TELKOM
NIKA Indonesia
n
Journal of
Electrical En
gineering
Vol. 12, No. 8, August 201
4, pp. 5779 ~ 5783
DOI: 10.115
9
1
/telkomni
ka.
v
12i8.627
2
5779
Re
cei
v
ed Ap
ril 7, 2014; Re
vised Ma
y 17
, 2014; Accep
t
ed Jun
e
2, 2014
Automatic Monitoring of Pest Insects Traps Using
Image P
r
ocessing
Aka
s
h J. Up
adhy
ay
*
1
, P.
V. Ingole
2
Dep
a
rtment of
Electronics a
n
d
telecomm
uni
cation En
gi
nee
ring, G. H. Raisoni C
o
ll
ege
of Engin
eeri
ng,
Amravati
*Corres
p
o
ndi
n
g
author, e-ma
i
l
: Ajupa
dh
ya
y1
3@gma
il.com
1
, prasha
nt.ing
ol
e@rais
on
i.net
2
A
b
st
r
a
ct
Monitori
ng
p
e
s
t
insect
po
pul
a
t
ion
is c
u
rrentl
y
an
i
m
p
o
rtant
issu
e i
n
cr
op
protectio
n
. At f
a
rm l
e
vel
insect po
pu
lati
on
mon
i
tori
ng
is consiste
ntl
y
operat
e
d
by
repeate
d
sur
v
eys by a hu
ma
n op
erator
of
adh
esive tr
aps
, disse
mi
nate
d
throug
h the fi
eld, w
here
ins
e
cts remai
n
stuck w
hen
attracted. T
h
is is
a
lab
o
rio
u
s an
d
time-c
onsu
m
i
ng activity, an
d it w
oul
d b
e
of great adv
antag
e for farmers to h
a
ve
a
n
affordable system
perfor
m
i
ng
this task automatically.
A system
based on a distribut
ed imaging device that
is able to a
u
tomatica
lly acq
u
ire a
nd trans
mit i
m
a
ges
of
the trappin
g
area to a re
mote host statio
n is
prop
osed. T
h
e
station
eva
l
ua
tes the
i
n
sect
dens
ity evo
l
uti
on at
differe
nt
farm
s
i
tes
a
nd sends a mess
ag
e
to the
server
that p
e
stici
des
are r
equ
ire
d
,
w
hen i
n
se
ct
d
ensity
go
es ov
er thres
hol
d
or
an
al
ar
mi
ng
l
e
ve
l
throug
h GSM techn
o
lo
gy. T
he netw
o
rk arch
itecture cons
ists of a master n
ode h
o
sted
in
a PC an
d a set
of
client n
o
d
e
s, sprea
d
in the fi
elds, w
h
ich ac
t as
mon
i
torin
g
stations. T
he mast
er nod
e
coordi
nates t
h
e
netw
o
rk and re
trieves the cap
t
ured i
m
a
ges from the cl
ient n
odes. A GSM
mo
de
m w
h
ich i
s
interfaced w
i
t
h
PC throug
h U
SB port is use
d
to send
mes
s
ages to the
p
a
rticul
ar nu
mb
ers for requ
ire
d
attentio
n in t
h
is
regar
d for further action.
Key
w
ord
:
M
A
TLAB, GSM m
odem
, cam
e
ra
Copy
right
©
2014 In
stitu
t
e o
f
Ad
van
ced
En
g
i
n
eerin
g and
Scien
ce. All
rig
h
t
s reser
ve
d
.
1. Introduc
tion
An accurate and timely monitorin
g
of pes
t in
se
cts
popul
ation is a key issue
in cro
p
prote
c
tion. At farm level, p
e
st po
pulatio
n leve
ls duri
n
g
crop sea
s
o
n
are ro
utinel
y
monitore
d by
mean
s of ad
hesive trap
s
dissemin
ated
throug
h
the
field at app
ro
priate
spatial
den
sity, when
inse
cts rem
a
in stu
c
k whe
n
attra
c
ted i
n
to by colou
r
ap
pea
ran
c
e or odo
ro
u
s
b
a
its li
ke
sex
pheromo
ne. This ta
sk
req
u
ire
s
re
peate
d
su
rveys in
the field with
visual ob
se
rvations of tra
p
s
and
re
co
rdin
g of the
nu
mber of
in
se
cts
ca
pture
d
by a hu
man
operator. Bei
ng the
s
e
su
rvey
pro
c
e
s
s a hi
g
h
ly labori
o
u
s
and time
co
n
s
umin
g op
era
t
ion, they are
unlik
e to be
condu
cted
at the
necessa
ry m
onitorin
g
freq
uen
cy, o
ften yielding poo
r results, sig
n
if
icantly affe
cte
d
by
observe
r’s
skill
o
r
fatigu
e, or by
su
rvey
ing
con
d
itions [1-2]. Th
e on
goin
g
d
e
v
elopment
s i
n
ne
w
micro,
low
co
st imaging
device
s
an
d in wirele
ss co
mmuni
cation tech
nol
ogy coul
d g
i
ve a valuable
contri
bution i
n
facing auto
m
atic monito
ring of pe
st inse
cts by e
s
tablishi
ng a came
ra a
b
le
to
remotely a
s
sess the adh
e
s
ive trap
s ca
pture
s
in the field.
Indeed a
su
ccessful dev
elopme
n
t of such sy
ste
m
would p
o
tentially allow takin
g
deci
s
io
ns ab
out in
se
ct co
ntrol
strate
gi
es
at fa
rm
ba
sed
on
continuou
sly up
da
ted map
s
of
pest
inse
ct pop
ula
t
ion levels which a
r
e retri
e
ved in
a se
rver from a di
stribute
d
net
work of cam
e
ra
s
transmitting from the field. An alternativ
e soluti
o
n
, explore
d
in this study, is ba
sed on ad
opti
n
g
very simpl
e
stations o
n
ly h
a
ving ability of capt
u
r
ing
and tra
n
smitting imag
es to
the ho
st in the
netwo
rk. A h
o
st statio
n wil
l
pro
c
e
ss all t
he re
ce
ive
d
image
s an
d count the tra
p
ped in
se
cts.
This
same station will eventually provide
an interfac
e to the operator to
sh
ow i
n
real time t
he
popul
ation m
ap and its trend, as
well
as a me
ssag
e will be sen
t
through GS
M module
when
inse
ct den
sity goes
over a
n
alarmi
ng le
vel. Al
so, this sy
stem can
be appli
ed to the are
a
s
h
a
rd
to reach [1-5]
.
2. Related Work
It is given that, pests a
r
e those that direct
ly spoil th
e crop, and
p
e
st control h
a
s always
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 8, August 2014: 577
9 –
5783
5780
been
con
s
id
ered th
e mo
st com
p
licate
d
chall
eng
e
to overcome.
A well-kno
w
n tech
nique
to
perfo
rm pe
st control monit
o
ring i
s
ba
se
d on the u
s
e
of inse
ct trap
s co
nvenie
n
tly spre
ad ove
r
the
spe
c
ified
co
n
t
rol area.
De
pendi
ng o
n
the vari
ety of
inse
ct, ea
ch t
r
ap i
s
p
r
o
perl
y
installed
wi
th
pheromo
n
e
s
or othe
r ch
e
m
ical sub
s
tan
c
e
s
that attr
a
c
t the inse
cts we wa
nt to capture. Th
e traps
are d
e
si
gne
d
in su
ch a
way that inse
cts enteri
ng in
the trap a
r
e
unable to le
ave it, so pe
st
monitori
ng
sy
stem
s will
pe
riodically coll
e
c
t the
data
of
each tra
p
cap
t
ured to
pe
rfo
r
m a
n
effici
ent
pest control
monitori
ng [2].
To get
mo
st reliabl
e results, imag
e p
r
o
c
e
ssi
ng
sho
u
l
d be
pe
rformed o
n
ly on
the tra
p
active a
r
ea [
1
]. Energy
saving of
wire
less sen
s
o
r
netwo
rk no
d
e
is
usually
available f
r
o
m
the
hard
w
a
r
e, op
erating
sy
ste
m
s
a
nd co
m
m
unication pr
otocol
s.
In
thi
s
p
r
oje
c
t we are usi
ng GS
M
modem
whi
c
h use
s
lo
w p
o
we
r [5]. The large
scal
e
, long di
stan
ce,
and long
-term monitorin
g
for
agri
c
ultu
ral i
n
formation
can
be
achieved
by us
i
ng
pro
p
o
se
d mo
nitori
ng
system. M
u
ch
imp
r
oved
spatial resol
u
tion and te
mporal re
sol
u
tion is obta
i
ned compa
r
ed to traditio
nal method
s for
monitori
ng th
e data of pest
populatio
n [6].
Usi
ng GSM
wa
s 100% b
a
se
d on cro
s
s ch
eckin
g
the sent an
d receive
d
data,
and the
integrity of tra
n
smi
ssi
on i
s
guarant
ee
d.
Also
by u
s
ing
GSM, d
a
ta o
r
me
ssag
e
ca
n be
saved f
r
om
attacks an
d i
n
formatio
n is provide
d
to the use
r
[7]. They proposed method they firs
ts
identif
y
Aphids after t
hey go
for ot
her two
in
se
cts. Aphi
d
s
we
re ea
sy
to discrimi
nate be
cause
they
ha
ve
small vari
atio
n in colo
r inf
o
rmatio
n and
the body
si
ze was
sub
s
tantially different from oth
e
r
s
p
ec
ies
.
T
hey p
r
o
p
o
s
e
d
pe
s
t
de
te
c
t
io
n
s
y
s
t
e
m
in
c
l
ud
in
g
fo
ur
s
t
ep
s
n
a
m
e as
c
o
lor
c
o
n
v
er
sio
n
,
segm
entation
,
reductio
n
in noise and
co
unting [8].
An automatic electro
n
ic trap is de
sign
e
d
to m
onitor the flight of codling moth
males, to
identify the p
e
st an
d to fo
rward the i
n
fo
rmati
on
on trappe
d mal
e
s.
The
re
sults
sho
w
e
d
ho
w
the
modificatio
n
s perform
ed o
n
the standa
rd sticky
trap
s did not ca
use
s
a re
du
ction of the trap
cap
ability to catch the male
s, and al
so th
at the phot
os
taken by the
system an
d sent to a remo
te
unit allo
wed
an ea
sy moth
identificatio
n
and d
e
tectio
n. Also, they
sho
w
e
d
that the modifi
cati
ons
introdu
ce
d in the stand
ard t
r
ap
s do not af
fect
the captu
r
e ca
pa
city of the traps [9].
This
pape
r d
e
scrib
e
s the
monitori
ng of
pest in
se
ct p
o
pulatio
n by
usin
g wi
rele
ss sen
s
o
r
netwo
rk i
n
fie
l
d.In this proj
ect they ju
st found
the i
n
sect count a
n
d
they are
n
o
t cont
rolling
so
further [10]
.
The GSM tra
n
smi
ssi
on te
chniqu
es e
nab
le this sy
ste
m
to engag
e
in long
- dista
n
ce
data tran
smission, so the
sy
stem can b
e
deployed t
o
the area
s h
a
rd to rea
c
h
[11].
One of the
major a
d
vant
age
s of using
inse
ct phero
m
one tra
p
s
wi
th lure is that
they are designed to attra
c
t
a singl
e parti
cula
r sp
eci
e
s of insect. So
me other
n
o
n
targeted in
sects m
a
y accidentally visit the
trap b
u
t their numb
e
rs
wil
l
be lo
we
r th
an the t
a
rg
et pe
st. The
r
ef
ore, in
se
ct id
entification
is
automatic wit
h
lure
-b
ased
(phe
rom
one
) trap
s an
d
th
is tra
p
data
can
sig
n
ifica
n
tly comple
m
ent
field scouting
.
But problem
s like, Lu
re
s
sho
u
ld
be
re
placed on
ce
every month.
Too many n
o
n-
target insect
s could al
so b
e
probl
emati
c
issu
e [12].
3. Flo
w
Char
t
Figure 1. The
desig
ned
system perfo
rms trap monitori
ng task
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Autom
a
tic Monitoring of Pe
st Inse
cts Tra
p
s Us
in
g Im
a
ge Pro
c
e
ssi
n
g
(Aka
sh
J. Upadh
ya
y)
5781
The de
sig
ned
system pe
rfo
r
ms trap mo
n
i
toring
ta
sk a
nd cal
c
ul
ates the inse
ct de
nsity of
the trap
by
capturin
g the
trap im
age
a
nd the
n
p
r
ep
rocessin
g it.
The p
r
e
p
ro
ce
ssi
ng i
n
cl
ude
s
image bin
a
ri
zation, ero
s
ion
,
dilation, filte
r
ing et
c. Late
r
on sca
nnin
g
the numb
e
r of insect
s a
nd
perfo
rming th
e further a
c
tio
n
depe
nding
on the insect
den
sity using
the followin
g
algorith
m
.
4. Working
The develo
p
ed netwo
rk is based on
the GSM communi
catio
n
proto
c
ol. It allows
impleme
n
ting
seve
ral
networki
ng
archite
c
ture
s wi
th
m
oderate e
n
e
r
gy co
ns
umpti
on. Th
e
syste
m
netwo
rk con
s
ists
of a m
a
st
er n
ode,
call
e
d
ba
se
st
at
io
n
, whi
c
h
is h
o
sted
in
side
a PC, a
nd
a
set
of client node
s. The maste
r
node ha
s th
e role
of coo
r
dinating the
netwo
rk. Thi
s
node se
nd
s to
the client
s the requ
est for
sho
o
ting a pi
cture
(imag
e
) and then col
l
ects all the i
m
age
s, sen
d
i
ng
them to the PC. The im
age
s are pr
ocessed by the PC
that prod
uces
the inse
ct co
unt. The cli
e
n
t
node
s h
a
ve the role
of a
c
qui
ring th
e i
m
age
s up
on
requ
est a
n
d
transfe
r the
m
, towards t
he
maste
r
n
ode.
As th
e m
e
a
s
ured
phy
sical qu
antities
are
re
prese
n
t
ed by im
ag
es, the
d
e
vice
compl
e
xity is an issu
e here: each no
de
should
cont
ain a came
ra
, host is requ
ired to mana
ge
image ca
ptu
r
ing and
proce
s
sing. Rece
ntly,
su
ch sm
all
cap
t
uring
device
s
h
a
ve b
e
come
available. Th
e cam
e
ra is con
n
e
c
ted
to
the Com
puter. Th
e communi
catio
n
between t
h
e
Comp
uter
an
d the came
ra is throug
h
the USB po
rt. Commu
nica
tion between
host a
nd G
S
M
modem
con
s
i
s
ts of
three p
hases: co
nne
ction, se
tup,
download.
Th
e first on
e i
s
necessa
ry for
cha
nnel synchroni
zatio
n
.
In the do
wnl
oad
pha
se, t
he GSM
mod
e
m receives
messag
e to
be tra
n
smitted & the
n
transmit it to
the sp
ecifie
d
numbe
r [1].
Inse
ct re
co
gn
ition and
cou
n
ting in im
ag
es i
s
n’t a t
r
ivial
task a
s
lighti
ng con
d
ition
s
in real environment ar
e subje
c
ted to large
cha
nge
s. To tackl
e this
chall
enge, th
e implem
ent
ed ima
ge p
r
oce
s
sing
ha
s to
cop
e
with light
ch
ange
s a
nd
with
irre
gula
r
ities i
n
the glu
e
ov
er the trap. A
first
po
ssibili
ty is to apply
an imag
e tra
n
sform in
col
o
r
spa
c
e
su
ch t
hat insect
s b
e
com
e
line
a
rl
y sepa
rabl
e from the
ba
ckgrou
nd. A si
ngle pl
ane
can
indee
d sep
a
rate the pixels belo
nging t
o
inse
cts.
Ho
wever, the p
o
sition a
nd o
r
ientation of this
plane do
es
chang
e acco
rding to lightin
g and surfa
c
e con
d
ition that make
s th
is tran
sform
a
tio
n
usel
ess
on th
e field. Th
ere
f
ore a
mo
re
prin
ciple
d
a
p
p
roa
c
h
ha
s
b
een
develop
e
d
. We
con
s
id
er
each si
ngle i
m
age a
nd ext
r
act it
s lumin
ance an
d we
estimate the
backg
rou
nd i
m
age by a
ppl
ying
a media
n
filter to the l
u
minan
ce im
a
ge. On
ce
th
e ba
ckgro
u
n
d
is avail
abl
e, inse
cts
be
come
clea
rly visible
on the i
m
ag
e obtain
ed
subtra
cting
th
e
backg
ro
und
from the
origi
nal ima
ge, af
ter
this pro
c
e
s
sin
g
the GSM module
which is co
nne
ct
ed t
o
comp
uter th
roug
h USB p
o
rt will se
nd the
messag
e to the users wh
e
n
the inse
ct p
opulatio
n goe
s over thresh
old.
We d
e
velop
e
d
the
syste
m
whi
c
h
pe
rforms
the m
onitorin
g
ta
sk auto
m
atica
lly using
image p
r
o
c
e
ssi
ng an
d ta
king a
c
tion t
o
wa
rd
s controlling the p
e
s
t by sen
d
in
g messa
ge t
o
the
use
r
thro
ugh
GSM modem
.
The ho
st stati
on perfo
rm
s image p
r
o
c
e
s
sing o
n
the trap ca
pture
s
i
n
the followin
g
step
s:
Binari
z
ation, thre
shol
ding a
nd co
unting.
We a
pplie
d the ab
ove pro
c
ed
ure to
the
tr
ap captu
r
e
s
then we go
t
the
re
sults sho
w
n
belo
w
in
Fig
u
r
e
2, Fig
u
re
3
and
Fig
u
re
4
.
And d
epe
nd
ing o
n
the
n
u
m
ber of
cou
n
t
the me
ssag
e
is sent to th
e user. If the co
unt is
g
r
eate
r
than t
he defin
ed t
h
re
shol
d val
ue the
system
comm
uni
cate
s to the GSM modem a
nd
messag
e is sent for furthe
r action.
Figure 2. Cap
t
ure Image
Figure 3. Black a
nd White Image
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 8, August 2014: 577
9 –
5783
5782
Figure 4. Image Showi
ng
Cou
n
t of Inse
cts
5. Result
The de
sign
e
d
system h
a
s
su
cce
ssfull
y
show
n re
sults of autom
atic monito
rin
g
of pest
inse
cts tra
p
s;
also p
r
ovide
d
the
step
to
wards c
ontrol
ling of
pe
st. By using
ph
e
r
omo
ne t
r
ap
s this
system
can b
e
mo
re
efficie
n
t that it will
send m
e
ssa
g
e
of pa
rticul
ar
pesti
cide
de
p
endin
g
o
n
typ
e
of pest.
6. Conclusio
n
In this pape
r, we give brief
summa
ry ab
out variou
s m
e
thod
s and t
e
ch
niqu
e whi
c
h were
provide
d
by variou
s auth
o
rs fo
r ag
ricultural
p
e
st
s detectio
n
. Every method
ol
ogy is havin
g its
own meth
od
s as well a
s
e
n
vironm
ents i
n
whi
c
h it ap
plied. The
r
e
are some me
thods
whi
c
h
are
applie
d to
onl
y gre
enho
use
enviro
n
me
nts
whe
r
ea
s so
me of m
e
tho
d
s i
s
able
to i
dentify only o
ne
pest
s
n
a
me
as
white
fly. From
survey
it is
co
n
c
lu
de
that the
r
e i
s
ne
ce
ssity fo
r propo
sing
o
n
e
uniqu
e metho
d
whi
c
h will i
dentify not only large pe
sts as well a
s
la
rvae stag
e of various p
e
st
s.
Our futu
re
scope i
s
to p
r
o
pose on
e me
thodolo
g
y fo
r identificatio
n
of different
pest
s
of vari
ous
cro
p
s which
not only d
e
te
cts
and tell
s
whi
c
h p
e
st
h
a
s affe
ct the
cro
p
a
s
well
as it p
r
ovide
the
solutio
n
to p
r
event it or to
control it. Ou
r pro
p
o
s
ed
sy
stem i
s
hel
pful to farme
r
s
as
well a
s
ro
botic
vehicle
syste
m
which is u
s
ed for
sp
rayi
ng pe
stic
id
es automaticall
y
when
any p
e
sts o
c
cu
r in the
field.
Referen
ces
[1]
T
i
relli P, Borgh
e
se NA, Ped
e
r
s
ini
F, Galassi
G, Oberti R. Automatic mo
n
i
torin
g
of pest in
sects traps
b
y
Z
i
gbe
e-bas
ed
w
i
rel
e
ss
net
w
o
rking
of im
ag
e
sens
ors.
Instrumentati
o
n
an
d Me
asure
m
en
t T
e
chno
lo
g
y
Confer
ence (I2
M
T
C
)
. 2011; IEEE 1-5.
[2]
Otoniel
L
opez,
Ming
uel
Martin
ez R
a
ch, H
e
ct
or Mi
gal
lon,
Manu
el P
Ma
lum
b
res, Al
berto
B
onastre, J
u
a
n
J Serrano.
Monitor
i
ng P
e
st
Insect T
r
aps
by Mea
n
s of
Low
Pow
e
r Ima
ge S
ens
or
T
e
chno
log
i
es
.
Sensosrs. 20
1
2
; 12: 158
01-1
581
9.
[3]
Hussei
n
AN,
Masho
hor S, Iqba
l Sari
pa
n M.
A T
e
xture-Based A
ppro
a
c
h for Co
ntent
Based I
m
a
g
e
R
e
tri
e
va
l
System
fo
r Pl
an
t Lea
ve
s Im
ag
e
s
.
Signal Pr
ocess
i
ng
and its A
p
plications
(CS
PA), IEEE 7th
Internatio
na
l C
o
llo
qu
ium o
n
. 2011; 11-
14.
[4]
Nirmal K
u
mar
K, Ranjith P,
Prabak
aran
R.
Real T
i
me P
addy Cr
op F
i
e
l
d Mon
i
torin
g
Using Z
i
gbe
e
Netw
ork.
Emergin
g
T
r
ends in Electrica
l
and Com
put
er T
e
chnolo
g
y
(ICET
E
C
T
). Internatio
nal
Confer
ence. 2
011; 11
36-
114
0.
[5]
Che
n
Xi
an
yi,
Ji
n Z
h
i
gan
g, Ya
ng
Xion
g.
D
e
si
gn
of T
r
op
ical
Crops
Pests M
onitor
i
ng
Syste
m
B
a
se
d
o
n
wireless sensor network
. Co
nsumer E
l
ectr
onics, C
o
mmu
nicati
ons a
nd
Net
w
orks (CE
CNet). 20
12;
253
0-25
32.
[6]
Joe-Air Ji
an
g, Ch
w
a
n-L
u
T
s
eng, F
u
-Ming
Lu
, En
-Chen
g Ya
ng, Z
ong-S
i
ou
W
u
, Chia-Pa
n
g
Chen, Sh
ih-
Hsian
g
Lin, K
u
ang-
Cha
ng
Li
n
,
Chih-S
he
ng
L
i
ao. A GSM B
a
sd Rem
o
te W
i
r
e
less A
u
tomati
c Monitor
i
n
g
S
y
stem for
F
i
eld I
n
formatio
n
: A Cas
e
St
ud
y
or Ecol
og
ical M
onit
o
rin
g
of T
he Orien
t
al F
r
uit F
l
y,
Bactroceradors
a
lis (Hendel).
Co
mp
uters an
d Electron
ics in
Agricultur
e
. 20
08; 62(2): 2
43-
259.
[7]
Ch
w
a
n-L
u
T
s
eng, Jo
e-Air Ji
a
ng, R
en-Gue
y
Lee,
fu-Mi
ng
L
u
, Ch
eng-S
h
i
o
u Ou
ya
ng, Yi
h-
Shai
ng
Che
n
,
Chih-
H
sia
n
g
Cha
ng. F
e
asi
b
iklit
y st
ud
y
on
ap
plic
ation
of GSM-SM
S
techn
o
l
o
g
y
to F
i
e
l
d
Dat
a
Acquis
i
tion.
Co
mp
uters an
d El
ectronics i
n
Ag
riculture
. 2
006;
53(1): 45-5
9
.
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Autom
a
tic Monitoring of Pe
st Inse
cts Tra
p
s Us
in
g Im
a
ge Pro
c
e
ssi
n
g
(Aka
sh
J. Upadh
ya
y)
5783
[8]
Gaurav Y Ka
n
dalk
a
r, AV De
orank
ar, Dr PN Chat
ur
. A Revie
w
of Liter
a
ture on A
ppl
ic
ation
of Image
Processi
ng
for Identificati
on of
Agricultur
al P
e
sts on Vari
ou
s Crops. ISSN
: 2319-
596
7 IS
O 9001:2
0
0
8
Certifie
d Intern
ation
a
l Jo
urna
l of Engin
eer
i
ng
Scienc
e an
d Innovativ
e T
e
chn
o
lo
gy (IJESIT
)
.
2013; 2(6).
[9]
Adria
no GUAR
NIERI, Stefano MAINI,
Giova
nni MO
LARI, Vald
a RONDE
LLI. Automatic
trap for moth
detectio
n
in
int
egrate
d
p
e
st mana
geme
n
t.
Bull
etin of Inse
ctology.
2
0
1
1
; 64(2): 2
47-2
5
1
.
ISSN 1721
-
886
1.
[10]
C T
hulasi Pri
y
a, K Praveen,
A Srivid
ya. Monitor
i
ng Of Pest Insect
T
r
a
p
s Using Imag
e Sensors
&
Dspic.
Indi
a Internati
ona
l Jour
nal of Eng
i
n
eer
ing T
r
en
ds an
d
T
e
chnol
ogy
. 2
013; 4(9).
[11]
J
y
h
C
hern
g
S
h
ie
h, Jii
ng-Y
i
W
ang, T
z
u-Shi
ang
LIN,
Chi-
Hun
g
LIN,
En-
C
he
ng Y
a
n
g
,
Yao-Je
n T
s
ai,
Hsien-T
s
ung
T
s
ai, Ming-T
z
u Ch
iou, F
u
-
M
ing
LU, Joe
-
Air Jian
g.
A
GSM-based F
i
eld M
onitor
i
n
g
System for Sp
odo
ptera lit
ura
(F
abricius)
. 2011; EAEF 4(3): 77-82.
[12]
A
y
a
nav
a Maj
u
mdar E
x
tens
io
n Entomo
lo
gist
, Al
abam
a Co
o
perativ
e E
x
ten
s
ion S
y
stem. I
n
troducti
on to
Insect Pest Monitor
i
ng Us
ing
Pheromo
ne T
r
aps. T
e
xt book of W
i
reless Commun
i
cati
o
n
s: Principl
e
s
and Practic
e
b
y
T
heodor
e S. Rap
pap
ort 2nd
Edition.
[13]
T
e
xt book of Cr
y
p
to
gra
p
h
y
a
n
d
Net
w
ork
Se
curit
y
b
y
W
i
l
lia
m Stallin
gs 5th
Edition.
[14]
Qing Ya
o, Qin
g
jie
Li
u a, T
homas G Dietteri
ch, Sini
s
a
T
odorovic, Jeffre
y
Lin, Gua
n
g
q
ia
n
g
Dia
o, Bao
j
un
Yang, Jia
n
T
a
ng.
Segm
en
ta
ti
o
n
o
f
tou
c
hi
n
g
i
n
se
cts ba
se
d o
n
op
ti
ca
l fl
ow a
n
d
N
C
u
t
s
. 201
2 IAg
r
E
.
Publ
ishe
d b
y
E
l
sevier
Ltd.
Evaluation Warning : The document was created with Spire.PDF for Python.