TELKOM
NIKA Indonesia
n
Journal of
Electrical En
gineering
Vol.12, No.6, Jun
e
201
4, pp. 4353 ~ 4
3
6
0
DOI: 10.115
9
1
/telkomni
ka.
v
12i6.487
2
4353
Re
cei
v
ed O
c
t
ober 2
2
, 201
3; Revi
se
d Decem
b
e
r
31, 2013; Accept
ed Ja
nua
ry 2
3
, 2014
Construction and Application of Virtual Reality
Geographic Information System
Ke Xi-Lin*
1,2,3
, Guo Qing-Sheng
1
, Zha
ng Yue-Peng
2,3
,Zhou Ya
n
2,3
1
School of Res
ource a
nd Env
i
ronme
n
t Scien
c
e of W
u
Han U
n
iversit
y
, W
u
h
an 43
00
79, Chi
n
a
2
State
Ke
y
Lab
orator
y of Geo-inform
at
ion En
gi
neer
i
ng, Xi’a
n 71
00
54, Chi
n
a
3
Xi’a
n R
e
searc
h
Institute of Surve
y
in
g
and Ma
ppi
ng
,
Xi
’a
n 710
05
4, Chin
a
*Corres
p
o
ndi
n
g
author, em
ail
:
kexil
i
n
_
tek@1
63.com
A
b
st
r
a
ct
In the
pa
per,
virtual
re
ality
ge
ogra
p
h
i
c i
n
formatio
n
sy
stem
is d
i
scu
ssed
and
the
relat
e
d
techno
lo
gies i
n
volves the co
n
c
ept, character
i
stics,
classific
a
tion, key tech
nol
ogy, and th
e ma
in ap
plic
at
ion
and
dev
el
op
ment situ
atio
n, i
t
ana
ly
z
e
s th
e
virtual
re
ality
techn
o
lo
gy
a
nd th
e n
e
cess
ity of co
mbin
i
n
g
geographic
a
l information system
te
chnology. As to the
design and dev
elopm
ent pr
oc
ess of the whole
system
, inc
l
uding the system
'
s
user
interfac
e, system
r
e
quirem
ents, th
e structure of th
e system
des
ign,
system dev
elo
p
ment mode
l
a
nd
re
late
d
tec
h
nol
ogy are disc
u
ssed
in
this
p
aper. In v
i
ew
of
the virtu
a
l r
e
a
l
i
t
y
geographic information system
technol
ogy,
the 3-D (three
Dim
ens
ional) s
patial data
acquis
ition, 3-D data
mo
de
l and d
i
g
i
tal terrain
mo
d
e
l an
d dig
i
tal e
l
evati
on
mod
e
l
and 3-D d
a
ta of level of deta
il mode
l an
d th
e
express
i
on of
3-D data vis
uali
z
a
t
io
n,
etc, on the bas
is
of
theoretica
l
research, w
i
th the he
lp of
a
professi
ona
l vi
rtual rea
lity mode
lin
g softw
are MultiG
en C
r
eator, it bui
lts the
scenic
spot of the 3-
D
envir
on
me
nt. A
nd
it pr
esents
the v
i
rt
ual
sce
n
e
mo
del
in
g pr
o
c
ess o
n
th
e
ba
sis of M
u
ltiGe
n
Creator
co
mp
l
e
x
mo
de
lin
g, the i
m
p
l
e
m
e
n
tatio
n
plan, scen
e
mode
lin
g metho
d
are als
o
giv
e
n out.
Ke
y
w
ords
:
virtual reality, geogra
phy infor
m
ation system
, c
onstruction, application
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
GIS (Geo
gra
phic Info
rmat
ion System,
GIS) is the
bord
e
rli
ne su
bject of ge
o
g
rap
h
y,
surveyin
g, ca
rtography, re
mote sen
s
in
g
,
it
now is develope
d towa
rds soci
alization.
Virtual Realit
y is refe
rre
d to as V
R
, this
word
is
a prof
essor
at Ame
r
ica
n
scie
ntist
s
Jaro
n
Lanie
r
first p
u
t forward in
the ea
rly 1
9
8
0
s, al
so
kno
w
n
as virtual
environme
n
t, artificial
re
al
ity,
simulato
r te
chnolo
g
y, etc.
Its origi
n
can
be tra
c
e
d
b
a
ck in the
19
6
0
s, u
n
til the 1
990
s it be
co
me
the co
ncern
of the scie
nce
and e
n
g
inee
ring te
chnolo
g
y. With the ra
pid
developm
ent
of
informatio
n scien
c
e, virtu
a
l reality te
chnolo
g
y is b
a
se
d on
co
mputer
scien
c
e, math
ema
t
ics,
mech
ani
cs, aco
u
sti
cs, o
p
tics,
m
e
cha
n
ics,
bi
olo
g
y, and
even t
he ae
stheti
c
and th
e
so
cial
sci
ences and other di
sci
pli
nes.
Com
p
rehensive utilization of the
com
puter virtual reality, tactile
feedba
ck, virtual stereo technolo
g
y, such as hig
h
ly
re
alistically can
simulate the
behavio
r of man
in the
natu
r
al
environme
n
t. Thi
s
kind
of
virtual
envi
r
o
n
ment
i
s
a ki
nd
of environ
ment
g
ene
rat
e
d
by compute
r
, it can be a re
al simulatio
n
worl
d, and ca
n also b
e
a vision of the world [1, 2].
In the contin
uou
s develo
p
m
ent of virtual realit
y tech
nology, the rese
arch of it involves
the content
of relate
d di
sci
pline
s
, su
ch
as
co
mp
uters, cro
s
s and synthe
si
s,
this
compl
e
xity
make
s it
so f
a
r, there is
n
o
uniform defi
n
ition.
Simply spe
a
ki
ng, it can
be u
nde
rstood
as th
ro
ugh
the e
s
tabli
s
h
m
ent of
a virt
ual e
n
viron
m
ent, and
ma
ke the
pe
rson
feels
as in
re
al envi
r
onm
e
n
t.
No
w its re
co
gnized préci
s
ed definition i
s
thro
ugh u
s
i
ng co
mpute
r
techn
o
logy a
s
the co
re of
the
mode
rn high
-tech to gen
e
r
ate re
alisti
c visual, hea
rin
g
, touch, the
integration o
f
the particul
a
r
rang
e of virtual environ
me
nt, it can through a n
a
tural way and from entity objects in a virt
ual
environ
ment i
n
tera
ction, th
us it can p
r
o
duce t
he real
environ
ment
of feeling a
n
d
experi
e
n
c
e
[3
-
6].
The
esse
nce
of virtual
reality tech
no
logy is
invisi
ble ima
g
inati
on into
the
form of
scenarios, it emphasi
zes t
he visibility of lifelike and
sci
ence, it
is
a key basi
c f
eatures are it
s
three
cha
r
a
c
teristics of
“I”, namely th
at the
Immers
ion charac
teris
t
ic
, Interac
t
ion features
,
Imagination f
eature
s
[7, 8].
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 6, June 20
14: 4353 – 4
360
4354
2. Nece
ssit
y
of Virtual Re
alit
y
and Geography
Information Sy
stem
VR an
d GIS
system
integ
r
ation b
e
com
e
ri
ch vi
rtual
reality world
and th
e b
e
st
way to
enha
nce the
geog
rap
h
ic i
n
formatio
n system pe
rfor
mance. VR
system ha
s p
e
rfect inte
ra
ctive
ability and inspired
by the idea of multi-d
i
mensi
onal inf
o
rmatio
n envi
r
onm
ent. The
GIS has go
o
d
perfo
rman
ce
in the processing
of spati
a
l relati
o
n
s, t
he two
-
dime
nsio
nal techn
o
logy, re
sea
r
ch
conte
n
ts, and
methods h
a
ve become
the
application p
r
ospe
ct and trend.
(1) As to GIS, integration
of VR tec
hnology can ex
tend existing
GIS graphic display
function, GIS
theory and
rich
c
onn
otation, expa
ndi
ng the
area
of
GIS ap
plication; it i
s
t
h
e
inevitable tre
nd of spatial
informatio
n visuali
z
at
io
n virtual exp
r
e
ssi
on. VR sy
ste
m
is a
kind
of
man-m
a
chine
relation
shi
p
, throug
h the
visual, hea
ri
ng, touch, a
nd so on, th
e user of it
can
immersive fe
el the multi-di
mensi
onal i
n
formatio
n in
te
ractive visual
simulatio
n
system of the
real
worl
d, it can
meet the current
exploratio
n and
study o
f
earth scie
nce, now the
GIS is develop
e
d
from
simple
descri
p
tion t
o
the
whol
e
earth
spa
c
e
of high
an
d
new inform
ation technol
og
y
dire
ction.
(2) As V
R
system, GIS is
the tool in sol
v
ing the prob
lem of user
s
in the virtual
scene.
Multi-dime
nsi
onal informa
t
ion spa
c
e
gene
rat
ed b
y
the VR
system i
s
called a virtual
environ
ment.
Although virtu
a
l enviro
n
me
nt with ea
rth sci
en
ce can
p
r
odu
ce 3-D
g
eologi
cal
virtu
a
l
environ
ment
with realisti
c
visual effe
cts,
but
wh
en th
e sp
ace is to
o larg
e o
r
co
mplicate
d
, users
can only see
obje
c
ts within
the scop
e
of vision, it
is lack of integ
r
al feeling to the environ
ment, it
is often in
roaming l
o
st
feeling a
nd
explorat
io
n p
r
ocess. At this mom
ent it need
s the
2-D
navigation sy
stem co
rresp
ond
s with the
3-D virtual world, so
the e
x
pressio
n
of the 2-D GIS are
indispen
sabl
e
.
(3) F
r
om the data level, in
orde
r to build
lar
ge-scale
scen
e in the real wo
rld, the virtual
scene m
odeli
ng of obje
c
t
s
can
be ad
o
p
ted into
a u
n
ified geo
gra
phic
sp
atial databa
se in t
he
referen
c
e
system, and the
unified ma
na
gement m
u
st
be op
erate
d
by GIS syste
m
, GIS has g
r
eat
advantag
es a
nd potential i
n
3-D,
real-ti
m
e dynami
c
, multi-re
sol
u
tion and hu
ge
amount
s of data
spa
c
e
scen
e simplifi
c
ati
on, com
p
ression,
stora
g
e
and q
u
e
r
y structu
r
e,
extraction
and
informatio
n re
cov
e
ry
, etc.
In the geologi
cal re
se
arch
and practi
ce, peopl
e have grad
ually real
ized the impo
rtance
of com
b
inati
on of
GIS a
nd VR sy
ste
m
in
solv
ing
pra
c
tical p
r
oblem
s, thu
s
forme
d
the
new
resea
r
ch field
of VRGIS. Integratio
n of
techn
o
logy in
tegration
will
be no
do
ubt
applie
d for th
e
further
combi
nation of the two sy
stem
s a
nd
guid
e
its d
e
velopme
n
t directio
n [9-1
4].
3. Design of
Virtual Reality
Geographi
c Informatio
n Sy
stem
In orde
r to m
a
ke the
syste
m
to be fully func
tio
nal an
d
stru
ctural opt
imized, the fo
llowing
prin
ciple
sho
u
ld be ad
opte
d
.
(1) Simpl
e
practical prin
cip
l
e: the syste
m
shoul
d be
as sim
p
le a
s
possibl
e, and
suitable
for different l
e
vels an
d the
kno
w
led
ge structu
r
e of the use
r
. It sho
u
ld have con
c
ise de
sign a
nd
friendly interf
ace, ea
sy op
eration
cha
r
a
c
teri
stics.
(2) Standa
rd
prin
cipl
e: the sy
stem
sh
ould
m
eet th
e ba
sic requi
reme
nt of V
R
GIS an
d
standard,
system data types
, such
as
coding, schem
a
tic illust
ration should
conf
orm to the st
ate
and
ind
u
stry stand
ard req
u
irem
ents.
(3) Pro
s
p
e
cti
v
e pri
n
cipl
e: the info
rma
t
ion technol
o
g
y is develo
p
ing ve
ry fa
st, the
upgradin
g
of
hardware
is also
very
ra
pidly, softw
a
r
e version
up
grad
es al
so
very fast, in
the
desi
gn, the full con
s
ide
r
atio
n to the developm
ent tre
n
d
of technolo
g
y
should b
e
g
i
ven.
(4) The
secu
rity and
stabi
lity princi
ple:
sy
stem
ha
s
certai
n fault t
o
lera
nce a
n
d
goo
d
remin
ders, so
me simpl
e
mistake w
ill not
lead to the sy
stem cra
s
h.
(5) T
he ratio
n
a
lity principl
e: in the config
ur
ation of
sof
t
ware
and h
a
r
dware, the software
and ha
rd
wa
re sho
u
ld fully con
s
ide
r
pe
rforman
c
e
rati
o and the rationality of the config
urati
on,
and shoul
d
also
con
s
i
der the
real
ization of
th
e function
and the
sp
ecific a
ppli
c
ation
requi
rem
ents.
The m
a
ximum vario
u
s reso
urce
s co
nsumi
ng pro
b
lem
of com
puter hardwa
r
e
perfo
rman
ce
sho
u
ld al
so b
e
cared of.
(6) Th
e p
r
in
ciple of
ope
n
ness
and
scalabilit
y: the
system
data
has exchan
g
eability,
provide in
du
stry popula
r
d
a
ta transmission and ex
ch
ange fun
c
tion
. Modular d
e
s
ign
sho
u
ld be
adopte
d
, ea
ch fun
c
tion
mo
dule i
s
i
nde
p
ende
nce, an
d
the m
odul
e
of ch
ang
e
wil
l
not
brin
g m
u
ch
impact to the
system.
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Const
r
u
c
tion and Appli
c
a
t
ion of Virtual Rea
lit
y Geo
graphi
c Inform
ation System
(Ke Xi-Lin)
4355
4. Establish
m
ent of Virtual Realit
y
G
e
ograp
h
ic Informatio
n Sy
stem
With the dev
elopme
n
t of comp
uter g
r
a
phics te
chnol
ogy in re
cent
years, TIN
(i
rre
gula
r
triangle n
e
t) in 3D visuali
z
ation also g
o
t more
and m
o
re wi
dely used. TIN has t
he advantag
e
of
less data red
unda
ncy, it is better to take the terra
in feature
s
into accou
n
t, it can fully reflect the
terrai
n
com
p
l
e
xity details,
etc. its wea
k
ness is
the m
o
re complex
algorith
m
imp
l
ementation, the
spa
c
e o
p
e
r
ati
on and it
s sto
r
age. Th
e G
R
ID modeli
ng i
s
anoth
e
r m
e
thod to built DEM. It
feature
s
is simpl
e
stru
cture, Be
cau
s
e of the spa
c
e point
in the plane carri
e
d out in accorda
n
ce with the
rule
s of gri
d
form a
rra
nge
ment, plane
coo
r
din
a
tes
b
y
starting at the ori
g
in a
r
e
cal
c
ulate
d
, so it
just nee
d to
record point
elevation a
r
ray, gener
al
perfo
rman
ce
for the vertical value of two-
dimen
s
ion
a
l coordi
nate a
r
rays,
Today'
s
GIS system
s gen
erally su
ppo
rt the
two methods of data
repre
s
e
n
tation. The
GRID and
th
e algo
rithm
o
f
TIN have
b
een m
a
ture,
now it i
s
n
o
l
onge
r affe
cte
d
virtual
roa
m
ing
function
reali
z
ation
of the
key facto
r
s o
f
urba
n
lan
d
scap
e. Creato
r can b
e
abl
e to identify the
DEM form
at for USGS
DEM and oth
e
r
spe
c
ific fo
rmat, so
wit
h
all ki
nd
s o
f
GIS, RS tools
inclu
d
ing A
r
cGIS, MapGis,
Erda
s an
d E
N
VI softwar
e
to gene
r
ate
DEM, the fin
a
l format
sho
u
ld
be converte
d
into USGS
DEMO
Creat
or, its inte
rnal
format ge
nerated by US
G
S
DEM and
then
by using the
polygon alg
o
rithm and the level of
detail (LO
D
) 3
D terrain can be g
e
nerated.
In this syste
m
, the whol
e regi
on are
a
is not
larg
e, it demand
s
terrain d
e
tails, final
sele
ction by
way of irregu
lar trian
g
le n
e
t (T
IN) to g
enerate DEM
.
Rende
rin
g
s are a
s
follo
ws:
Figure 1 scen
ic terrain DE
M image
Figure 1. Sc
enic
Terrain DEM Image
4.1. Establis
hment of Vir
t
ual Env
i
ronment Model
4.1.1. Basic
Cons
tru
c
tio
n
of Env
i
ronmental Scen
ario
3-D e
n
viron
m
ent scene bui
lding is the foundatio
n of the entire mod
e
ling pro
c
e
s
s. On the
basi
s
of p
a
rtit
ion to compl
e
te each pa
rtition of
the
spe
c
ific e
n
tity modelin
g an
d g
eometry m
o
d
e
l
of the la
nd
scape, bi
g e
n
tities to
the
co
m
p
lex con
s
tru
c
tion, small
to
road
sid
e
flo
w
ers g
r
a
ss. E
a
ch
model
de
scri
bed i
n
the
o
b
ject
sh
ape
i
s
d
e
termi
ned
by p
o
lygon,
vertice
s
and
triangle
s
,
obj
ect
appe
ara
n
ce is determine
d by its surfa
c
e
texture,
color, material, illumination coef
ficient, etc.
Landfo
r
m la
ndsca
pe is
one of the
most im
po
rt
ant geog
ra
p
h
ical o
b
je
cts in the
environ
ment,
the terrain
model is the basi
s
of
virtual geo
grap
hic e
n
vironment of three-
dimen
s
ion
a
l spa
c
e, an
d it is the ca
rri
er of other
feat
ure
s
. Terrain
landform dat
a
modeli
ng is to
use vi
sual a
r
ea, a
c
cording to ce
rtai
n model
i
ng
algorith
m an
d de
scribe
s
the ch
ange
s of
contin
uou
s re
lief area an
d reflects the
re
al pro
c
e
ss.
Whe
n
the
reli
ef is
not bi
g a
nd the
terrain
of
topo
gra
p
h
i
cal fe
ature
s
i
s
n
o
t compli
cated, it
can throug
h the way of field investigati
on and t
he referen
c
e pl
a
ne desi
gn, modelin
g tools,
th
e
basi
s
of usin
g MultiGen p
l
ane co
mbin
a
t
ion inst
ead
of terrain mo
del, namely that the terrai
n
elevation dat
a is ze
ro, the
n
the re
st of the model
i
ng
can b
e
com
p
l
e
ted in the terrain surfa
c
e.
As for
more
complex terrai
n, Terrain to
o
l
s is
provided
by MultiGen
module, th
e
module
is a quickly creation and la
rge
topog
rap
h
y
database tool, through u
s
ing the unifie
d manag
eme
n
t
of proje
c
t re
source
s (te
rrai
n
data,
texture, cultural ch
ara
c
teri
stic
s, etc.). It can m
a
ke top
ograp
hic
accuracy
clo
s
e to the rea
l
worl
d, with
a high
fidelit
y in three-di
mensi
onal
cu
lture feature
and
texture feature. It uses
a
seri
es
of tria
ngulat
io
n alg
o
rithm a
nd t
he mod
e
l of
the earth,
a
n
d
automatically build
s an
d transfo
rm
s the
terrai
n
, while
it is
con
s
i
s
tent with th
e
prototype
of the
geog
rap
h
ic l
o
catio
n
. Through the te
xture map,
photog
rap
h
s of the terra
i
n are g
ene
rated,
inclu
d
ing
ro
a
d
s,
rivers, cit
y
or oth
e
r, th
e charac
te
ri
stics of the
are
a
. Its path
di
scovery
algo
rit
h
m
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 6, June 20
14: 4353 – 4
360
4356
is
sup
e
rio
r
t
o
line
a
r feat
ure
ge
neration al
gorith
m
. Terrai
n m
o
deling
is a
very compli
cated
pro
c
e
ss; ge
n
e
rally sp
ea
kin
g
, the analysi
s
of the
data prep
aration, the orig
i
nal m
odel gen
eration,
and topo
gra
p
h
y optimizatio
n are the
key step
s.
4.1.2. Ph
y
s
io
gnomy
Building
4.1.2.1. Scenic Area Cons
truc
tion La
y
out
The e
n
tire
scenic area m
a
inly con
s
i
s
ts
of se
ve
ral
group
s of b
u
ildi
ngs an
d la
ke
s, in th
e
pro
c
e
ss of
modelin
g, the coo
r
din
a
te
s and shap
e
should b
e
relatively accurate. After man
y
experim
ents,
whe
n
we g
e
t CAD d
r
a
w
ing
of the sce
nic spot, CAD d
r
awi
ng is i
n
DXF format,
we
can im
port it
into the Cre
a
tor, the ba
sic po
sition i
s
accurately. The un
ne
ce
ssary p
a
rt
s can
omitted, keep
the sce
nic a
r
ea peri
phe
ry and entity
sh
ape, location
informatio
n, take the
CAD
as
a reprodu
ctio
n, and dra
w
terrain
co
ntou
r line, and the
n
the contou
r line alone ca
n be saved a
s
a
DXF format a
nd PLT forma
t, the outline
of the
sceni
c
area
small rul
e
s shap
e and
basi
c
physi
cal
locatio
n
information ca
n a
c
hieve the p
r
edetermine
d requi
rem
ent.
4.1.2.2. Path Modeling
On the
surfa
c
e of the
ro
ad
feature
mod
e
ling,
the
key
parts a
r
e th
e
every path
s
o
f
each
positio
n an
d road info
rmati
on. Plane
roa
d
inform
ation
is ea
sy to ex
pre
ss,
while
surfa
c
e
reli
ef or
ways of ro
ad
information i
s
more difficult to i
ndicate
, if we mark the roa
d
locat
i
on in the CAD
conve
r
si
on of
creato
r
expressed
go
od t
e
rrain file, thi
s
lo
catio
n
p
r
oblem
can
b
e
ea
sy
solve
d
.
After the posi
t
ioning of the featur
e mo
del
ing, the cre
a
tor offers an o
pen mod
u
le o
f
road tools.
The mo
dule
has
po
werful
function
s, it def
ine
s
the
transve
rse
sl
ope a
nd lon
g
itudinal
slop
e of the
road, yo
u ca
n define th
e
road
cro
ss section of
tran
sverse slop
e
and
lo
ngitud
i
nal
slop
e, turnin
g
radiu
s
, the g
eometry
stru
cture
of the side of the ro
a
d
, and even
can
be di
rect
ly
transfe
rred e
x
ternal refe
re
nce to defin
e
the way stre
et light, reflector, traffic
s
i
gns
,
s
u
c
h
as
t
he
road g
eom
etry model, it also can be di
rectly app
lie
d road texture, road g
r
e
enin
g
belt pro
c
e
s
sing
and m
u
ltiple
LODs. As to road g
r
e
ening
belt, we
ca
n
adopt the
me
thod on
the
road to
esta
bli
s
h
the sub
part.
Acco
rdi
ng to the need
s of system simu
lati
on, in the pape
r, we ch
oose the pol
ymesh
algorith
m
an
d
stan
da
rd sampling rate and set
the
numbe
r of b
enchma
r
ki
ng,
throu
gh
syst
em
cal
c
ulatio
n, the gene
rated t
e
rrain is mo
re
reasona
bl
e. If the driven sampling
rate i
s
too larg
e, the
grid effect is
not so
good;
when
the
sampling
rate i
s
too
small,
area block si
ze XY will
be too
large, the g
e
nerate
d
polyg
on is ove
r
mu
ch, and it
will
cau
s
e
slow t
e
rrain converting spe
ed, a
nd
affect the
re
al-time
sp
ee
d of the
sce
ne. The
r
ef
o
r
e, it nee
ds to weigh
the
variou
s fa
cto
r
s
of
polymesh alg
o
rithm pa
ram
e
ters, a
nd set each
param
eter re
asona
bly. Area block si
ze
sho
u
ld
be
cho
s
e
n
as far as po
ssi
ble b
i
g, and then the app
rop
r
iat
e
sampli
ng ra
te is necessa
ry.
4.1.2.3.Textu
r
e Mapping
The mo
st
commonly u
s
ed in the
texture a
r
e t
he ae
rial
ph
otogra
p
h
s
, p
a
ste the
corre
s
p
ondin
g
aeri
a
l ph
oto
g
rap
h
s i
n
the
corre
s
p
ondi
n
g
regi
on eve
n
make th
e visual effect of t
he
terrai
n
model
achi
eve gen
u
i
ne, the relate
d figure
s
are as Figu
re 2 to
Figure 7.
Figure 2. Sce
n
ic La
nd
scap
e Ren
deri
n
g
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Const
r
u
c
tion and Appli
c
a
t
ion of Virtual Rea
lit
y Geo
graphi
c Inform
ation System
(Ke Xi-Lin)
4357
Figure 3. He
misph
e
re Sky Rend
erin
g
Figure 4. Vision Effect
Figure 5. Impendin
g
Roa
d
Ren
deri
n
g
Figure
6. Vertical an
d Ho
rizontal Ro
ad Slope
Ren
deri
n
g
s
Figure 7. Surface
Ren
deri
n
g
4.1.2.4. Riv
e
r Modeling
The enti
r
e
scenic
are
a
ha
s three p
ond
s;
the ar
e
a
in t
he entire sce
n
ic a
r
ea i
s
la
rger.
On
the su
rface
modelin
g, it can
use th
e
floor pla
n
det
ermin
e
its po
sition, si
ze, shape a
n
d ot
her
factors, and then it also ta
ke
s pi
ctu
r
e
s
of water ri
ppl
e, vegetat
ion image
s even
ban
k su
rface as
the texture i
m
age
s. Ensu
re the
a
c
tual
effect is th
e
key to the
la
ke
rippl
e
s
of
image
pro
c
e
s
sing
and texture repeat joinin
g together, the
small wave
texture image repeate
d
in large are
a
, until it
covers the
whole a
r
ea. As long a
s
the
edge
of t
he g
r
aphi
cs is
pro
c
e
s
sed
carefully, and ma
ke it
repeatability, feeling
of
abrupt changes i
n
texture j
o
ined together won'
t be appeared, thus make
the whol
e are
a
look mo
re
smooth and n
a
t
ure.
Figure 8. Exhibition Hall Effect
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 6, June 20
14: 4353 – 4
360
4358
4.2. En
v
i
ron
ment Land
s
cape Cons
tr
uction
Land
scap
e m
a
inly incl
ude
s the grass, trees, fl
o
w
e
r
s,
street l
a
mp
s, hedg
es, flo
w
er b
eds,
street si
gn
s, fences, etc. De
co
rated
with
landscap
e
, realism a
n
d fidelity can gre
a
tly enhance the
scene.
But the tree
s and
flowers everywhere,
su
ch
as disp
er
se
d,
the numb
er is mo
re, and
mostly with irregula
r
sh
ape
s, irre
gula
r
ob
je
ct modeli
n
g
has be
en the
rese
arch h
o
tspot.
If acco
rdin
g to the re
al form of scene
modelin
g in d
e
tail, the polygon n
u
mbe
r
will be i
n
the hundreds of millions, it
will not only greatly
exceed the modeling tool
capacity
limits,
but
also
will lead to co
llapse eventu
a
lly ov
erwh
el
med by roami
ng syste
m
.
This sy
stem
mostly adopt
s simpl
e
geo
metry to
express its app
ea
ran
c
e, then u
s
e texture
mappin
g
to ensu
r
e the aut
henticity of scenery. In t
he belo
w
, the grass, tree
s, flowers an
d sh
rubs
are taken a
s
example
s
in the discu
s
sion
of the above method.
In this sy
ste
m
, the gra
s
s
coverage
are
a
is la
r
ge; it i
s
al
so the i
m
portant
conte
n
t in the
scene.
Rel
a
tive to oth
e
r scen
ario
s, th
e
grass
ca
n
b
e
blu
r
red. T
h
ere
is a
direct partial
map
p
ing
method; it is to set up the grass
of the irregula
r
su
rfac
e deformatio
n
,
acco
rdin
g to certain de
gre
e
of repetition
with grass te
xture imag
es.
But this
met
hod i
s
the la
ck of ste
r
eo
sense; truth sense
will be poor after ca
reful observation.
In the syste
m
, it mainly
use
s
an
other met
hod to solve the pro
b
l
em, due to the si
ze of
the gra
s
s is
different, a variety of sh
a
pes
and lo
cat
i
on di
stributio
n is different, and
so the e
a
ch
block
sh
ould
be m
odel
ed.
And the
n
it
can
be
divid
e
d into
surro
undin
g
ed
ge
and th
e
cent
ral
area, a
certai
n height (0.1
m~0.2 m
)
is chosen to
en
sure ste
r
e
o
se
nse of the rea
l
grass.
As
for
detail parts su
ch as
door, windo
w,
ra
ilin
gs, it gene
rally ado
pt the texture
mappin
g
method, nam
ely that on the co
rre
sp
ondi
ng po
sition of
the polygon, it stick on th
e
corre
s
po
ndin
g
texture imag
es, in
stead
o
f
the detaile
d
model.
T
he
advantag
e is that the n
u
m
ber
of poly
gon
s
can g
r
eatly redu
ced an
d model compl
e
xity is al
so redu
ce
d, thus it
can improve the displ
a
y
spe
ed of the
output of the image. Belo
w Figure 9
an
d Figure 10
are the
cha
r
ts of sceni
c a
r
ea
exhibition hall
and the final rend
eri
ng.
Figure 9. Exhibition Hall Chart
(a)
(b)
Figure 10. Exhibition Hall Effect
4.3. Model Optimization
In virtual reali
t
y sce
ne, the
model i
s
ve
ry
com
p
lex; a l
a
rge
num
be
r
of polygon
s t
a
ke
s u
p
a lot of memory spa
c
e a
nd red
u
ce the spe
ed
syst
em of real-ti
m
e intera
ctio
n. Although the
hard
w
a
r
e of
virtual reality
is greatly imp
r
oved n
o
w, b
u
t in the pro
c
ess of the hu
ge amo
unts
of
data, the hardware
also shows po
werl
ess
capability. Aiming at
th
is probl
em, this paper i
s
to
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Const
r
u
c
tion and Appli
c
a
t
ion of Virtual Rea
lit
y Geo
g
raphi
c Inform
ation System
(Ke Xi-Lin)
4359
redu
ce
the
a
m
ount of
dat
a proc
essin
g
method
in t
he virtual
scene m
odel
a
nd o
p
timize
the
model, in ord
e
r to improve
the rend
eri
n
g
s
pee
d, redu
ce the lag time when roami
n
g.
4.3.1. Adjustment of Str
u
cture
Before th
e
scene m
odel
is
set u
p
,
all e
n
tities in
the virt
ual
scene
mo
del hie
r
a
r
chy
sho
u
ld
be determi
ne
d according to the geomet
ry of eac
h en
tity in
the virtual scene
sp
ace po
sition,
and
the structu
r
e
of the intern
al relatio
n
s
b
e
twee
n mod
e
ls a
nd mo
d
e
l.
After the scene
hierarchy
division, it
can facilitate the solid
model
of organization and
management,
and
definite ta
rget
in
model buil
d
in
g, thus greatl
y
lightens the
load of
the model. Even the simple
st
model al
so n
eed
s
to adjust the
model of hie
r
archical struct
ure and to a
c
hieve the goa
l of optimization.
Adjusting p
r
in
ciple of the hi
era
r
chy view is as follo
ws:
(1) An
obje
c
t may be form
ed by multipl
e
form
s, it does n
o
t need t
o
put every b
ody in its
own
set
of no
des, but
th
e obje
c
t
of a ce
rtain scal
e
of
the bo
dy
sho
u
ld b
e
put to
gether,
un
der a
comm
on
set
of node
s. As to a file with
a hierarch
i
c
al stru
ctu
r
e,
executio
n efficien
cy is
hig
her
than files with
no hi
era
r
chy. Ren
d
e
r
the
b
ody doe
s
not
belon
g to top
level set
of n
ode
s, and
the
n
the top unde
r the set of nodes of all forms can be ig
nore
d
.
(2) Visual adj
ace
n
t form should al
so b
e
adja
c
ent in
the hiera
r
ch
y. The same
level of
data n
ode
s in
turn f
r
om l
e
ft to rig
h
t, wh
e
n
no
de
s ex
ist
and th
e o
b
je
ct is not visi
bl
e; it ca
n di
spl
a
y
the node p
o
si
tion adju
s
tme
n
t.
(3) T
r
y to avo
i
d creating
sp
an larger
obj
ects
i
n
sp
ace
.
Although on
ly fall within our field
of vision can
dra
w
the pa
rt
of the object,
alt
hough
we
only can
se
e part of the ob
jects, while th
e
system
shoul
d cal
c
ulate th
e obje
c
t of the big sp
atial span.
(4) T
he ulti
mate goal
o
f
model is t
o
rea
c
h
wh
a
t
extent definition and
u
s
e
what
techn
o
logy. Targ
et imple
m
entation m
odel sy
stem
of restri
ctio
ns; such as colo
r, polyg
ons,
material, light
and texture,
the backg
ro
und of the
m
odel sy
stem
requi
rem
ent i
s
as
simpl
e
as
possibl
e, as real as p
o
ssibl
e
; spe
c
ial
re
q
u
irem
ents in t
he wh
ole mo
del system.
4.3.2. Segmenta
tion Mod
e
l Unit
Cell
division
is to
divide t
he virtual
scene m
odel
in
to sm
aller un
its, only g
e
n
e
rate
s a
transitio
n
zon
e
between
two layers of
L
O
D
.
Befo
re h
i
gh p
r
e
c
isio
n
in LO
D mo
de
l gene
rate
d i
n
the tran
sition
zo
ne, a
nd
high
pre
c
i
s
io
n of L
O
D in
every ve
rte
x
corre
s
p
ond
ing mo
rp
hing
, its
scope
is lo
ca
ted in th
e a
d
j
a
ce
nt verti
c
e
s
b
e
tw
e
en t
w
o LO
D
mod
e
l
s
, they a
r
e
g
enerally sele
cted
adja
c
ent vert
ex of the low pre
c
i
s
ion L
O
D mo
del
v
e
rti
c
e
s
for it
re
c
ently
morp
hin
g
v
e
rtic
es,
thus
formed
the
morp
hing t
w
o ed
ge
s of
the verti
c
e
s
in
the p
r
oce
s
s of transitio
n. Before
transfo
rmatio
n of LO
D m
odel, the tra
n
sition
of
re
al-time
simul
a
tion sce
nari
o
will di
spl
a
y the
transitio
n val
ue of th
e vertex and
upda
te the tra
n
siti
on g
r
ad
ually, until the
nex
t LOD mod
e
l
is
displ
a
yed.
(1) Sha
pe Re
tention
As to
simplif
y the algo
rith
m mu
st preserve
the
sha
pe an
d
su
rfa
c
e
ch
ara
c
te
ristics of
model
as fa
r
as p
o
ssibl
e
, so the al
gorith
m
must
fi
rst fi
nd out th
e fe
ature i
n
form
a
t
ion mod
e
l (e.g.,
surfa
c
e
cu
rva
t
ure, sha
r
p p
o
int and feat
ure ed
ge
),
an
d then throu
g
h
the fusion
of flat area and
cha
nge ch
aracteri
stics of
linear
edg
e t
o
sim
p
lify the model. T
oda
y, most algo
rithms a
dopt t
he
edge
coll
ap
se is to
sim
p
li
fy the model
or
smal
l
e
r
curvature of t
he adj
acent surfa
c
e, and also
throug
h the thre
shol
d met
hod to simplif
y the control.
(2) App
r
ox
im
ation Erro
r Es
timates
In orde
r to si
mplify control
pro
c
e
ss, e
a
ch st
ep i
s
to si
mplify all evaluation mo
del
of local
approximatio
n erro
r, such
as
som
e
e
r
ro
r e
s
timati
on
algorith
m
ad
o
p
ts lo
cal
or di
stan
ce
crite
r
i
o
n
to judge the simplified erro
r; geometry an
d some
al
go
ri
thms are use
d
to limit the
simplified.
5. Conclusio
n
In the study, it makes the study and discussi
on of virtual reality ge
ogra
phi
c information
system te
ch
n
o
logy sy
stem
atically ba
se
d
on th
e
data
colle
ction
an
d analy
s
is,
a
nd throug
h u
s
ing
real time
3
-
D modeli
ng to
ol Creato
r
, visual
drive
r
software
com
b
ined with G
I
S
softwa
r
e, and
throug
h usi
n
g
VC langua
ge
, it designs a
nd implem
ent
s a reali
s
tic 3
D scen
e mod
e
l.
In the paper,
many kind
s of modeling
method
to co
nstru
c
t the virtual enviro
n
m
ent of
sceni
c spot
s are
ado
pted, con
s
id
erin
g
t
he sceni
c
sp
ot, entity mod
e
l ba
se
d te
rrain, environm
ent
and othe
r obj
ects of the p
r
oblem of mod
e
l establi
s
hm
ent.
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 6, June 20
14: 4353 – 4
360
4360
It also analy
z
ed co
mmon
probl
em
s in the proc
ess o
f
virtual environment mo
d
e
ling, the
model
of
structural
adju
s
tment, texture ma
pping,
l
e
vels
of det
ail, extern
al
referen
c
e
s
a
n
d
instantiate th
e tech
nical m
e
thod
s are u
s
ed i
n
optimi
z
ing th
e entire model. T
h
rough
usi
ng th
ese
method
s, ove
r
all pe
rform
a
nce
of roami
n
g scena
ri
o
s
can be im
prov
ed sig
n
ifica
n
tly; requiremen
t
s
of the real-tim
e rend
eri
ng can be met.
Referen
ces
[1]
Gao Jian
xin. T
he develo
p
ment
trend of GIS in recent revie
w
.
Jo
urna
l
of surveying
and
ma
ppi
n
g
infor
m
ati
on an
d eng
in
eeri
n
g
. 200
3; (5): 15-1
8
.
[2]
Z
heng Go
ngb
o
,
luo zhij
un. T
he deve
l
o
p
ment
trend of GIS techn
o
lo
g
y
to e
x
p
l
or
e.
Journ
a
l
of souther
n
land and resources.
2003; (8):
67-70.
[3]
Shi J
un, Qiu
Xin C
u
i
Li
nli,
etc. T
he CAD gr
aph
ic
d
a
ta is
c
onverte
d to t
h
e tech
nol
og
y
r
e
searc
h
a
n
d
app
licati
on of
GIS spatial dat
a.
Journa
l of n
o
rtheast forestr
y
univers
ity.
2001; 29(6): 8
3
-8
5.
[4]
Don
g
R
enc
ai,
cong
sh
engr
i
eds, D
e
n
g
Go
ngb
ing,
etc.
T
he ap
plic
ation
of
3
s
tech
no
l
o
g
y
in urb
a
n
gree
n spac
e s
y
stem stud
y
.
F
o
rest resources
ma
na
ge
me
nt.
200
6; 4(2): 83-
85.
[5]
Gong J
un, Z
h
u
Qing S
u
i
Ha
i
Gang, etc.
F
r
o
m
the CA
D mo
del t
o
the
seve
ral pr
ob
lems
of dig
i
tal
urb
a
n
GIS model. Journa
l
of W
uhan
University (e
n
g
in
eeri
ng sci
en
ce).
2003; 3
6
(3
): 64-68.
[6]
He Jing. 3-D m
ode
lin
g techn
o
l
o
g
y
.
Journ
a
l of
comp
uter ap
pl
icatio
ns
. 200
6; 24(2
39): 33-
35
.
[7]
BuLiJ
i
ng W
a
n
g
Guhai Z
h
a
ng Z
hen
gp
en
g. 3DGIS visualiz
atio
n s
y
st
em.
Journa
l of Liao
nin
g
Engi
neer
in
g T
e
chno
logy U
n
iv
ersity (supp
le
ment)
. 2006; (2
5
)
: 63-63.
[8]
Don
g
Z
hank
un
. Venture Mult
iGen mo
del
ing
techno
lo
g
y
in
the rese
arch
and a
p
p
licati
on of visu
a
l
simulati
on.
Jou
r
nal of el
ectron
ic techno
lo
gy
. 200
6; (2): 14-1
7
.
[9]
Chu Z
h
e
ng W
e
i Ch
en
Xi
n, H
an W
enq
ua
n, etc.
T
he cit
y
of
three-dim
ens
i
ona
l ge
ogra
p
h
i
c informatio
n
s
y
stem to esta
blish, ma
intai
n
, update
and a
p
p
licati
o
n
.
Jo
urn
a
l of eng
in
eeri
ng survey.
20
0
6
; (11): 9-13.
[10]
Z
hu Qin
g
. Co
n
s
ider
ation
of
3
D
GIS techn
o
l
o
g
y
progr
ess.
Jo
u
r
na
l
o
f
ge
og
ra
ph
i
c
al
in
fo
rm
a
t
io
n wo
rl
d
.
200
4; 2(3): 6-7
.
[11]
Liu
xi
ao
ya
n,
macros, zha
ng Lv c
ountri
es, et
c.
T
o
mana
ge th
e inf
o
rmatio
n
s
y
stem of num
ber
accord
ing to t
he qu
asi n
o
w
w
i
t
h
virtua
l to
real mutu
al o
p
e
rati
ons research test.
Journal of pai
nted
new
spap
er
. 20
03; (2): 17-21.
[12]
Ni Kai, Z
h
a
n
g
Yao-ting, Ma
Y
ue-pe
ng. S
hortest Path
Anal
ys
is
Base
d on Di
jkstra'
s
Algorithm i
n
Emerge
nc
y R
e
spo
n
se S
y
ste
m
.
T
E
LKOMNIKA Indo
nesi
a
n
Journ
a
l
of El
ectrical E
ngi
ne
erin
g
. 20
14
;
12(5): 34
76-
34
82.
[13]
Yu Ka
nh
ua.
Urba
n Pl
an
nin
g
Su
pp
ort
S
ystem Base
d
o
n
T
h
ree D
i
me
nsio
nal
GIS.
TEL
K
OMNIKA
Indon
esi
an Jou
r
nal of Electric
al Eng
i
ne
eri
n
g
.
2014; 1
2
(5): 3
928-
393
5.
[14]
Hon
g
ren J
i
an
g
.
T
he Shari
ng
T
e
lecommunic
a
ti
on Inform
ati
on R
e
sourc
e
s
Mana
geme
n
t S
y
stem Bas
e
d
on B/S model
and GIS
.
T
E
LKOMNIKA Indon
esia
n Jour
nal of
Electric
al Eng
i
ne
eri
n
g
.
2014; 12(
3)
:
238
4-23
89.
Evaluation Warning : The document was created with Spire.PDF for Python.