in
encryption method is filed in [3]. Cipher images aremadeup
of using the chaotic function. After that GA initialize the
population and in each stage the respond is gained from
earlier stage iteration to produce a most excellent encrypted
image is explained in [4].
Clients have the adaptability to pick an) any current or
recently created picture as source picture b) any
deterioration strategy for producing the bitplane c) any
disintegrated bitplanes as the precautions key bitplane d)
any scrambling technique for the bit dimension change are
designed in [5]. A plausibility reproduction for separate the
circulations of picture contrasts dependent on bitplane
imaginable potential and carvings among bitplanes.Analyze
through the present picture disparity portrayals which
envision the disseminations incorporate itemized setups.
This model offered a group parametric showing that can be
utilized to shape arbitrary conveyances with no usage a few
careful restrictions on the appropriations [6].
The regular fractal measurement process is diminished in
sectioning lopsided substance beginning multifaceted
surroundings since it can't separate the nearby grayscale
among the substance and the composite surroundings
additionally divergent sizes of the substance, whichareeasy
to starting point under-division. To get better division
impact, an improved picture division technique dependent
on bit-plane and morphological remaking is proposedin[7].
Three vital contributions are presented in [8] a) reason for
critical bitplanes, b) encryption of justthereal biplanesmost
imperative to diminish in computational intricacy and c)
rejection of the required for part channel for pass on the
succession by means of reverence to the major bitplanes.
Another strategy to use a security framework for
correspondence of advanced stuffing over network systems
[9]. Stretch out change process based ensured manyshading
picture encryption framework which is totally unique than
the nearly utilized a few picture encryption framework [10].
Two-dimensional unadulterated picture is changed to one
measurement. From that point onward, with the motivation
behind trim down the exchanging procedure time, stageand
dispersion ventures for a few pixels are completing in the
comparative time [11]. There are six dissimilar permutation
methods based on confused and non messy producers are
explained [12]. The stage is accomplished by doughpuncher

map and the substitution by a key identified with plain
picture calculation dependent on the altered Logistic guide.
Adjustment of the Logistic guide is advanced to support the
opportunity of the encryption key, and along these lines
support the security [13]. In paper [14] contains fourstages:
dissemination, substitution, dispersion and change.
Dissemination is finished by bitwise XOR activity and new
clamorous guide, substitution is finished by solid S boxes,
third technique is finished by calculated guide are
advertised. In conclusion a square stage is capable by a
change work [14]. Thearrangedstrategymakesutilizationof
coupled cross section guide to deliver the measure of
ensured figure pictures as essential position of changed
hereditary calculation. What's more, after that it relates the
changed hereditary calculation to both lift the entropyofthe
encryption and drop off the computational time [15]. The
calculated guide is use to make number of veil and GA is
utilized to locate the most magnificent cover for encoding
[16].
Change is finished by round move by lines and sections and
substitution is finished by legitimate administrators. This
framework is awfully sympathetic to each key and it
reproduces the picture well great quality [17]. A irregular
arrangement is utilized to locate the beginning spotposition
in the wave and produces a dissemination vehicle for
measured procedure [18]. This technique is isolated into
three sections: mystery key age, picture change and
dissemination. Mystery keys contains of two phases. One is
arbitrary keys and another one is hash clusterkeys[19].The
plain image is transformed into two binary sequences of the
same size. Subsequently, a fresh diffusion stratagem is
initiated to circulate the two sequences reciprocally [20].
2. PROPOSED SYSTEM
To begin with two color image as input of the system. The
plain images are segmented using split and merge
segmentation. Then the segmented blocks are shuffled
randomly. Again the shuffled blocks pixels are randomly
scrambled after that block permutation is performed. The
permuted image decomposed into eight planes and then
combined finally hybrid substitution is performed. Three
types of substitution areperformedhere;theyaretranspose,
swap, and circular shift.Using3dlogistic mapkeygeneration
is performed.
Fig. 1 Encryption system
2.1 Segmentation
Segmentation is the progression of separating an image into
having an important effect part, here we using a split and
merge segmentation. We perform different block sizetosplit
the image. In this analysis we came to know that splitting
small size of blocks are gave more efficient results.
Fig. 2 Divided image
2.2 Random shuffling of pixels
Scrambling affords excellent security to the system.
Scrambling is nothing but change the pixels of an image
randomly. We have donewith4x4,8x8and16x16.Outcomes
shown in fig.2

(a)
(b)
(c)
Fig. 3(a) Scrambling with size of 4x4 (b) Scrambling with
size of 8x8 (c) Scrambling with size of 16x16
2.3 Bitplane slicing algorithm
With the help of this algorithm we separatetheLSBandMSB
of the image. And the n combine the LSB and MSB to get
partial image with random noise.
Fig. 4 Bit plane slicing of Scrambled image
2.4 Genetic algorithm realization
Two main operators are in GA. They are crossover and
mutation. In first stage block swapping permutation is done
and in mutation three types are hybrid.
The hybrid administrator is practically equivalent to
generation and natural hybrid. In this more than one parent
is chosen and at least one off-spring is delivered utilizingthe
hereditary material of the guardians. Hybrid is generally
connected in a GA with a high likelihood.
P1 P2
C1 C2
Fig. 5 Block crossover using GA

2.5 Mutation
In basic terms, change might be characterized as a little
irregular change in the chromosome, to get another
arrangement. It is utilized to remain awake and present
assorted variety in the hereditary populace and is normally
connected with a low likelihood – pm. On the off chance that
the likelihood is high, the GA gets decreased to an irregular
hunt.
Change is the piece of the GA which is identified with the
"investigation" of the pursuit space. It has been seen that
transformation is fundamental totheinterminglingofthe GA
while hybrid isn't. three types of mutation.
 Random Swap
 Transpose
 Circular shift
1 2 3
4 5 6
7 8 9
Fig. 6 Random swap
1 2 3
4 5 6
7 8 9
Fig. 7 Transpose
1 2 3
4 5 6
7 8 9
Fig. 8 Circular shift
(a) (b) (c)
Fig. 9 Mutation results. (a) Circular shift mutation (b)
Transpose mutation (c) Flip mutation.
3. EXPERIMENTAL ANALYSIS
The planned encryption algorithm is capable of encrypt
dissimilar sizes of images. This part affords testing
consequences to demonstrate its encryption presentation.
Every single one testing’s executed with MATLAB2010a ina
mainframe with the configurations of Intel(R) Pentium(R)
CPU 2.10 GHz; with 2 GB RAM, and Microsoft Windows 7
operation system. Our arrangement has two CPUs and one
GPU.
Fig. 10 Plain image and Encrypted, decrypted image
3 2 1
5 9 6
7 8 4
1 4 7
2 5 8
3 6 9
3 6 9
2 5 8
1 4 7

3.1 Histogram
We can obtain the allocation of pixel values of a figure from
its histogram. If it is not smooth enough, a little quantity of
sequence able to be leaked by the numerical attack.
Therefore, a uniform and even allocation is attractive for a
high-quality encryption method. The histograms of plain
image and its encrypted image by the planned algorithm are
shown in Fig. 6. It is understandable that histogram of the
encrypted image is unvarying and considerably unlike from
that of the plain image. So our algorithm able to build the
statistical attack unacceptable.
Fig. 11Histogram of original image and encrypted
3.2 Correlation analysis
Correlation analysis is worn to experiment the connections
of adjoining pixels in the original image and encrypted
image. 3000 pairs of adjacent pixels from the primary
picture and scrambled picture are chosen arbitrarily in the
event, vertical and corner to corner headings to execute this
examination. Figure 7 plots the correlations of two
adjacent pixels in three directions for Lena.
Fig .12 Comparisons of correlation coefficients
Fig. 13 Correlation of two adjacent pixels for standard
image
Fig .14 Correlation coefficients of two adjacent pixels in
plain image and cipher image

3.3 Information entropy
The information entropy able to be designed by,
H(m) =
In which H(m) represents the information entropy of an
sequence source m, x(mi) denotes the probability of symbol
mi.
Fig .15 Information entropy of original and encrypted
image
3.4 NPCR and UACI
In the discrepancy harass, a minor variation is
prepared to the original image, and the proposed algorithm
is engaged to encrypt the plain image earlier than and
behind this transform. These twocipheredimageshavebeen
evaluated to discover every achievable connection between
the plain image and the ciphered image. Unified Average
Changing Intensity(UACI)andNumberofPixelsChangeRate
(NPCR) are two criterion generally used by researchers to
inspect the differential attack resistance of any encryption
algorithm. The NPCR and UACI are shown as Eqs.
Subjected to:
where M and N denote height and width of the image,
respectively. C 1 and C 2 are two encrypted-images with one
pixel difference. Fig .12 lists the best and average NPCR and
UACI values.
Fig .16 Security comparisons between different encryption
method by number of pixel change rate (NPCR) and
unified average changing intensity (UACI)
3.5 Speed analysis and performance comparison
Running speed is an important characteristic
parameter for encryptionalgorithms, whenthesecuritylevel
may meet therequirements.Genetic-basedimage encryption
scheme is mostly composed of crossover process and
mutation process. In the proposed algorithm, block
swapping and hybrid mutation process is used. This leads to
a speed advantage when compared with other algorithms.
Fig .17 Encryption speed comparison

Fig .18 Performances of the proposed scheme and five
comparable methods for (256 x 256)
4. Conclusion
This paper suggested an encryption based on bitplanedecay
with block permutation of pixels. In our system original
image is split using segmentation algorithm. And then
scrambling pixels are in first stage. Then the scrambled
image is sliced into eight bitplanes using bitplane slicing
algorithm, and swapping its blocks randomly then and
mutation operation is completed. To increase the security of
the system block swapping crossoverandhybridmutationis
performed. Some analysis are completed to validate the
system rate and safety.
REFERENCE
1. Yong Zhang, Yingjun Tang(2018), ‘A Plaintext-
Related Image Encryption Algorithm Based on
Chaos’ , Multimedia Tools and Applications, Vol. 77,
No. 6, pp. 6647-6669.
2. Aref Miri, Karim faez(2017), ‘Adaptive Image
Stenography Based on Transform Domain via
Genetic Algorithm’, Optik – International Journal
For Light and Electron Optics,Vol.145,pp.158-168.
3. Yong Wang, Kwok-Wo Wong, Changbing Li, Yang
Li(2012), ‘ A Novel Method to Design S-box Based
on Chaotic Map and Genetic Algorithm’, Physics
Letters A, Vol. 376, No. 6-7, pp. 827-833.
4. Abdul Hanan Abdullah, Rasul Enayatifar, Malrey
Lee(2012), ‘ A Hybrid Genetic Algorithm and
Chaotic Function Model For Image Encryption’,
International Journal of Electronics and
Communication,Vol. 66, No. 10, pp. 806-816.
5. Yicong Zhou, WeijiCao, Philip Chen C.L (2014),
‘Image Encryption Using Binary Bitpane’ Signal
Processing, Vol. 100, pp. 197-207.
6. Choy S.K, Kevin Yuen, Carisa Yu (2018), ‘ Fuzzy Bit
plane dependence Image Segmentation’, Elsevier –
Signal Processing, Vol. 154, pp. 30-44.
7. Yanfen Lin, Liuxi wu (2018), ‘Improved Abrasive
Image Segmentation Method Based on Bit-plane
And Morphological reconstruction’, Springer –
Multimedia Tools and Applications, pp. 1-14.
8. Sukalyan som, Abhijt Mitra, Sarbani Palit,
B.B.Chaudhuri (2018), ‘A Selective Bitplane Image
Encryption Scheme Using Chaotic Maps’, Springer –
Multimedia Tools and Applications, pp. 1-28.
9. Ranjeet Kumar singh, Binod Kumar, Dilip Kumar,
Danish Ali Khan(2018), ‘ Level by Level Image
Compression-Encryption Algorithm Based On
Quantum Chaos map’, Journal of King Saud
University – Computer and Information Sciences,
Vol. 75, No. 22, pp. 14685-14706.
10. Abimanyu Kumar Patro K, Bibhudedra Acharya
(2018), ‘Secure Multi-level Permuation Operation
Based Multiple Colour Imge Encryption’, Elsevier –
Journal of Information Security and Applications,
Vol. 40, pp. 111-133.
11. Rasul Enayatifar, Adul HananAbdullah,Ismail Fauzi
Isnin, Ayman Altmeen (2017), ‘ Image Encryption
Using A Synchronous Permutation-diffusion
technique’, Elsevier – Optics and LaserEngineering,
Vol. 90, pp. 146-154.
12. Salwa K. Abd –El-Hafiz, Sherif H.AbdElHaleem,
Ahmed G.Radwan (2016), ‘Novel Permutation
MeasuresForImageEncryptionAlgorithm’,Elsevier
– Optics and Laser Engineering, Vol. 85, pp. 72-83.
13. Ramadan Noha, Ahmed HossamEldin H, El-khamy
said, Abdul El-samie fathi E (2017), ‘ Permutation-
substitution Image encryption based on a modified
chaotic map in transform domain’, Springer –
Journal of Central South University, Vol. 24, No. 9,
pp. 2049-2057.
14. Akram Belazi, Ahmed A. Abd El-Latif, Safya Belghith
(2016), ‘A Novel Image Encryption Scheme Based
On Subsitution-Permutation Network and Chaos’,
Signal Processing, Vol. 128, pp. 155-170.
15. Hossein Nematzadeh, Rasul Enayatifar, Homayun
Motameni, Frederico Gadelha Guimmaraes, Vitor
Nazario Coelho (2018), ‘Medical Image Encryption
Using a Hybrid Model Of Modified Genetic

Algorithm and Coupled Map Lattices’, Optics and
Laser Engineering, Vol. 110, pp. 24-32.
16. Rasul Enayatifar, Abdul Hanan Abdullah, Ismail
Fauzi Isnin (2014), ‘Chaos-Based Image Encryption
Using A Hybrid Genetic Algorithm And A DNA
Sequence’, Elsevier – Optics and Laser Engineering,
Vol. 56, pp. 83-93.
17. Zahra Parvin, Hadi Seredarabi, Mousa Shamsi
(2014), ‘A New Secure And Sensitive Image
Encryption Scheme Based On New Substitution
Chaotic function’, Springer – Multimedia tools and
Applications, Vol. 75, No. 17, pp. 10631-10648.
18. Guodong Ye (2014), ‘A Block Image Encryption
Algorithm Based On Wave Transmission and
Chaotic Systems’, Springer – Nonlinear Dynamics,
Vol. 75, No.3, pp, 417- 427.
19. Mingxu Wang, Xingyuan Wang, Yingqian Zhang,
Zhenguo Gao(2018), ‘ A Novel Chaotic Encryption
Scheme Based On Image SegmentationandMultiple
Diffusion Models’, Opticsand LaserTechnology,Vol.
108, pp. 558-573
20. Lu Xu, Zhi Li, Jian Li, Wei Hua (2015), ‘A Bit-Level
Image Encryption Algorithm Based on Chaotic
Maps’ Optics and Lasers in Engineering, Vol. 78, pp.
17-25
BIOGRAPHIES
Surya R, BE(Electronics and
communication engineering) –
Anna university, Trichy,
ME(communication systems) –
Mount zion college of engineering
and technology, Pudukottai.
R. Premkumar was born in india
1984. He received a B.E degree in
Electronis and Communication
Engineering from anna university
Chennai in 2005. He received the
M.E degree in VLSI Design from
Anna university, Chennai in 2017.
He is currently submitted the Ph.d
thesis at anna university, Chennai.
His area of interest are media,
security, image Processing, vlsi
design, cryptography. HeistheLife
time member ISTE,IETE and CSI

IRJET- Efficient Image Encryption with Pixel Scrambling and Genetic Algorithm

More Related Content

What's hot (16)

Similar to IRJET- Efficient Image Encryption with Pixel Scrambling and Genetic Algorithm (20)

More from IRJET Journal (20)

Recently uploaded (20)

IRJET- Efficient Image Encryption with Pixel Scrambling and Genetic Algorithm