Inter
national
J
our
nal
of
Electrical
and
Computer
Engineering
(IJECE)
V
ol.
10,
No.
5,
October
2020,
pp.
5217
5226
ISSN:
2088-8708,
DOI:
10.11591/ijece.v10i5.pp5217-5226
r
5217
Cyber
-ph
ysical
system
based
on
image
r
ecognition
to
impr
o
v
e
traffic
flo
w:
A
case
study
Vladimir
Sanchez
P
adilla,
Ronald
A.
P
onguillo,
Ar
naldo
A.
Abad,
Lissette
E.
Salas
ESPOL
Polytechnic
Uni
v
ersity
,
Escuela
Superior
Politecnica
del
Litoral,
ESPOL,
F
aculty
of
Electrical
and
Computer
Engineering,
Campus
Gusta
v
o
Galindo,
Ecuador
Article
Inf
o
Article
history:
Recei
v
ed
Sep
10,
2019
Re
vised
Feb
2,
2020
Accepted
Mar
25,
2020
K
eyw
ords:
Image
capture
Object
recognition
P
attern
matching
Raspberry
Pi
UV
camera
V
ehicle
detection
ABSTRA
CT
V
ehicular
traf
fic
in
metropolitan
areas
turns
congested
along
either
paths
or
periods.
As
a
case
study
,
we
ha
v
e
considered
a
mass
transport
system
with
a
b
us
fleet
that
rides
o
v
er
e
xclusi
v
e
lanes
across
streets
and
a
v
enues
in
an
urban
area
that
does
not
allo
w
the
circulation
of
lightweight
v
ehicles,
car
go,
and
motorc
ycles.
This
traf
fic
flo
w
becomes
congested
due
to
the
absence
of
restriction
policies
based
on
criteria.
Moreo
v
er
,
the
e
xclusi
v
e
lanes
are
at
ground
le
v
el,
decreasing
lanes
for
other
v
ehicles.
The
main
objecti
v
e
of
this
proposal
consists
of
controlling
the
access
to
the
e
xclusi
v
e
lanes
by
a
c
yber
-ph
ysical
system
follo
wing
authorizati
on
conditions,
v
erifying
the
per
-
mission
status
of
a
v
ehicle
by
the
accurate
recognition
of
license
plates
to
reduce
traf
fic
congestion.
Therefore,
in
the
case
of
in
v
ading
an
e
xclusi
v
e
lane
without
permission,
the
v
ehicle
o
wner
gets
a
notification
of
the
fine
with
the
respecti
v
e
e
vidence.
Copyright
©
2020
Insitute
of
Advanced
Engineeering
and
Science
.
All
rights
r
eserved.
Corresponding
A
uthor:
Vladimir
Sanchez
P
adilla,
F
aculty
of
Electrical
and
Computer
Engineering,
ESPOL
Polytechnic
Uni
v
ersity
,
Escuela
Superior
Politecnica
del
Litoral,
Campus
Gusta
v
o
Galindo,
Km
30.5
V
ia
Perimetral,
P
.
O.
Box
09-01-5863,
Guayaquil,
Ecuador
.
Email:
vladsanc@espol.edu.ec
1.
INTR
ODUCTION
Metropolitan
areas
usually
ha
v
e
high
occupanc
y
v
ehicle
lanes
duly
identified
based
on
criteria
or
conditions,
e.g.,
when
the
dri
v
er
of
a
lightweight
v
ehicle
tra
v
els
with
at
least
one
pass
enger
.
In
t
h
e
city
of
Guayaquil,
Ecuador
,
the
mass
transport
service,
named
Metro
via
System,
is
a
sol
ution
for
the
mobility
of
ground
transportation
passengers,
either
from
a
b
us
stop
(taking
a
feeder
b
us
to
arri
v
e
at
an
inte
gration
station
or
secondary
station)
or
directly
from
an
articulated
b
us
(from
a
secondary
station
for
mo
ving
to
an
specific
area).
An
articulated
b
us
transits
along
an
a
v
enue
or
street
lane,
occup
ying
an
e
xclusi
v
e
ground
lane.
The
density
of
an
articulated
b
us
is
nearly
160
passengers
in
maximum
capacity
,
which
is
higher
compared
to
the
number
of
people
tra
v
eling
in
non-e
xclusi
v
e
lanes.
These
lanes
are
shared
along
some
paths
wi
th
other
v
ehicles
with
the
authorization
of
the
transit
agenc
y
[1].
The
local
municipality
dictates
ordinances
for
the
use
of
the
e
xclusi
v
e
lanes,
applying
a
fi
ne
to
unauthorized
v
ehicles
that
enter
,
in
v
ade
or
obstruct
them
as
sho
wn
in
Figure
1,
which
consists
of
one
vital
minimum
w
age
[2–4]
(currently
equi
v
alent
to
USD
400),
and
tw
o
vital
minimum
w
ages
in
case
of
recidi
vism
[4].
Photo
capture
from
a
closed-circuit
surv
eillance
system
e
vidences
the
infraction,
notifying
to
the
of
fenders
through
electronic
means
if
possible,
or
communicating
them
at
the
time
the
y
approach
the
transit
agenc
y
for
an
y
process.
An
ar
gument
for
setting
an
e
xclusi
v
e
lane
is
the
ef
ficienc
y
in
terms
of
tra
v
eling
time
of
a
higher
dens
ity
of
users.
At
the
same
time,
e
v
en
if
the
route
frequencies
are
well
scheduled,
the
b
uses
usually
e
xperience
delays.
Contro
v
ersies
highlight
because
of
deri
v
ed
problems
in
non-e
xclusi
v
e
lanes
such
as
tra
v
eling
time
inef
ficienc
y
due
to
traf
fic
congestion,
which
may
increase
along
some
trunks
at
rush
hour;
more
fuel
consumption
and
J
ournal
homepage:
http://ijece
.iaescor
e
.com
Evaluation Warning : The document was created with Spire.PDF for Python.
5218
r
ISSN:
2088-8708
mechanical
parts
wear
due
to
continuous
jams;
noise
contamination
and
g
as
pollution.
Because
of
traf
fic
jam
e
xperiences,
lane
e
xclusi
v
eness
tak
es
out
during
some
periods
and
paths.
As
traf
fic
intensifies,
traf
fic
of
ficers
allo
w
lightweight
v
ehicles
to
transit
e
xclusi
v
e
lanes
[5]
as
traf
fic
intensifies,
whether
reducing
or
intensifying
congestion,
no
t
pro
viding
ef
fecti
v
eness
and
ef
ficienc
y
.
Researches
pursue
traf
fic
allocation
to
less
congested
routes,
optimizing
schemes
for
rerouting
traf
fic,
socializing
direct
benefits
to
dri
v
ers
for
enhancing
path
choices.
V
ehicular
congestion
issues
can
be
addressed
by
system
automation
to
pro
vide
alternati
v
e
paths
through
not
congested
roads
[6].
Figure
1.
Lane
in
v
asion
signboard
w
arning
[3]
(Note:
the
fine
is
not
updated)
This
w
ork
is
or
g
anized
as
follo
ws:
Section
2
presents
related
w
ork
aspects,
addressing
that
our
case
study
focuses
on
occupanc
y
controlling.
Section
3
depicts
the
system
design
co
v
ering
the
image
management
process
by
databases
for
detecting
occupanc
y
violations
and
for
the
notification
process
to
a
v
ehicle’
s
o
wner
.
Section
4
describes
the
analysis
of
the
results
considering
f
alse
posit
i
v
e
situations
in
captured
images.
In
the
end,
Section
5
presents
conclusions
and
lessons
learned
throughout
the
de
v
elopment
of
this
proposal.
2.
RELA
TED
W
ORK
A
related
w
ork
directly
associated
with
the
Ecuadorian
situation
for
controlling
the
lane
access
of
a
mass
transport
system,
particularly
in
the
city
of
Guayaquil,
has
not
been
de
v
eloped.
Ne
v
ertheless,
similar
studies
in
other
countries
are
a
v
ailable,
approaching
security
by
controlled
systems
for
persuading
drastic
lane
changes,
such
as
collision
a
v
oidance.
These
studies
look
forw
ard
to
a
confidence
pro
vision
to
decrease
stress
le
v
els
of
dri
v
ers,
helping
them
to
detect
situations
for
collision
pre
v
ention,
such
as
the
researches
presented
in
[7]
and
[8].
On
the
other
hand,
we
present
a
c
yber
-ph
ysical
system
that
allo
ws
lane
sharing
to
optimize
traf
fic
flo
w
circulation,
considering
lanes
that
are
on
a
ground-le
v
el
not
occupied
all
the
time.
3.
SYSTEM
DESIGN
The
system
design
implements
an
electronic
circuit
with
infrared
(IR)
diodes
for
v
ehicle
detection,
a
Raspberry
Pi
camera
[9]
for
photo
capturing,
and
a
Raspberry
Pi
3
board
for
a
v
alidation
process
of
the
occupanc
y
of
the
e
xclusi
v
e
lanes
as
sho
wn
in
Figure
2.
W
e
use
imported
libraries
that
w
ork
with
MySQL
databases
to
send
te
xt
messages
through
a
messaging
platform.
Both
the
photo
capturing
and
the
license
plate
v
alidation
allo
w
the
image
reading.
Three-dimensional
arrays
are
set
up
for
representing
colors
where
rectangles
are
formed
to
indicate
the
position
of
the
license
plate
and
the
color
font
of
the
signboards
that
o
v
erwrite
the
generated
images.
Moreo
v
er
,
the
system
reads
a
signal
from
an
electronic
circuit
with
IR
diodes,
which
act
as
an
infrared
barrier
to
detect
v
ehicles
that
ingress
to
an
e
xclusi
v
e
lane
for
taking
a
license
plate
photo.
It
is
necessary
to
connect
a
pin
of
the
Ra
spberry
Pi
3
for
the
corresponding
signal
con
v
e
ying
through
the
IR
diode
[10,
11]
by
the
General-Purpose-Input-Output
pin
(GPIO
23)
v
ariable
definition
to
count
the
fines
and
set
anti-re
bo
und
in
the
electronic
circuit.
The
Raspberry
Pi
camera
w
orks
as
an
array
for
photo
capture
using
searching
functions.
The
array
images
successions
depict
real-time
video
recorded
by
the
capture
function
that
tak
es
the
photo
at
the
instant
the
barrier
determines
it
[12,
13].
T
o
get
an
appropriated
image
is
necessary
to
adjust
some
camera
features
such
as
the
resolution
(320,
240),
the
angle
of
vie
w
(in
this
case
180
)
and
Int
J
Elec
&
Comp
Eng,
V
ol.
10,
No.
5,
October
2020
:
5217
–
5226
Evaluation Warning : The document was created with Spire.PDF for Python.
Int
J
Elec
&
Comp
Eng
ISSN:
2088-8708
r
5219
the
frame
rate
to
the
maximum,
aiming
to
capture
the
photos
as
f
ast
as
possible,
setting
it
the
maximum
of
15
fps.
The
programming
of
an
infinite
loop
allo
ws
the
photo-taking,
continuously
displaying
them
into
a
frame.
The
setup
of
an
anti-bounce
code
determines
if
a
v
ehicle
passes
through
the
barrier
for
t
he
photo
taking.
The
process
car
ries
out
for
once
until
the
v
ehicle
passes
the
barrier
as
sho
wn
in
Figure
3,
sa
ving
the
information
for
the
license
plate
recognition.
Figure
2.
Connecti
vity
of
electronic
de
vices
for
the
v
alidation
process
Figure
3.
Lab
test
with
a
mock-up
An
algorithm
processes
the
photos
based
on
neural
netw
orks
trained
for
character
recognition
[14].
Data
acquisition
and
an
OpenCV
library
training
function
return
a
Boolean
v
alue:
If
f
alse,
there
are
problems
in
the
file
loading
process;
otherwis
e,
if
true,
the
process
continues
asking
the
photo
captured
after
passed
the
infrared
barrier
,
sa
ving
it
into
the
coding
files
to
identify
the
license
plate,
creating
copies
to
generate
the
fine.
F
or
other
sensed
v
ehicle,
the
photo
o
v
erwrites
the
photo
captured
at
that
instant.
A
reading
function
of
the
OpenCV
library
loads
the
photo
image
[15],
and
a
detection
function
returns
a
re
gion
list
where
a
rectangular
area
for
license
plate
recognition
finds
alphanumeric
characters
whereas
a
preprocessing
function
performs
color
con
v
ersion
to
HSV
in
the
photo
image
by
e
xtracting
v
alues
[16]
from
the
OpenCV
library
[15].
A
se
gmentation
function
links
to
the
photo
image
channels,
setting
addition
and
subtraction
operations
to
maximize
contrasts
for
remo
ving
gray-scale
image
noises
[17],
con
v
erted
pre
viously
into
erosion
and
dilation.
Moreo
v
er
,
the
first
filtering
remo
v
es
the
Gaussian
noise,
making
it
possible
to
change
images
to
gray-scale
as
sho
wn
in
Figure
4(a)
for
comparing
each
pix
el
of
the
threshold
and
getting
its
binary
form
as
sho
wn
in
Figure
4(b).
W
ith
the
binary
image,
the
algorithm
starts
searching
the
license
plate
v
arying
the
gray-scale
image
and
marking
rectangles
until
finding
the
lar
gest
one
as
depicted
in
Figure
5.
Once
detected
and
e
xtracted
a
license
plate
image,
the
system
proceeds
to
turn
it
to
a
gray-scale
as
sho
wn
in
Figure
6(a).
Character
recognition
starts
with
a
license
plate
recognition
function
as
sho
wn
in
Figure
6(b).
Pre
viously
,
the
creation
of
a
back-up
is
necessary
due
to
the
binary
image
can
alter
the
contours.
Then,
the
analysis
of
v
alid
characters
by
a
neural
netw
ork
learning
pro
vides
the
files
loaded
when
the
process
s
tarts.
If
Cyber
-physical
system
based
on
ima
g
e
r
eco
gnition...
(V
.
Sanc
hez
P
adilla)
Evaluation Warning : The document was created with Spire.PDF for Python.
5220
r
ISSN:
2088-8708
it
is
into
the
deli
mited
area
of
the
license
plate,
the
recognition
analyzes
a
returning
character
list.
The
area
that
contains
the
non-null
characters
in
the
license
plate
re
gion
dra
ws
a
rectangle
around
it.
Both
console
and
dated
images
present
the
identified
license
plate
as
sho
wn
in
Figure
7,
displaying
into
an
original
image
windo
w
and
sa
ving
it
into
a
folder
if
the
process
analysis
determines
the
permission
status
has
e
xpired.
(a)
(b)
Figure
4.
License
plate
image,
(a)
in
gray-scale
(b)
in
binary
Figure
5.
Character
detection
process
(a)
(b)
Figure
6.
License
plate
e
xtracted,
(a)
in
gray-scale
(b)
for
character
recognition
Figure
7.
Character
recognition
simulation
Int
J
Elec
&
Comp
Eng,
V
ol.
10,
No.
5,
October
2020
:
5217
–
5226
Evaluation Warning : The document was created with Spire.PDF for Python.
Int
J
Elec
&
Comp
Eng
ISSN:
2088-8708
r
5221
A
database
that
stores
the
information
links
a
MySQL
connector
with
the
hostname
to
v
alidate
a
username
and
a
passw
ord,
specifying
the
database
name
that
sa
v
es
the
table
with
the
v
ehicle
information
that
passed
with
permission.
A
cursor
creation
is
necessary
to
specify
the
lecture
param
eters
for
information
depiction
by
console
re
g
arding
a
v
ehicle
that
has
occupied
the
e
xclusi
v
e
lane.
The
action
selects
license
plate
information
depicting
as
a
string
per
console
as
sho
wn
in
Figure
8,
and
checking
permission
parameters
and
its
use.
Additionally
,
the
system
v
erifies
whether
the
permission
is
e
xpired
or
not
by
comparing
its
v
alidation
period
with
the
current
date.
Re
gistry
commands
use
the
updated
information
for
depicting
information
through
the
database
(e.g.,
name,
e-mail,
cell
phone
number
,
permission
v
alidity)
of
the
authorized
v
ehicles
that
paid
the
fee
to
occup
y
the
e
xclusi
v
e
lanes
as
depicted
in
Figure
9.
An
object
calls
a
method
to
use
sending
ar
guments
such
as
subject
and
recipient
for
the
e-mail
notification
process,
pre
viously
associating
an
account
with
the
mail
serv
er
for
linking
the
object
ar
gument
for
the
e-mail
recipient.
Figure
8.
V
ehicle
o
wner
information
described
in
Spanish
Figure
9.
Re
gistered
v
ehicles
information
F
or
loading
captured
images
with
the
modifications
made
in
the
analysis
process,
another
object
i
ndi-
cates
the
path
where
the
image
loads,
adding
a
file
name
to
the
header
and
continuing
with
an
authentication
process
by
the
configuration
of
an
SMTP
client
session
object
that
sends
e-mails.
Furthermore,
a
command
with
parameter
information
of
the
sender
(administrator)
and
the
recipient
(database)
sends
an
e-mail
retrie
v-
ing
the
image
with
rele
v
ant
information
of
the
v
ehicle
that
in
v
aded
the
e
xclusi
v
e
lane
as
sho
wn
in
Figure
10,
closing
the
pre
vious
object
created.
Cyber
-physical
system
based
on
ima
g
e
r
eco
gnition...
(V
.
Sanc
hez
P
adilla)
Evaluation Warning : The document was created with Spire.PDF for Python.
5222
r
ISSN:
2088-8708
Figure
10.
E-mail
with
information
of
the
fine
Short
message
service
(SMS)
noti
fications
ask
some
parameters
re
gistered
in
the
database,
such
as
information
of
a
cloud
account
with
the
respecti
v
e
authorization
[18].
A
client
parameter
function
object
sends
a
message
specifying
the
recipient
and
the
cell
phone
number
re
gist
ered
in
the
system
w
arning
to
the
o
wner
that
his/her
v
ehicle
has
occupied
an
e
xclusi
v
e
lane
with
an
e
xpired
permission
as
sho
wn
in
Figure
11.
After
sending
a
notification,
another
table
adds
information
about
the
number
of
the
generated
fines
such
as
the
date
and
time
of
the
infraction,
image
name,
user
information,
and
the
license
plate
as
depicted
in
Fi
g
ur
e
12.
Case
studies
where
SMS
notifications
were
deplo
yed
in
dif
ferent
scenarios
with
the
respecti
v
e
results
are
a
v
ailable
in
[19–24].
Figure
11.
SMS
with
information
of
the
fine
Figure
12.
Database
with
information
of
fined
v
ehicles
4.
AN
AL
YSIS
OF
RESUL
TS
T
ests
were
carried
out
in
outdoor
en
vironments
during
daytime
hours
to
ease
the
reading
of
the
license
plates,
with
a
scenario
that
resembles
e
xternal
conditions.
The
results
of
the
captured
images
had
v
ariations
due
to
either
the
position
or
the
type
of
character
in
the
plate.
A
dif
ferent
camera
than
the
one
used
in
the
prototype
system
to
check
the
coding
ef
ficienc
y
captured
the
images
of
the
license
plates.
At
the
mome
n
t
of
reading
an
image
a
fter
loading
pre
vious
ones,
the
program
recognized
characters
that
did
not
e
xist,
presenting
a
f
alse
positi
v
e
as
sho
wn
in
Figure
13.
F
or
e
xample,
depending
on
the
font
used
in
the
license
plate,
letters
with
closed-form
lik
e
the
letter
C
and
G
depicted
a
distortion
during
the
recognition
as
sho
wn
in
Figure
14,
deli
v
ering
wrong
characters
such
as
0
and
8
,
respecti
v
ely
.
Int
J
Elec
&
Comp
Eng,
V
ol.
10,
No.
5,
October
2020
:
5217
–
5226
Evaluation Warning : The document was created with Spire.PDF for Python.
Int
J
Elec
&
Comp
Eng
ISSN:
2088-8708
r
5223
Figure
13.
F
alse
positi
v
e
depiction
Figure
14.
Closed-form
characters
depiction
Alteration
during
the
recognition
process
occurs
depending
on
the
inclination
angle
of
the
captured
image
[25,
26],
limiting
the
recognition
area
or
altering
the
charact
ers.
There
are
successful
cases
where
an
image
is
captured
with
either
a
front
or
a
rear
profile
without
inclinations
and
with
standard
font
letters
with
adequate
accuracies,
such
as
the
ones
mentioned
in
[27].
In
our
case
study
,
a
f
actor
that
af
fects
the
accurac
y
relationship
w
as
the
inclination
angle
during
the
capturing
of
the
characters
as
sho
wn
in
the
Figures
15(a)
and
15(b).
T
able
1
depicts
a
match
percentage
of
t
he
algorithm
when
recognizing
the
characters
from
all
the
captures
done.
(a)
(b)
Figure
15.
Captured
images,
(a)
with
altered
information
(b)
with
correct
information
T
able
1.
Sample
coincidences
License
plate
Recognized
characters
Wrong
characters/T
otal
Accurac
y
(%)
GSM-9640
OSM-9640
1/7
85.71
GSN-5505
OSN-5505
1/7
85.71
GO
Y
-515
CO
Y
-SI5
3/6
50.00
PCN-6934
PON-6934
1/7
85.71
GSP-7329
8SP-73Z9
2/7
71.43
GRL-992
RL-99
2/6
66.67
GSR-7446
6SR-7145
3/7
57.14
MDF-275
MDF-275
0/6
100.00
PB
A-1827
PB
A-1827
0/7
100.00
Cyber
-physical
system
based
on
ima
g
e
r
eco
gnition...
(V
.
Sanc
hez
P
adilla)
Evaluation Warning : The document was created with Spire.PDF for Python.
5224
r
ISSN:
2088-8708
5.
LESSONS
LEARNED
AND
CONCLUSIONS
The
prototype
w
as
b
uilt
using
Python
and
Raspberry
Pi
3
with
neural
netw
ork
algorit
hms.
According
to
the
tests
carried
out,
we
obtained
an
accurac
y
rate
a
v
erage
of
78.04%,
a
percentage
not
reliable
for
this
sort
of
application.
Future
w
ork
should
include
algorithm
testing
that
not
only
w
ork
with
images
in
the
front
and
rear
profile
b
ut
also
to
correct
inclinations
for
a
better
image
cleaning
before
t
he
characters’
e
xtraction.
On
the
other
hand,
the
implementation
costs
can
diminish
by
either
doing
the
detection
by
Python
codes
or
decreasing
the
capturi
ng
time
of
the
lice
n
s
e
plate.
Moreo
v
er
,
the
Raspberry
Pi
camera
w
as
a
con
v
enient
tool
for
capturing
images
due
to
its
resolution
of
5
Me
g
apix
els
and
static
image
processing
of
up
to
2592x1944
pix
els.
Ho
we
v
er
,
other
options
can
be
considered
for
capturing
motion
images,
to
attain
instant
captures
with
higher
resolution
and
e
xpanded
co
v
erage
to
o
v
ercome
speed
constraints.
Re
g
arding
the
te
xt
message
notification
platform,
its
scope
w
as
limited
insomuch
as
it
w
as
not
able
to
w
ork
w
orldwide,
co
v
ering
to
cell
phones
of
the
mobile
netw
ork
operators
w
orking
in
Ecuador
.
This
situation
can
be
o
v
ercome
with
the
implementation
of
GSM
modules.
Also,
as
this
project
w
orks
through
the
Internet,
it
is
required
a
system
connection
with
enough
bandwidth
to
ensure
notification
deli
v
ery
in
case
of
generation
of
a
fine.
REFERENCES
[1]
T
en
sections
of
the
e
xclusive
lane
of
the
Metr
o
via
System
in
Guayaquil
ar
e
shar
ed;
Diez
tr
amos
del
carril
e
xclusivo
de
la
Metr
o
via
en
Guayaquil
son
compartidos
,
Diario
El
Uni
v
erso,
Oct.
2017.
[Online].
A
v
ailable:
www
.eluni
v
erso.com/guayaquil/2017/10/23/nota/6446788/diez-tramos-carril-
e
xclusi
v
o-metro
via-
guayaquil-son-compartidos.
[2]
V
alue
of
infr
actions
and
faults
incr
eases
by
s
alary;
V
alor
de
infr
acciones
y
faltas
sube
por
salario
,
Diario
El
Uni
v
erso,
Jan.
2018.
[Online].
A
v
ailable:
www
.eluni
v
erso.com/guayaquil/2018/01/01/nota/
6546166/v
alor
-
infracciones-f
altas-sube-salario
[3]
F
r
om
today
,
a
fine
applies
for
occupying
the
Metr
o
via
lane
in
A
venida
de
las
Americas;
Desde
hoy
,
multa
por
in
vadir
carril
de
Metr
o
via
en
A
venida
de
las
Americas
,
Diario
El
Uni
v
erso,
Jun.
2017.
[Online].
A
v
ailable:
www
.eluni
v
erso.com/noticias/2017/06/12/nota/6227956/ho
y-multa-in
v
adir
-carril
[4]
Thir
d
r
eform
of
the
r
eforming
or
dinance
and
coding
of
the
or
dinance
that
cr
eates
and
r
e
gulates
the
inte
gr
ated
system
of
urban
mass
tr
ansport
of
the
City
of
Guayaquil
-
Metr
o
via
Syst
em
,
Gaceta
Ofici
al,
Dec.
2014.
[Online].
A
v
ailable:
https://guayaquil.gob
.ec/Gacetas/Periodo
2014-2019/Gaceta
17.pdf
[5]
The
occupancy
of
the
Metr
o
via
lanes
occur
s
at
thr
ee
places;
La
in
vasion
del
carril
de
la
Metr
o
via
se
da
en
tr
es
sitios
,
Diario
Expreso,
Jun.
2018.
[Online].
A
v
ailable:
www
.e
xpreso.ec/guayaquil/atm-metro
via-
carriles-
transito
v
ehicular
-multas-paraderos-XD2216356
[6]
D.
Ni,
H.
Liu,
W
.
Ding,
Y
.
Xie,
H.
W
ang,
H.
Pishro-Nik,
and
Q.
Y
u,
“Cyber
-ph
ysical
inte
gration
to
connect
v
ehicles
for
transformed
transportation
safety
and
ef
ficienc
y
,
”
in
International
Confer
ence
on
Industrial,
Engineering
and
Other
Applications
of
Applied
Intellig
ent
Systems
.
Springer
,
2012,
pp.
88–94.
[7]
H.
Muslim,
M.
Itoh,
and
M.-P
.
P
acaux-Lemoine,
“Dri
ving
with
shared
control:
Ho
w
support
system
performance
impacts
safety
,
”
in
2016
IEEE
International
Confer
ence
on
Systems,
Man,
and
Cybernetics
(SMC)
.
IEEE,
2016,
pp.
000
582–000
587.
[8]
T
.
Aiza
w
a
and
M.
Itoh,
“Shared
automation
of
lane
change
for
a
v
oiding
forw
ard
obstacle,
”
in
2016
IEEE
International
Confer
ence
on
Systems,
Man,
and
Cybernetics
(SMC)
.
IEEE,
2016,
pp.
003
782–003
787.
[9]
T
.
C.
W
ilk
es,
A.
J.
McGonigle,
T
.
D.
Pering,
A.
J.
T
agg
art,
B.
S.
White,
R.
G.
Bryant,
and
J.
R.
W
illmott,
“Ultra
violet
imaging
with
lo
w
cost
smartphone
sensors:
De
v
elopment
and
applicati
on
of
a
raspberry
pi-
based
uv
camera,
”
Sensor
s
,
v
ol.
16,
no.
10,
p.
1649,
2016.
[10]
R.
A.
P
on
gui
llo
and
C.
V
.
Medina,
“Using
open
source
embedded
hardw
are
and
softw
are
tools
in
au-
tomatic
control
from
mathematical
model,
”
in
Pr
oceedings
of
the
W
orld
Congr
ess
on
Engineering
and
Computer
Science
,
v
ol.
1,
2016.
[11]
R.
A.
Ponguillo,
“Deplo
yment
of
a
lo
w
cost
fuzzy
controller
using
open
source
embedded
hardw
are
and
softw
are
tools,
”
in
Pr
oceedings
of
the
International
MultiConfer
ence
of
Engineer
s
and
Computer
Scientists
,
v
ol.
2,
2018.
[12]
C.
S.
Baidal,
D.
S.
Salazar
,
V
.
S.
P
adilla,
R.
L.
Estrada,
and
N.
X.
Arreag
a,
“Design
of
a
wireless
sensor
netw
ork
to
detect
car
accidents
on
highw
ays,
”
in
2018
International
Symposium
on
Networks,
Computer
s
and
Communications
(ISNCC)
.
IEEE,
2018,
pp.
1–6.
Int
J
Elec
&
Comp
Eng,
V
ol.
10,
No.
5,
October
2020
:
5217
–
5226
Evaluation Warning : The document was created with Spire.PDF for Python.
Int
J
Elec
&
Comp
Eng
ISSN:
2088-8708
r
5225
[13]
C.
V
accaro,
D.
T
orres,
A.
Collaguazo
Jaramillo,
W
.
V
elasquez,
V
.
Sanchez
P
adilla,
R.
Ponguillo,
and
M.
Chalen,
“De
v
elopment
of
a
road
w
arning
monitoring
system
by
deplo
yment
of
sensors
netw
ork
and
reacti
v
e
signaling,
”
DEStec
h
T
r
ansactions
on
Computer
Science
and
Engineering
,
no.
MMST
A,
2017.
[14]
M.
Be
yeler
,
Mac
hine
Learning
for
OpenCV
.
P
ackt
Publishing
Ltd,
2017.
[15]
J.
Ho
wse,
P
.
Joshi,
and
M.
Be
yeler
,
Opencv:
computer
vision
pr
ojects
with
python
.
P
ackt
Publishing
Ltd,
2016.
[16]
G.
Sara
v
anan,
G.
Y
amuna,
and
S.
Nandhini,
“Real
time
implementation
of
r
gb
to
hsv/hsi/hsl
and
its
re
v
erse
color
space
models,
”
in
2016
International
Confer
ence
on
Communication
and
Signal
Pr
ocessing
(ICCSP)
.
IEEE,
2016,
pp.
0462–0466.
[17]
N.
Jamil,
T
.
M.
T
.
Sembok,
and
Z.
A.
Bakar
,
“Noise
remo
v
al
and
enhancement
of
binary
images
using
morphological
operations,
”
in
2008
International
Symposium
on
Information
T
ec
hnolo
gy
,
v
ol.
4.
IEEE,
2008,
pp.
1–6.
[18]
P
.
V
amsikrishna,
S.
D.
K
umar
,
S.
R.
Hussain,
and
K.
R.
Naidu,
“Raspberry
pi
controlled
sms-update-
notification
(sun)
system,
”
in
2015
IEEE
international
confer
ence
on
electrical,
computer
and
communi-
cation
tec
hnolo
gies
(ICECCT)
.
IEEE,
2015,
pp.
1–4.
[19]
N.
H.
A.
Aziz,
W
.
N.
W
.
Muhamad,
N.
A.
W
ahab,
A.
J.
Alias,
A.
T
.
Hashim,
and
R.
Mustaf
a,
“Real
time
monitoring
critical
parameters
in
tissue
culture
gro
wth
room
with
sms
alert
system,
”
in
2010
International
Confer
ence
on
Intellig
ent
Systems,
Modelling
and
Simulation
.
IEEE,
2010,
pp.
339–343.
[20]
M.
F
.
Sikder
,
S.
Halder
,
T
.
Hasan,
M.
J.
Uddin,
and
M.
K.
Bao
w
aly
,
“Smart
disaster
notification
system,
”
in
2017
4th
International
Confer
ence
on
Advances
in
Electrical
Engineering
(ICAEE)
.
IEEE,
2017,
pp.
658–663.
[21]
N.
A.
Samsudin,
S.
K.
A.
Khalid,
A.
M.
Y
usof
f,
M.
N.
Ihkasan,
and
Z.
Senin,
“Procedure
automation
with
immediate
user
notification:
A
case
study
,
”
in
2011
IEEE
Symposium
on
Business,
Engineering
and
Industrial
Applications
(ISBEIA)
.
IEEE,
2011,
pp.
253–258.
[22]
R.
A.
Ponguillo,
V
.
S.
P
adilla,
and
D.
M.
Ordonez,
“Alarm
system
based
on
gsm
netw
ork
and
fpg
a
using
nios
ii
embedded
processor,
”
International
J
ournal
of
Engineering
and
Advanced
T
ec
hnolo
gy
,
v
ol.
8,
no.
6,
pp.
4144–4148,
2019.
[Online].
A
v
ailable:
www
.scopus.com
[23]
K.
Lappanitchayakul,
“De
v
elopment
of
email
and
sms
based
notification
system
to
detect
abnormal
net-
w
ork
conditions:
A
case
study
of
f
aculty
of
b
usiness
administration,
rajamang
ala
uni
v
ersity
of
technology
phra
nakhon,
thailand,
”
in
2018
International
Confer
ence
on
Intellig
ent
Informatics
and
Biomedical
Sciences
(ICIIBMS)
,
v
ol.
3.
IEEE,
2018,
pp.
98–105.
[24]
A.
A.
Lapada,
“De
v
elopment
and
e
v
aluation
of
a
management
information
system
with
sms
notification
technology
in
a
philippine
military
camp,
”
International
J
ournal
of
Engineering
and
Advanced
T
ec
hnolo
gy
,
v
ol.
8,
no.
5,
pp.
1174–1177,
2019.
[Online].
A
v
ailable:
www
.scopus.com
[25]
S.
T
odoro
vic
and
N.
Ahuja,
“Unsupervised
cate
gory
modeling,
recognition,
and
se
gmentation
in
images,
”
IEEE
T
r
ansactions
on
P
attern
Analysis
and
Mac
hine
Intellig
ence
,
v
ol.
30,
no.
12,
pp.
2158–2174,
2008.
[26]
N.
B.
A.
Mustaf
a,
F
.
B
akri,
and
S.
K.
Ahmed,
“Identification
of
image
angle
using
projecti
v
e
trans-
formation:
application
to
banana
images,
”
in
2014
IEEE
REGION
10
SYMPOSIUM
.
IEEE,
2014,
pp.
408–413.
[27]
K.
Y
ogheedha,
A.
Nasir
,
H.
Jaaf
ar
,
and
S.
Mamduh,
“
Automatic
v
ehicle
license
plate
recognition
sys-
tem
based
on
image
processing
and
template
matching
approach,
”
in
2018
International
Confer
ence
on
Computational
Appr
oac
h
in
Smart
Systems
Design
and
Applications
(ICASSD
A)
.
IEEE,
2018,
pp.
1–8.
BIOGRAPHIES
OF
A
UTHORS
V
.
Sanchez
P
adilla
is
w
orking
at
the
Escuela
Superior
Politecnica
del
Litoral
as
an
adjunct
lecturer
in
the
T
elematics
Engineering
Program
a
nd
collaborates
with
the
academic
committee
of
the
master’
s
program
in
telecommunications.
He
recei
v
ed
a
master’
s
de
gree
in
telecommunications
engineering
from
Geor
ge
Mason
Uni
v
ersity
,
USA
and
a
master’
s
de
gree
in
quality
management
fr
om
the
Escuela
Superior
Politecnica
del
Litoral,
Ecuador
.
His
research
interests
focus
on
public
polic
y
,
wireless
sen-
sor
netw
orks,
and
engineering
science
education.
Memberships
include
IEEE
ComSoc,
IEEE
TEMS,
Institute
of
Research
Engineers
and
Doctors,
and
the
Ecuadorian
Re
gional
Colle
ge
of
Electrical
and
Electronics
Engineers.
Cyber
-physical
system
based
on
ima
g
e
r
eco
gnition...
(V
.
Sanc
hez
P
adilla)
Evaluation Warning : The document was created with Spire.PDF for Python.
5226
r
ISSN:
2088-8708
R.A.
P
onguillo
is
w
orking
to
w
ards
his
doctorate
in
engineering
in
the
Department
of
Industrial
Systems
Engineering
and
Product
Design
at
Ghent
Uni
v
ersity
,
Belgium.
He
obtained
his
master’
s
de
gree
in
management
inform
ation
system
from
the
Escuela
Superior
Politecnica
del
Litoral
(ESPOL),
Ecuador
.
At
ESPOL,
he
w
ork
ed
as
an
adjunct
lecturer
in
the
Digital
Electronics
Depart-
ment
of
the
F
aculty
of
Electrical
and
Computer
Engineering.
His
research
interests
lie
primarily
in
fuzzy
logic,
technology
de
v
elopment
by
FPGA,
and
embedded
systems.
Memberships
include
IEEE,
IAENG,
and
the
Re
gional
Colle
ge
of
Electrical
and
Electronics
Engineers
from
Ecuador
.
A.A.
Abad
is
w
orking
as
biomedical
engineer
at
Sistagrosa,
part
of
the
Elicrom
group
in
the
city
of
Guayaquil,
Ecuador
.
He
obtained
an
engineering
de
gree
in
electronics
and
telecommunications
from
t
he
Escuela
Superior
Politecnica
del
Litoral
(ESPOL),
Ecuador
.
While
at
ESPOL,
he
w
as
teaching
assistant
of
the
laboratories
of
microcontrollers
and
digital
systems.
His
e
xpertise
includes
VHDL
program
ming
with
research
interests
in
electronics
applications,
embedded
systems,
and
bio-engineering.
He
holds
technical
certifications
from
dif
ferent
v
endors.
L.E.
Salas
is
w
orking
as
a
lecturer
at
the
CREAR
Institute,
teaching
topics
related
to
ph
ysics,
mathematics,
and
robotic
s,
holding
the
position
of
Math
&
T
echnology
Coordinat
or
.
She
obtained
her
engineering
de
gree
in
electronics
and
telecom
munications
from
the
Escuela
Superior
Politecnica
del
Litoral
(ESPOL),
Ecuador
.
She
has
attended
se
v
eral
courses
related
to
electronics,
operating
systems,
and
programming.
Int
J
Elec
&
Comp
Eng,
V
ol.
10,
No.
5,
October
2020
:
5217
–
5226
Evaluation Warning : The document was created with Spire.PDF for Python.