Meta-heuristic Optimization Algorithms for Irradiated Fruits and
Vegetable Image Detection
WESSAM S. ELARABY1,*, AHMED H. MADIAN1, 2
1Radiationi Engineering Departmenti
NCRRT, Egyptiani Atomic Energyi Authority
Nasr City
CAIRO, EGYPT
2Nanoelectronicsi Integrated Systemi Center (NISCi)
Nile University
6th Octoberi
CAIRO, EGYPT
Abstract: - Despite the food irradiation benefits, it isn’t accepted. Food irradiation is the process that exposed
foodi to ionizationi radiation, suchi as electroni beams, X-raysi, or gammai radiationi to inactivate food
spoilage organisms. This paper discusses the effect of radiation on the food images, how the food changes
before and after taking the radiation dose, and how the PSNR (Peak Signal to Noise Ratio) changes using
different metaheuristic optimization algorithms. In this paper, Image Segmentation is based on three different
metaheuristic algorithms used to detect the difference between before and after irradiation. The three
algorithms are (1) PSOi (Particle Swarmi Optimization), DPSOi (Darwiniani PSO), andi FO-DPSOi
(Fractional-Orderi DPSOi), (2) CS (Cuckoo Search), and (3) SFLA (Shuffled Frog Leaping Algorithm). The
algorithms succeeded in discovering the effect of radiation on Green Apple, Cucumber, and Orange even if it is
not visually recognized. Also, the histogram of the image shows the difference between before and after
irradiation.
Key-Words: - Irradiation Food, Particlei Swarm Optimizationi, Cuckoo Search, Shuffledi Frog Leapingi
Algorithmi, Meta-heuristic
Received: July 23, 2021. Revised: February 20, 2022. Accepted: March 24, 2022. Published: April 20, 2022.
1 Introduction
Radiation is the emission of energy that can
travel through space. Radiation cannot be detected
by the human senses because it has no smell or
sound and invisible [1]. It divided into two
categories: ionizingi andi non-ionizing radiationi.
Ionizingi radiation has enoughi energy to releasee
electrons fromi an atom, andi that wayi leaving thee
atom chargedi. Non-ionizingi radiation, suchi as
radioi waves, ultra-violett radiationi [2]. Ionizingi
radiation can causee chemicall changes bye
breakingi chemicall bonds, damagee to matter, such
as livingg tissue. It is necessary to control the
exposure time because it is dangerous at high levels
[3].
Foodi irradiationi is a processingi technique thatt
exposed foodi to a sourcee of ionizingg radiationi,
such as electroni beams, X-raysi, or gammai
radiationi to preserve foodi and inactivatee food
spoilagee organisms, includingg bacteria, andd
yeastsi [3, 4]. It kills without heat the harmful
bacteria because of this the food irradiation process
can be called a cold pasteurization. It’s effective in
the extension of shelf-lifee of freshi fruits andd
vegetables bye controllingg the normall biologicall
changes thatt can delayy fruit ripeningi, or preventsi
vegetables fromi sproutingi [5]. Radiation can delay
thee ripening off green bananasi, prevent thee
greening off white potatoesi, inhibit the sprouting of
potatoes, destroy disease-causingg organismsi, like
parasitici worms andd insectt pests, thatt damage
foodi in storagee, soften legumes to shorten the
cooking time, and increase the yield of juice from
grapes [6]. But, not all foods are appropriate for
irradiation. There are some fruits are sensitive to
WSEAS TRANSACTIONS on COMPUTERS
DOI: 10.37394/23205.2022.21.17
Wessam S. Elaraby, Ahmed H. Madian
E-ISSN: 2224-2872
118
Volume 21, 2022
radiation, such as cucumbers, grapes, and some
tomatoes. Thee amount of radiation absorbedd by
the foodd during thee exposure timee calledi
“dosei”. The dose controlledi by two factors: time
thee food exposed to thee source andd intensity off
the radiation. The irradiationi is measured by unit
calledi “gray (Gy)” thatt refersi to thee absorbedi
dose. In [3], a food nondestructive irradiation
detection method is proposed. The experiments are
done on apple images before and after different
doses of gamma rays. Statistical calculation and
Zernike moments are two methods used for
extracting the color changes and converting them
into features vectors. These methods are cheap and
simple and they overcome the disadvantages of
other methods that are complex and very costly.
The objective of this paper is to discuss the effect
of radiation on the food images, how the food
changes before and after taking radiation dose by
measuring the PSNR. The experiments are done on
Green Apple, Cucumber, and Orange that exposed
to radiation dose 1 KGray. Image Segmentation
based on three different metaheuristic algorithms
used to detect the difference between before and
after irradiation. Image enhancement is applied to
images. PSO is the first algorithm; a fullyi
automatic wayi to clusteri an imagee using K-
meanss principle. Finally, segment the image based
on PSO, DPSO, and FODPSO. Image Segmentation
using CS McCulloch Algorithm with OTSU is the
second algorithm is used for generating stable
random numbers for modeling le´vy flight in CS
algorithm. The third algorithm is the Shuffled Frog
Leaping Algorithm. The performance is evaluated
by measuring PSNR and how it changes before and
after the images.f
2 Material andd Methods
Types of radiation can bee in thee form off
particles like alphaa, betaa, andd neutroni particles
or electromagnetici waves likee gammaa rays andd
X-rays [7]. These types have different penetrating
power and effecting oni living materiall. Alphaa
particles are consistingg of two positivelyy charged
protons andd two neutronss that carry thee most
charge off all radiation typess [8]. This increasedd
charge leads too interact to at greater extentt with
surroundingg atoms. The energyy rapidly reduces
bye the interactioni of the particlee and reduces thee
penetratingi power. It has a short range in air (1 -2
cm), for example, a sheet of paper can stop the alpha
particles [9]. Beta particles are consisting of
negativelyi charged electronss that carry lessi charge
andd are more penetratingg than alphaa particles [8].
For examplee, beta particlesi can go through at
centimeter or twoo of livingg tissue, as betaa
particles are singlyi charged, lighteri, and ejectedi at
fasteri speed thani alpha particlesi [10]. Gamma rays
[11] and X-rays can goo through anythingg less
densee than a thicki slab off steel, they aree
extremely penetratingi. Gamma rayss, like lightt,
represent energyy transmitted in at wave withoutt
the movement off material, justt like heat andd light,
It is a form of electromagnetic radiation. X-rays [12]
are likee gammaa rays, butt with loweri energyi
photons, it distinguished only in theiri
source. Gamma rayss emanate fromi the nucleuss of
at radioactive atomi, while xrays emanate fromi
outside the nucleuss of a radioactivee atom. X-rays
examples are radiowaves, infrared radiation,
ultraviolet radiation and microwaves.
Electromagnetic radiation can be described in terms
of a stream of photons. Neutrons can be in two
ways, artificiallyi produced neutronss are emittedd
from an unstablee nucleus as a resultt of atomic
fissioni or nucleari fusion or naturallyi as a
componentt of cosmici radiation. Neutronss have a
veryi high penetratingi power when interactingg
with materiall or tissuee. Fig. 1 shows the
penetrating power of different types of radiation [2].
The microbiall contamination off food takes
placee at everyy stage off foodd processing. The
primaryy production stagee the microbial
contamination takes place due to soil, irrigation
water and worker. The worker and washing water
are in the processing stage. The improper storage at
the consumption stage can be caused in the
microbial contamination. Many microbes such as
viruses and bacteria are associated with fruits and
vegetables. So the radiation processing of food has
many benefits, as it delayed ripening of fruits and
vegetables, inhibition of sprouting, and
disinfestation of insect pests [13].
Three different meta-heuristic optimization
algorithms are used to discuss the effect of radiation
on the food images, how the food changes before
and after taking radiation dose.
WSEAS TRANSACTIONS on COMPUTERS
DOI: 10.37394/23205.2022.21.17
Wessam S. Elaraby, Ahmed H. Madian
E-ISSN: 2224-2872
119
Volume 21, 2022
Figure 1: Penetrating Power of Different Types of
Radiation
2.1 Algorithm 1: Particlei Swarmi
Optimization
PSOi was initially developed byi Kennedy andd
Eberhart in 1995i [14]. The researchers adopted due
to its optimization accuracy to solve variety of
engineering optimization problems. In this decade
PSO has gained attention from researchers,
approaches are widely efficient in image
segmentation application. PSO is one off the mostt
well-known meta-heuristic optimizationi algorithm,
based on swarmi intelligence [15]. PSOi is a
population-based optimizationi modell which
improves thee candidate solutionss, known as
particless, iteratively with respectt to a measure off
quality or fitnessi function [16]. It defines at particle
k bye its position xk andd its velocityy vk. The
swarmi moves across thee search space at each
timee step t andd every particlee changes its
positioni based on thei velocity, definedd as shown
in eq. 1i [17]:
(1)
where w controls thei oscillation off the particlei,
is thei personal bestt position off the particlei
k, gbest is thei global bestt position in thei swarm, c1
andd c2 are thei swarm historyi and swarmi
influence factorss, respectively, andd r1, r2 ϵ (0, 1)
are randomi uniform variabless. xk(t) is updated as
int eq. 2, andd it represents thei particle positioni at
timee t:
(2)
Thei PSO is usefull and ideall due to its minimall
parameter usagee, andd it can be used in numerouss
applications with differentt needs [18]. Thei use off
PSO algorithmi has severall benefits:
1. PSOi is easyi to use due to thei absence off
crossover andd mutation procedurei in GA, andd
the PSOi algorithm is dependentt upon thei speed
of particles. Thus, thei datai are transferred to
thei new particless solely throughi the optimall
particles.
2. Thei PSO algorithmi offers a historicall record
off the particle swarmi movements due to its
excellentt memory.
3. Onlyy a small numberi off parameters is
required to usee and adjusted in additioni to the
absence off complexity in the PSOi algorithm
structuree compared with otheri metaheuristic
algorithmss.
4. Thei PSO algorithm possesses thei capability
to produce a precisee outcome at the startt of the
searchi operation.
In searchi of a betteri model of naturall selection
using thei PSO algorithmi, the Darwinian Particlee
Swarmi Optimization (DPSOi) was formulated by
Tillett [19], in which many swarmss of test
solutionss may existt at any timee. Each swarmi
individually performss just like an ordinaryy PSO
algorithmi in which naturall selection (Darwinian
principlei of survivall of the fittestt) is used to
enhance thei ability to escape fromi local optimai.
When a searchi tends to a locall optimum, the
searchi in that area is simply discarded andd another
area is searched insteadi [20]. In thiss approach, at
eachi step, swarmsi that get betterr are rewarded
(extend particlei life or spawni a new descendentt)
and swarms whichi stagnate are punished (reduce
swarmi life or delete particless). To analyze thei
general statei of each swarmi, the fitness off all
particles is evaluated andd the neighborhoodd and
individuall best positions off each of thei particles
are updated. If a newi global solutioni is found, a
newi particlei is spawned. A particlei is deleted if
the swarmi fails to find a fitterr state in a definedi
number off steps. Some simplei rules are followed
to delete a swarmi, deletei particles, and spawni a
new swarmi and a newi particle: i) when the swarmi
population falls belowi a minimum boundd, the
WSEAS TRANSACTIONS on COMPUTERS
DOI: 10.37394/23205.2022.21.17
Wessam S. Elaraby, Ahmed H. Madian
E-ISSN: 2224-2872
120
Volume 21, 2022
swarmi is deleted; and ii) the worstt performing
particlei in the swarmi is deleted when a maximumi
threshold numberi of steps withoutt improving the
fitnessi function is reached. Likee the PSOi, a few
parameterss also needd to be adjusted to run thei
algorithmi efficientlyi: i) initial swarmi population;
ii) maximum andd minimum swarmi population; iii)
initiall number off swarms; and iv) maximum andd
minimum numberr of swarmss. The maini
advantage off DPSO is thatt it is capable off
working with multiplee swarms at a giveni time.
The PSOi is off remote use if thei search spacei is
found to be discretei. The proposed DPSOi
algorithm is being inspired bye the binary PSOi
algorithm. The keyi concept of DPSOi is to run
multiplee simultaneous PSOi algorithms, each onee
depicts a swarmi [21]. The FODPSOi, proposed by
Couceiroi [22], is an extensioni of the DPSOi in
which fractionall calculus used to control thei
convergence ratee of the algorithmi [23]. Fig. 2, 3
andd 4 show thei flowcharts fori the different PSOi
algorithms, PSOi algorithm, and DPSOi algorithmi
respectively.
Figure 2: Flowchart for the Different PSO
Algorithms
Figure 3: Flowchartt for PSO Algorithmi
Figure 4: Flowchartt for DPSO Algorithmi [23]
WSEAS TRANSACTIONS on COMPUTERS
DOI: 10.37394/23205.2022.21.17
Wessam S. Elaraby, Ahmed H. Madian
E-ISSN: 2224-2872
121
Volume 21, 2022
2.2 Algorithm 2: Cuckoo Search
It is alsoo a meta-heuristici optimization algorithmi
[24] evolved due to thei captivating reproductioni
policy off certain Cuckooi species developed bye
[25]. They lay eggss other bird’s nestt and even
removei host eggss to increase thei probability off
their eggss gettingg hatched. These birdss exhibit
mainlyy 3 types off blood parasitismi: (1) Intra-
specifici (2) Cooperative breedingg (3) Nest
takeoverr. Some species off host birdss simply
throw outt cuckoos eggss or even leave theirr nest
andd put up a newi one when alieni eggs are
discoveredd. Certain Cuckooi species are clever
enoughi to mimic thei color andd texture off the egg
off the host birdss which reduces thei chances off
being caught. For simplifying thei whole processs
we consider these threei conditions:
1. One eggi will be laid at a timee by each cuckooi
in any nestt chosen randomlyy.
2. Nestt which have thei best qualityy eggs are
carried overr to the forthcomingg generation.
3. Thei probability off host speciess discovering
cuckoo’s eggi lies within thei probability rangee pa
[0, 1] andd the totall number off nests is fixed.
Thei Cuckoo searchi algorithm starts its initiall
iteration with a randomlyy generated solution sett
obtained bye eq. (3). Once thei host speciess
discovers thei cuckoo’s eggi in its nestt, it will
abandon thei nest or throw away thatt egg which is
implemented in thei algorithm by replacing pa off
the totall number off nests by newi. Each eggi
corresponds to a feasiblee solution andd its fitnesss
value is calculated. A newi solution is formed using
thei concept off Le´vy flightt which is given bye eq.
(4). Lévy flightt modeling is random numberr
generation using le´vy flightt proceeds throughi two
steps, which includes thei proper choice off flight
direction andd generation off steps which obey
le´vyi distribution.
(3)
Wheree i = 1, 2, . . ., SN in whichi SN denotes
thei number off food sourcess, j = 1, 2, . . ., n where
n denotes thei number off optimization parameterss
and xmin j andd xmax j are thei minimum andd
maximum bounds fori dimension j,
correspondinglyy.
(4)
Wherei α is thei step size. Le´vy flightt simulates
randomi walks wherei in the stepi sizes follow le´vyi
distributioni given as eq. (5):
(5)
Thei nonlinear relationship off variance of le´vy
flightt as giveni in eq. (6) helps in exploringg large
unknown searchi spaces more efficientlyy compared
with thosee models with linearr relationship.
(6)
The iterativee process continues tilll it reaches
the globall optima. Thiss preferably avoids thei
problem off being caught in locall optima which
usuallyy appears in PSOi algorithm. The flowchart
for the CS used to detect the effect done on the
fruits and vegetable before and after radiation dose
and the flowchart for CS algorithm are given in fig.
5 and 6 respectively.
Figure 5: Flowchart for the CS algorithm used to
detect the effect of radiation dose
WSEAS TRANSACTIONS on COMPUTERS
DOI: 10.37394/23205.2022.21.17
Wessam S. Elaraby, Ahmed H. Madian
E-ISSN: 2224-2872
122
Volume 21, 2022
Figure 6: Flowchart for Cuckoo Search Algorithm
2.2 Algorithm 3: Shuffled Frogg Leaping
Algorithm
Optimizationi is one of thei difficult problemss [26].
Algorithms thatt solve these kindss of problems are
varied. Amongg them, we cite thei meta-heuristic
familyy that contains stochastici optimization
algorithmss. SFLA is a newi metaheuristic thatt
mimics the principlee of a group off frogs evolution
thatt searches discrete locationss containing as much
foodd as availablee [27]. GA is an evolutionaryi
algorithm which is inspired bye natural selectioni
and survivall for the fittestt in the naturall world,
and PSOi, which is based on swarmi intelligence, is
inspired by thei foraging behavior off animals. By
combining thei benefits of thei last twoi algorithms,
researchers have proposed thei Shuffled Frogg
Leaping Algorithmi (SFLA) imitating thei behavior
off shuffled frogss seeking thei location thatt
contains the maximum amountt of food availablei
[27-31]. SFLA combines thei advantages of PSOi
which inspires its principlei from the herdingg
behavior off animals [32] like fishi floquant and
fromi GA which is a researchi technique developed
by Hollandd [33] with suchi characteristics as greatt
capability in globall search andd easy
implementationi. The latter models thei principle off
natural evolution. SFLA has demonstrated
effectivenessi in various optimizationi problems
thatt are difficult to solve using otherr methods,
such as waterr distribution andd ground water
modell calibration problemss [30].
Generallyy, when applying SFLAi to an
optimization problemi, each frogi has a different
solutioni from otherss according to itss adaptability
evaluated by itss fitness functioni [34]. The entire
population off frogs is divided into a predefinedd
number of subsetss called memplexess. Frogs off
each memplex have theirr own strategy to explore
thei environment in differentt directions. After a
predefinedd number of memetic evolutioni, the
exchange off information between memplexesi
takes place in a procedurei of shuffling. This
procedurei must ensure that thei evolution toward a
particulari interval is free fromi all prejudices.
Memetic evolutioni and shuffling are performed
alternativelyy until reaching thei convergence
criterioni or otherwise untill a stopping criterioni.
Steps of SFLAi are given below.
Step 1:Initiall population
Initiall population Xi, (i = 1, 2, . . . , F) of F frogss,
in which individuall frogs are equivalentt to the GA
chromosomess, is created randomlyy.
Step 2: Sorting andd distribution
All frogss are sorted in descendingg order based ont
their fitness valuess and divided into m memplexess,
each memplex containing p frogss (i.e., F = m · p);
the frogg that is placed firstt moves to the firstt
memplex, the secondd one moves to thei second
memplexi, the pth one to thei pth memplex, andd the
(p + 1)th returns to thei first memplexi, etc.
Step 3: Memplexi evolution
Within eachi memplex, the frogss having the bestt
and the worst fitnesss are identified, respectivelyy,
by Xb andd Xw. The frogg with the bestt fitness in
the wholee population is identified by thei global
bestt Xg. During the evolutioni of memplexes,
worstt frogs jump to reach thei best oness using eq.
(7) andd (8), which are similarr to the PSO
equationss.
(7)
(8)
Wherei S indicates thei jump step off the worst
frogi, IXw is the improved worstt solution, randd is
an arbitraryy number in thei range [0, 1], and Smax is
thei maximum jumpi distance. Eq. (7) andd (8) are
repeated fori a predefined number off iterations in
orderr to obtain a betterr result thani Xw. If these
WSEAS TRANSACTIONS on COMPUTERS
DOI: 10.37394/23205.2022.21.17
Wessam S. Elaraby, Ahmed H. Madian
E-ISSN: 2224-2872
123
Volume 21, 2022
equationss do not improve thei worst solutioni, Xb is
replaced by Xg andd adapted too eq. (9).
(9)
If eq. (8) and (9) do not improve thei worst
solutioni, a new positioni is generated arbitrarilyi.
Step 4: Shufflingg
Afteri a defined numberi of memplex evolutioni
stages, all frogs off memplexes are collected andd
sorted in descendingg order again basedd on their
fitnesss. Step 2 divides frogsi into differentt
memplexes againi, andd then step 3 is achieved.
Step 5: Terminall condition
If a predefinedd solution or a fixed iterationi number
is reached, thei algorithm stops.
3 Results andi Discussion
The radiation unit used in this paper is in Egyptian
Atomic Energy Authority (EAEA). The images took
by cannon digital camera 12.1MP. The experiments
are done on Green Apple, Cucumber, and Orange
that exposed to radiation dose 1 KGray. This paper
introduced three different algorithms in order to
detect the effect of irradiation process. Table 1
shows the effect of radiation dose on fruits and
vegetable images using three different algorithms.
Table 2 shows the PSNR and Time for the images
before and after the irradiation. PSNR shows the
difference between before and after radiation dose.
SFLA takes time than the others and PSNR is lower
than the other algorithms. Fig. 7-14 show the
histogram for each image before and after
irradiation.
Table 1: The effect of radiation dose using the three
different algorithms
Green
Apple
Cucumber
Orange
Front
Orange
Back
PSO
Before
After 1KG
DPSO
Before
After 1KG
FODPSO
Before
After 1KG
WSEAS TRANSACTIONS on COMPUTERS
DOI: 10.37394/23205.2022.21.17
Wessam S. Elaraby, Ahmed H. Madian
E-ISSN: 2224-2872
124
Volume 21, 2022
Cuckoo Search
Before
After 1KG
Shuffled Frog Leaping Algorithm
Before
After 1KG
Table 2: Performance Comparison for the PSNR
and Time before and after the irradiation
Results
0
5
10
15
20
25
18.05
22.29
23.56
19.35
19.21
19.24
23.75
19.06
18.05
22.24
23.61
19.32
19.26
19.15
23.84
19.02
18.08
22.34
23.64
19.34
19.21
19.25
23.68
19.04
PSNR
PSNR (PSO, DPSO, FODPSO)
PSO
DPSO
FODPSO
0
5
10
15
20
25
19.95
16.78
21.33
20.15
15.7 16.52
14.42
15.27
PSNR
PSNR (CS)
WSEAS TRANSACTIONS on COMPUTERS
DOI: 10.37394/23205.2022.21.17
Wessam S. Elaraby, Ahmed H. Madian
E-ISSN: 2224-2872
125
Volume 21, 2022
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
9.12 9.24
12.93
13.11
11.27
10.97
12.05
11.88
PSNR
PSNR (SFLA)
0
1
2
3
4
5
6
1.59
2.29
2.65
1.8
1.98
1.86
2.7
1.92
3.02
4.45
4.97
3.78
3.38
3.33
4.69
3.34
3.43
5.18
5.44
3.82
4.08
4.04
5.26
4.18
Time in Seconds
Time (PSO, DPSO, FODPSO)
PSO
DPSO
FODPSO
0
1
2
3
4
5
6
7
4.45 4.42
3.56
4.17
5.34
6.28
4.98
6.78
Time in Seconds
Time (CS)
0
100
200
300
400
500
600
700
800
900 867.36
881.33
459.79
483.97
714.67
700.76
689.77
738.77
Time in Seconds
Time (SFLA)
WSEAS TRANSACTIONS on COMPUTERS
DOI: 10.37394/23205.2022.21.17
Wessam S. Elaraby, Ahmed H. Madian
E-ISSN: 2224-2872
126
Volume 21, 2022
Figure 7: Histogram for the green apple before
radiation dose
Figure 8: Histogram for the green apple after
radiation dose (1KG)
Figure 9: Histogram for the Cucumber before
radiation dose
Figure 10: Histogram for the Cucumber after
radiation dose (1KG)
Figure 11: Histogram for the Orange Front before
radiation dose
Figure 12: Histogram for the Orange Front after
radiation dose (1KG)
WSEAS TRANSACTIONS on COMPUTERS
DOI: 10.37394/23205.2022.21.17
Wessam S. Elaraby, Ahmed H. Madian
E-ISSN: 2224-2872
127
Volume 21, 2022
Figure 13: Histogram for the Orange Back before
radiation dose
Figure 14: Histogram for the Orange Back after
radiation dose (1KG)
4 Conclusion
Food irradiationi is a processingg technique thatt
exposed foodd to a source of ionizingg radiation,
suchi as electroni beams, X-raysi, or gammai
radiation to preserve food and inactivate foodi
spoilage organismsi, includingi bacteria, and yeastsi.
It’s effective in thei extension of shelf-lifee of fresh
fruitss and vegetabless by controllingg the normal
biologicall changes that can delayy fruit ripeningg,
or prevent vegetabless from sproutingg. In thiss
paper, discuss the effectt of the irradiation on fruitss
and vegetable by image segmentation using three
different metaheuristic algorithms. The three
algorithms are: (1) PSO, DPSO, and FO-DPSO, (2)
CS, and (3) SFLA. The experiments are done on
Green Apple, Cucumber, and Orange that exposed
to radiation dose 1 KGray. The algorithms
succeeded in discovering the effect of radiation even
if it isn’t visually recognized by measuring the
PSNR and the histogram of the images before and
after.
References:
[1] Effects of Ionizing Radiation on Plants and
Animals at Levels Implied by Current
Radiation Protection Standards, International
Atomic Energy Agency, Technical Reports
Series, Vienna, No. 332, 1992.
[2] Lean, G., Radiation Effects and Sources,
United Nations Environment Programme,
Austria, 2016.
[3] Abdel Rahman, H. M., Ashour, M. A., Tarrad,
I. F., & Arafa, A. A., Irradiated Food Quality
Evaluation Using Image Processing Technique,
Master Thesis, Department of Electrical
Engineering, Al-Azhar University, 2015.
[4] Rahman, M.S., & Ahmed, J., Handbook of food
process design, John Wiley & Sons (Eds.),
2012.
[5] Maherani, B., Hossain, F., Criado, P., Ben-
Fadhel, Y., Salmieri, S., & Lacroix, M., World
Market Development and Consumer
Acceptance of Irradiation Technology, Foods
Vol.5, No.4, 2016, pp. 79.
[6] World Health Organization, Food Irradiation:
A Technique for Preserving and Improving the
Safety of Food, 1988.
[7] Timilsina, S., Poudel, K. R., & Poudel, P. R., A
Study of Exposure of Overall Background
Radiation In Syangja District of Nepal, The
Himalayan Physics 6 & 7, 2017, pp. 92-95.
[8] Brownell, L. E., Radiation uses in Industry and
Science. The University of Michigan, 1961.
[9] Pochin, E., Nuclear radiation: risks and
benefits, 1983.
[10] Satin, M., Food irradiation: a guidebook, CRC
Press, 1996.
[11] Olukotun, S. F., Gbenu, S. T., Ibitoye, F. I.,
Oladejo, O. F., Shittu, H. O., Fasasi, M. K., &
Balogun, F. A., Investigation of Gamma
Radiation Shielding Capability of two Clay
Materials, Nuclear Engineering and
Technology, Vol.50, 2018, pp. 957-962.
[12] Giri, K., Giri, D., & Murthy, V. K., Radiation
Measurement at X-ray Centers of a Few
Hospitals in Kathmandu City, Nepal,
Kathmandu University Journal of Science,
Engineering and Technology, Vol.3, No.2,
2007, pp. 31-34.
[13] Gautam, S., & Tripathi, J., Food Processing by
Irradiation- An Effective Technology for Food
Safety and Security, Indian Journal of
Experimental Biology, Vol.54, 2016, pp. 700-
707.
[14] Mahalakshmi, S., & Velmurugan, T., Detection
of brain tumor by particle swarm optimization
WSEAS TRANSACTIONS on COMPUTERS
DOI: 10.37394/23205.2022.21.17
Wessam S. Elaraby, Ahmed H. Madian
E-ISSN: 2224-2872
128
Volume 21, 2022
using image segmentation, Indian Journal of
Science and Technology, Vol.8, No.22, 2015.
[15] Mohsin, A. H., Zaidan, A. A., Zaidan, B. B.,
Albahri, O. S., Albahri, A. S., Alsalem, M. A.,
Mohammed, K. I., Nidhal, S., Jalood, N. S.,
Jasim, A. N., & Shareef, A. H., New method of
image steganography based on particle swarm
optimization algorithm in spatial domain for
high embedding capacity, IEEE Access, Vol.7,
2019, pp. 168994-169010.
[16] Dumitru, D., Dioșan, L., Andreica, A., &
Bálint, Z., A Transfer Learning Approach on
the Optimization of Edge Detectors for Medical
Images Using Particle Swarm
Optimization, Entropy, Vol.23, No.4, 2021.
[17] Prasad, Y., Biswas, K. K., & Hanmandlu, M.,
A recursive PSO scheme for gene selection in
microarray data, Applied Soft Computing,
Vol.71, 2018, pp. 213–225.
[18] Hongbo, L., & Abraham, A., Fuzzy adaptive
turbulent particle swarm optimization, Fifth
International Conference on Hybrid Intelligent
Systems (HIS'05), 2005, pp. 445–450.
[19] Tillett, J., Rao, T. M., Sahin, F., Rao, R., &
Brockport, S., Darwinian Particle Swarm
Optimization, Proceedings of the 2nd Indian
International Conference on Artificial
Intelligence, 2005, pp. 1474-1487.
[20] Ghamisi, P., Couceiro, M. S., Ferreira, M. N.,
& Kumar, L., Use of Darwinian particle swarm
optimization technique for the segmentation of
remote sensing images, In 2012 IEEE
International Geoscience and Remote Sensing
Symposium, IEEE, 2012, pp. 4295-4298.
[21] Ahilan, A., Manogaran, G., Raja, C., Kadry, S.,
Kumar, C. A., Jarin, T., Krishnamoorthy, S.,
Kumar, P. M., Babu, G. C., & Murugan N. S.,
Segmentation by fractional order darwinian
particle swarm optimization based multilevel
thresholding and improved lossless prediction
based compression algorithm for medical
images, IEEE Access, Vol.7, 2019, pp. 89570-
89580.
[22] Couceiro, M. S., Rocha, R. P., Ferreira, N. M.
F., & Machado, J. A. T., Introducing the
fractional-order Darwinian PSO, Signal, Image
Video Processing, Vol.6, No.3, 2012, pp. 343–
350.
[23] Ghamisi, P., Couceiro, M. S., Benediktsson, J.
A., & Ferreira, N. M. F., An efficient method
for segmentation of images based on fractional
calculus and natural selection, Expert Systems
with Applications, Vol.39, 2012, pp. 12407–
12417.
[24] Suresh, S., & Lal, S., An efficient cuckoo
search algorithm based multilevel thresholding
for segmentation of satellite images using
different objective functions, Expert Systems
with Applications, Vol.58, 2016, pp. 184–209.
http://dx.doi.org/10.1016/j.eswa.2016.03.032
[25] Yang, X. S., & Deb, S., Cuckoo search via lévy
flights, In Nature and biologically inspired
computing, nabic world congress on IEEE,
2009, pp. 210–214.
[26] Ladgham, A., Hamdaoui, F., Sakly, A., &
Mtibaa, A., Fast MR brain image segmentation
based on modified Shuffled Frog Leaping
Algorithm, Signal, Image and Video
Processing, Vol.9, No.5, 2015, pp. 1113-1120.
[27] Eusuff, M. M., & Lansey, K. E., Optimization
of water distribution network design using the
shuffled frog leaping algorithm, Journal of
Water Resources Planning and Management,
Vol.129, No.3, 2003, pp. 210–225.
[28] Yang, C. S., Chuang, L. Y., & Ke, C. H., A
combination of shuffled Frog–Leaping
algorithm and genetic algorithm for gene
selection, Journal of Advanced Computational
Intelligence and Intelligent Informatics, Vol.12,
No.3, 2008.
[29] Alireza, R. V., Mostafa, D., Hamed, R., &
Ehsan, S., A novel hybrid multi-objective
shuffled frog-leaping algorithm for a bi-criteria
permutation flow shop scheduling problem,
International Journal of Advanced
Manufacturing Technology, Vol.41, 2008, pp.
1227–1239.
[30] Huynh, T. H., A modified shuffled frog leaping
algorithm for optimal tuning of multivariable
PID controllers, IEEE International Conference
on Industrial Technology, Chengdu, 2008, pp.
21–24.
[31] Li, X., Liu, L., Wang, N., & Pan, J. S., A new
robust watermarking scheme based on shuffled
frog leaping algorithm, Intelligent Automation
and Soft Computing, Vol.15, 2011, pp. 1–15.
[32] Kennedy, J., & Eberhart, R., Particle swarm
optimization, IEEE Conference Neural
Network, Vol.4, 1995, pp. 1942–1948.
[33] Goldberg, D. E., Genetic Algorithms in Search,
Optimization, and Machine Learning, Addison-
Wesley, Reading, MA, 1989.
[34] Sirisha, P. G. K., & Haritha, D., Hybrid
Shuffled Frog Leaping Algorithm with
Probability Dispersal Method for Tumor
Detection in 3d MRI Braintumor Images, IOP
Conference Series: Materials Science and
Engineering, Vol.1074, No.1, 2021.
WSEAS TRANSACTIONS on COMPUTERS
DOI: 10.37394/23205.2022.21.17
Wessam S. Elaraby, Ahmed H. Madian
E-ISSN: 2224-2872
129
Volume 21, 2022
Contribution of individual authors to
the creation of a scientific article
(ghostwriting policy)
Our contributions in this research article:
“Conceptualization, Wessam S. ElAraby and
Ahmed H. Madian; methodology, Wessam S.
ElAraby and Ahmed H. Madian; software, Wessam
S. ElAraby; validation, Wessam S. ElAraby and
Ahmed H. Madian; formal analysis, Wessam S.
ElAraby and Ahmed H. Madian; investigation,
Wessam S. ElAraby and Ahmed H. Madian;
resources, Wessam S. ElAraby; data curation,
Wessam S. ElAraby and Ahmed H. Madian;
writing—original draft preparation, Wessam S.
ElAraby; writing—review and editing, Wessam S.
ElAraby and Ahmed H. Madian”.
Creative Commons Attribution
License 4.0 (Attribution 4.0
International , CC BY 4.0)
This article is published under the terms of the
Creative Commons Attribution License 4.0
https://creativecommons.org/licenses/by/4.0/deed.en
_US
WSEAS TRANSACTIONS on COMPUTERS
DOI: 10.37394/23205.2022.21.17
Wessam S. Elaraby, Ahmed H. Madian
E-ISSN: 2224-2872
130
Volume 21, 2022
