Ci a ion: Pé ez, J.J.; Sende os, M.;
Casado, A.; Leon, I. Field Wo k’s
Op imiza ion o he Digi al Cap u e
o La ge Uni e si y Campuses,
Combining Va ious Techniques o
Massi e Poin Cap u e. Buildings
2022,12, 380. h ps://doi.o g/
10.3390/buildings12030380
Academic Edi o : Fahim Ullah
Recei ed: 21 Feb ua y 2022
Accep ed: 16 Ma ch 2022
Published: 18 Ma ch 2022
Publishe ’s No e: MDPI s ays neu al
wi h ega d o ju isdic ional claims in
published maps and ins i u ional a il-
ia ions.
Copy igh : © 2022 by he au ho s.
Licensee MDPI, Basel, Swi ze land.
This a icle is an open access a icle
dis ibu ed unde he e ms and
condi ions o he C ea i e Commons
A ibu ion (CC BY) license (h ps://
c ea i ecommons.o g/licenses/by/
4.0/).
buildings
A icle
Field Wo k’s Op imiza ion o he Digi al Cap u e o La ge
Uni e si y Campuses, Combining Va ious Techniques o
Massi e Poin Cap u e
JoséJa ie Pé ez, Ma ía Sende os, Amaia Casado and Iñigo Leon *
Depa men o A chi ec u e, Uni e si y o he Basque Coun y UPV/EHU, Plaza Oña i 2,
20018 Donos ia-San Sebas ián, Spain; joseja ie [email p o ec ed] (J.J.P.); ma ia.sende [email p o ec ed] (M.S.);
[email p o ec ed] (A.C.)
*Co espondence: [email p o ec ed]; Tel.: +34-943-01-7192
Abs ac :
The aim o he s udy is o ob ain as digi aliza ion o la ge u ban se ings. The da a o
wo uni e si y campuses in wo ci ies in no he n Spain was cap u ed. Challenges we e imposed
by he lockdown si ua ion caused by he COVID-19 pandemic, which limi ed mobili y and a ec ed
he ield wo k o da a eadings. The idea was o signi ican ly educe ime spen in he ield, using
a numbe o esou ces, and inc easing e iciency as economically as possible. The esea ch design
is based on he Design Science Resea ch (DSR) concep as a me hodological app oach o design he
solu ions gene a ed by means o 3D models. The digi aliza ion o he campuses is based on he
analysis, e olu ion and op imiza ion o LiDAR ALS poin s clouds cap u ed by go e nmen bodies,
which a e open access and ee. Addi ional TLS cap u e echniques we e used o complemen he
clouds, wi h he s udy o suppo o UAV-assis ed au oma ed pho og amme ic echniques. The
esul s show ha wi h poin s clouds o e lapped wi h 360 images, p oduced wi h a combina ion o
esou ces and echniques, i was possible o educe he on-si e wo king ime by mo e han wo hi ds.
Keywo ds: LIDAR; TLS; UAV; poin cloud; 3D modelling
1. In oduc ion
The global en i onmen al si ua ion is c i ical, wi h he deple ion o na u al esou ces,
global wa ming and CO
2
emissions leading o g ea e en i onmen al awa eness [
1
]. The
building sec o alone accoun s o 18.4% o o al an h opogenic g eenhouse gas emis-
sions [
2
]. I is necessa y o ci ies and he buil en i onmen o ul ill hei po en ial o
enhance ene gy e iciency [
3
]. Digi aliza ion, de ined as he de elopmen and deploymen
o digi al echnologies and p ocesses, is conside ed c ucial o he equi ed ans o ma ion
o he cons uc ion indus y o imp o e p oduc i i y acco ding o he epo o Wo ld
Economic Fo um [4].
In his sense, ou esea ch ook wo main lines om an a chi ec u al pe spec i e:
On he one hand, he op imiza ion o he digi al cap u e o a cons uc ed se ing [
5
], and
he use o digi al 3D models o en i onmen al assessmen o u ban se ings [
6
]. In ea ly
Ma ch 2020
, we commenced a esea ch p ojec ha linked bo h a eas o esea ch. We
we e hen aced wi h he c isis caused by he COVID-19 pandemic, which led o he
s a e o na ional con inemen dec eed on 14 Ma ch 2020 in Spain [
7
], as was he case
in many o he coun ies. The i s ask o he p ojec consis ed o he digi al cap u e o
wo la ge uni e si y campuses. This p ocess usually en ails a g ea deal o on-si e wo k.
Gi en he si ua ion o con inemen imposed by COVID-19, i was e y di icul o spend
long pe iods o ime on si e o ake da a eadings. Mass cap u e o poin s in he u ban
se ings was necessa y, o ob ain mul iple da a (coo dina es, dis ances, su ace a eas, angles,
empe a u es o
acades, e c.
), and his had o be pe o med a b eakneck speed wi h he
ewes possible esou ces.
Buildings 2022,12, 380. h ps://doi.o g/10.3390/buildings12030380 h ps://www.mdpi.com/jou nal/buildings
Buildings 2022,12, 380 2 o 32
Se e al op ions had o be s udied o ind a combina ion o esou ces and echniques.
The decision was made o s a he wo k by using Ligh Ampli ica ion by S imula ed
Emission o Radia ion (LiDAR) poin s clouds, cap u ed by manned Ai bo ne Lase Scan-
ning (ALS), ca ied ou by he go e nmen in bo h ci ies (a esou ce ha is ee in many
coun ies). The LiDAR ALS poin s clouds a e an easily accessible and cheap esou ce, bu
hei accu acy and pe o mance need o be complemen ed by o he cap u e echniques. The
esea ch conduc ed wi h he LiDAR ALS clouds, access o which is ins an , open and ee,
including analysis, edi ing and op imiza ion, enabled he addi ional echniques needed
o comple e he inal poin s cloud o each campus o be es ima ed. In his, pa icula ,
case s udy, Te es ial Lase Scanning (TLS) was conside ed o be he as es op ion o
complemen he inal poin s cloud o he u ban a ea, ex ending he s udy o he suppo o
au oma ed pho og amme ic echniques assis ed by Unmanned Ae ial Vehicle (UAV).
Design Science Resea ch (DSR) is a me hodology ha can p o ide solu ions o e-
sea ch h ough he use o h ee-dimensional models. I employs echniques such as case-
s udies, da a collec ion and documen analysis, among o he s, and cen e s on c ea ing and
op imizing a i ac s o imp o e p ocesses and hei ope a i e pe o mance [
8
]. Wi h his
me hodology, esea ch objec i es a e app oached mo e p agma ically han in explana o y
scien i ic in es iga ion [9].
A e he digi al cap u e o he u ban se ing in 3D, he second phase o he p ojec
ocused on he en i onmen al assessmen o he campuses. The ool used was Neighbo -
hood E alua ion o Sus ainable Te i o ies (NEST), a ool based on li e cycle e alua ion
me hodology (ACV) [
10
,
11
]. Al hough some esul s o his assessmen ha e al eady been
published [
12
], his a icle does no ocus on his phase, i only shows he minimum
in o ma ion necessa y o gi e con ex o he esea ch as a whole.
This a icle ocuses on he esul s o he op imiza ion wo k pe o med on he digi al
cap u e o he wo uni e si y campuses up o when he 3D simula ion models a e ob ained.
The conclusions include he esul ha a e p e iously wo king wi h he LiDAR ALS poin s
cloud o he Go e nmen o Na a a, he no mal on-si e eading pe iod wi h TLS, es ima ed
a 52 days o he campus o Pamplona, could be educed o 7 days. The combina ion o
de ices, so wa e and applica ions ha we e used made i possible o educe he scanning
ime wi h o e lapped 360 image cap u ing by mo e han 75%, in compa ison o cus oma y
scanning imes on he ma ke . We ound ha e en wi h such a as cap u e ime we we e
able o ob ain e o s o 1 mm, wi h a s eng h and o e lap ha was accep ed as alid by
he p ocessing so wa e. The e o e, his a icle could be o g ea bene i o he scien i ic
communi y engaged in wo k o his na u e, since i would help hem o be mo e e icien
and make e ec i e use o esou ces.
The a icle is s uc u ed as ollows: The in oduc ion consis s o wo sec ions. Sec ion 1
con ex ualizes he esea ch. Sec ion 2desc ibes he case s udy, including he cu en s a e o
esea ch in o di e en cap u e echniques. Sec ion 3desc ibes he Me hods and Ma e ials.
Sec ion 4desc ibes he esul s o gene a ion o LiDAR poin clouds, modelling and simula-
ion. Sec ion 5p esen s he Discussion, and he a icle ends wi h he inal conclusions.
2. Digi al Su ey o La ge U ban A eas in a Sho Time, S udy Cases
The case s udies ocus on wo uni e si ies in no he n Spain: he campus o he
Uni e si y o he Basque Coun y (UPV-/EHU) in Donos ia-San Sebas ián (DSS) and he
Uni e si y o Na a a (UNAV) in Pamplona (Figu e 1).
The UNAV campus has an a ea o app oxima ely 113 ha, which includes la ge open
g assy a eas, sligh ly wooded a eas and a i e bed lanked by a dense mass o ees. The
buildings co e only 6.8% o he o al a ea. The UPV/EHU uni e si y campus in DSS has a
much smalle a ea (app oxima ely 18 ha) and a much highe building densi y, wi h a much
lowe p opo ion o g een a eas. As p e iously men ioned, he aim o his wo k is, o ca y
ou he ield wo k in he sho es ime possible, wi h he ewes numbe o esou ces.
Buildings 2022,12, 380 3 o 32
Buildings 2022, 12, x FOR PEER REVIEW 3 o 33
(a) (b)
Figu e 1. (a) Ae ial iew o he UNAV uni e si y campus in Pamplona; (b) ae ial iew o he
UPV/EHU uni e si y campus in DSS.
The UNAV campus has an a ea o app oxima ely 113 ha, which includes la ge open
g assy a eas, sligh ly wooded a eas and a i e bed lanked by a dense mass o ees. The
buildings co e only 6.8% o he o al a ea. The UPV/EHU uni e si y campus in DSS has
a much smalle a ea (app oxima ely 18 ha) and a much highe building densi y, wi h a
much lowe p opo ion o g een a eas. As p e iously men ioned, he aim o his wo k is,
o ca y ou he ield wo k in he sho es ime possible, wi h he ewes numbe o e-
sou ces.
The modeling phase in NEST equi es a p io g aphic su ey o he cu en s a e [13].
Depending on he eason o using he model, i will equi e a su icien deg ee o p eci-
sion o accu a ely de e mine he geome ic con igu a ion o buildings and hei su ound-
ings [14]. Al hough he NEST 3D model does no equi e excessi e p ecision [15], he su -
ey does equi e de ini ion o he building en elopes [16], ma king he numbe o loo s,
window con igu a ions and opaque elemen s, and o he spaces occupied by oads and
highways; g een spaces and ees mus also be de ined [17], so ha o es biomass can be
calcula ed [18–20].
Gi en he limi ed ime a ailable o ca ying ou ieldwo k, he use o massi e poin
cap u e echniques will allow highly p ecise geome y o be ob ained in digi al o ma in
a e y sho ime. The combined use o digi al geome ic da a collec ion echniques [21],
is cu en ly he mos e ec i e p ocedu e o conduc ing a p ecise geo- e e enced a chi ec-
u al su ey [22]. In such cases, i includes a opog aphic su ey wi h a o al s a ion [23],
e es ial lase scanne and sho - ange pho og amme y assis ed by an RPA (Remo ely-
Pilo ed Ai c a ) o UAV [24–27].
LiDAR echnology also makes i possible o acqui e massi e amoun s o 3D geospa-
ial in o ma ion in u ban scena ios [28–31]. LiDAR echnology measu es he p ope ies o
e lec ed lase pulses o de e mine he ange o a dis an objec [32]. Tha ange is ob ained
by measu ing he delay ime be ween ansmission o a lase pulse and de ec ion o he
e lec ed signal [33]. Due o LiDAR’s abili y o gene a e 3D da a wi h high p ecision and
spa ial esolu ion, a new e a is opening o he de elopmen o esea ch objec i es such as
he one p esen ed [34]. LiDAR scanning can be classi ied in ou ca ego ies: Sa elli e-based
Lase Scanning (SLS), Ai bo ne Lase Scanning (ALS) [35–38], Mobile Lase Scanning
(MLS) and Te es ial Lase Scanning (TLS) [39]. ALS is ideal o la ge a eas o ci ies
[40,41]; i can be conduc ed by UAVs o by manned ai c a , which usually ly a highe
al i udes and cap u e la ge a eas han UAVs, hough he la e a e cheape and less pol-
lu ing, among o he ad an ages [42–44]. SLS da a poin s can be ens o me e s apa and
he espec i e poin clouds a e he e o e unsui able o ex ac ing geome ies om u ban
ea u es such as buildings o masses o ees [45]. TLS da a has he highes poin densi y
and can be used o speci y da a o hose u ban elemen s a indi idual le el [46–48]. Some
publica ions claim ha TLS some imes has poo mobili y and occlusion issues ha make
Figu e 1.
(
a
) Ae ial iew o he UNAV uni e si y campus in Pamplona; (
b
) ae ial iew o he
UPV/EHU uni e si y campus in DSS.
The modeling phase in NEST equi es a p io g aphic su ey o he cu en s a e [
13
].
Depending on he eason o using he model, i will equi e a su icien deg ee o p ecision
o accu a ely de e mine he geome ic con igu a ion o buildings and hei su ound-
ings [
14
]. Al hough he NEST 3D model does no equi e excessi e p ecision [
15
], he
su ey does equi e de ini ion o he building en elopes [
16
], ma king he numbe o loo s,
window con igu a ions and opaque elemen s, and o he spaces occupied by oads and
highways; g een spaces and ees mus also be de ined [
17
], so ha o es biomass can be
calcula ed [18–20].
Gi en he limi ed ime a ailable o ca ying ou ieldwo k, he use o massi e poin
cap u e echniques will allow highly p ecise geome y o be ob ained in digi al o ma in a
e y sho ime. The combined use o digi al geome ic da a collec ion echniques [
21
], is
cu en ly he mos e ec i e p ocedu e o conduc ing a p ecise geo- e e enced a chi ec u al
su ey [
22
]. In such cases, i includes a opog aphic su ey wi h a o al s a ion [
23
],
e es ial lase scanne and sho - ange pho og amme y assis ed by an RPA (Remo ely-
Pilo ed Ai c a ) o UAV [24–27].
LiDAR echnology also makes i possible o acqui e massi e amoun s o 3D geospa ial
in o ma ion in u ban scena ios [
28
–
31
]. LiDAR echnology measu es he p ope ies o
e lec ed lase pulses o de e mine he ange o a dis an objec [
32
]. Tha ange is ob ained
by measu ing he delay ime be ween ansmission o a lase pulse and de ec ion o he
e lec ed signal [
33
]. Due o LiDAR’s abili y o gene a e 3D da a wi h high p ecision and
spa ial esolu ion, a new e a is opening o he de elopmen o esea ch objec i es such as
he one p esen ed [
34
]. LiDAR scanning can be classi ied in ou ca ego ies: Sa elli e-based
Lase Scanning (SLS), Ai bo ne Lase Scanning (ALS) [
35
–
38
], Mobile Lase Scanning (MLS)
and Te es ial Lase Scanning (TLS) [
39
]. ALS is ideal o la ge a eas o ci ies [
40
,
41
]; i can
be conduc ed by UAVs o by manned ai c a , which usually ly a highe al i udes and
cap u e la ge a eas han UAVs, hough he la e a e cheape and less pollu ing, among
o he ad an ages [
42
–
44
]. SLS da a poin s can be ens o me e s apa and he espec i e
poin clouds a e he e o e unsui able o ex ac ing geome ies om u ban ea u es such as
buildings o masses o ees [
45
]. TLS da a has he highes poin densi y and can be used o
speci y da a o hose u ban elemen s a indi idual le el [
46
–
48
]. Some publica ions claim
ha TLS some imes has poo mobili y and occlusion issues ha make i di icul o collec
da a on an u ban scale. When TLS is no e ec i e, MLS has been used in some esea ch,
such as o collec ion and analysis o in o ma ion on ees in u ban a eas [49,50].
Va ious echniques ha e been s udied, and he me hod ha bes i he objec i es o
his wo k in ol ed he combina ion o di e en esou ces: ALS LiDAR clouds cap u ed by
public adminis a ions, poin clouds cap u ed using TLS and, inally, poin clouds p oduced
om au oma ed pho og amme y assis ed by UAVs.
Buildings 2022,12, 380 4 o 32
3. Me hods and Ma e ials
In his sec ion he echnologies enabling he massi e poin cap u e o engende he
poin cloud o each uni e si y campus in a e y sho ime [
5
], a e explained, speci ying
he echniques, me hods and ma e ials used. The esea ch design is based on DSR. DSR is
used o design and assess manmade a i ac s mean o esol e eal-wo ld p oblems [
51
].
This me hod helps ind p ac ical solu ions o common p oblems a ec ing design, wi h a
iew o achie ing expec ed esul s [
8
], and employs compu e -based ools o s eamline
p ocesses [
51
]. When a p oblem is associa ed o a physical objec , he espec i e solu ion
may appea as a 3D model, plan o d awing; when i equi es op imizing an ac ion,
he solu ion may ake he o m o new digi al so wa e o be de eloped as a lowcha
diag am [
51
]. DSR includes o he non-habi ual o ms o con eying knowledge, such
as models o cons uc s [
52
]. Tha is why DSR exp esses knowledge based on di e en
o ma s no commonly ound in o he scien i ic in es iga ions, such as, o example, 3D
models, a chi ec u es, design heo ies o p inciples and a i ac s [
53
]. Two main ac i i ies
a e pu o wa d in design science: o build he solu ion and o e alua e i [
54
]. The cons uc
is a s age wi hin he p ocess o c ea ing an a i ac ha can be used o esol e a speci ic
p oblem. The e alua ion is he ac ion ha mus alida e how e ec i ely ha a i ac se es
he pu pose o which i was c ea ed. This is p ecisely wha is going o be conduc ed in his
in es iga ion: o make, e alua e and op imize a 3D digi al model o he u ban a ea in he
o m o a poin cloud wi h 360 image ha con ains all he in o ma ion needed o achie e
he p ojec ’s objec i es. The cons uc s age mus necessa ily be i e a i e and inc emen al,
as he e alua ion phase will endow i wi h he eedback needed o op imize he solu ion.
DSR enables ele an p oblems o be esol ed based on applied esea ch appea ing in some
scien i ic in es iga ions linked o a chi ec u e [55].
3.1. Analysis o LiDAR Clouds Ob ained by Public Se ices
In Spain, di e en public se ices o e ha LiDAR da a, he eby simpli ying he
da a cap u e p ocess o hese kinds o p ojec s, wi h he espec i e poin clouds ob ained
using manned ai c a . The g ea ad an age o hese clouds is ha hey a e public and
can be consul ed o ee. In he case o poin clouds ob ained by ALS sys ems, he la es
senso echnology has signi ican ly inc eased he numbe o lase ligh beams pe squa e
me e . As a esul , he densi y o he poin clouds gene a ed du ing he da a collec ion
p ocess shows a ange o be ween 12–30 poin s/m
2
, compa ed o he ange o 1 poin /m
2
ob ained by p e ious senso s. The Cha e ed Communi y o Na a e was one o he i s
Eu opean egions o apply LiDAR echnology using hese new senso s, speci ically he
Leica Single Pho on LiDAR (SPL100). A senso is able o cap u e ligh pa icles wi h a lase
ligh beam ha can be di ided in o a 10
×
10 ma ix, ope a ing in p ac ice as
100 senso s
in pa allel, each o which is cap u ed by an independen channel o he de ec o . The
expe imen al ligh s we e conduc ed in 2017; a e p ocessing and classi ying he da a
ob ained using A i icial In elligence (AI) echniques, i was possible o co e an a ea o
10,391 km
2
, gene a ing meshes o 1
×
1 km, wi h a poin cloud densi y o 14 poin s/m
2
and a p ecision o 20 cm on he XY axis and 15 cm on he Z axis.
The Na a e go e nmen ’s pa ial LiDAR clouds om 2017 we e ini ially used, and a
poin cloud o he en i e UNAV Campus in Pamplona was composed. Di e en cu s o he
cloud we e pe o med a s a egic poin s; he p ecision and sui abili y o he cloud we e
also analyzed o s udy he combina ion o echniques (Figu e 2).
The campus cloud was segmen ed o o m de ailed se s o buildings and check hei
geome y wi h he densi y alue o 14 poin s/m
2
. The de ini ion o ha açade would
appa en ly su ice o ob ain a 3D simula ion model (Figu e 3).
Dimensional checks o he esul will, subsequen ly, be ca ied ou o asce ain which
buildings need o comple e he poin cloud wi h o he LiDAR echniques.
Buildings 2022,12, 380 5 o 32
Buildings 2022, 12, x FOR PEER REVIEW 5 o 33
he cloud we e pe o med a s a egic poin s; he p ecision and sui abili y o he cloud
we e also analyzed o s udy he combina ion o echniques (Figu e 2).
(a)
(b)
Figu e 2. LiDAR 2017 poin cloud, densi y: 14 poin s/m
2
. UNAV campus: (a) 3D colo cloud; (b)
e ical sec ion o he cloud h ough he no h açade o he cen al campus building.
The campus cloud was segmen ed o o m de ailed se s o buildings and check hei
geome y wi h he densi y alue o 14 poin s/m
2
. The de ini ion o ha açade would ap-
pa en ly su ice o ob ain a 3D simula ion model (Figu e 3).
(a)
(b)
Figu e 2.
LiDAR 2017 poin cloud, densi y: 14 poin s/m
2
. UNAV campus: (
a
) 3D colo cloud; (
b
)
e ical sec ion o he cloud h ough he no h açade o he cen al campus building.
Buildings 2022, 12, x FOR PEER REVIEW 5 o 33
he cloud we e pe o med a s a egic poin s; he p ecision and sui abili y o he cloud
we e also analyzed o s udy he combina ion o echniques (Figu e 2).
(a)
(b)
Figu e 2. LiDAR 2017 poin cloud, densi y: 14 poin s/m
2
. UNAV campus: (a) 3D colo cloud; (b)
e ical sec ion o he cloud h ough he no h açade o he cen al campus building.
The campus cloud was segmen ed o o m de ailed se s o buildings and check hei
geome y wi h he densi y alue o 14 poin s/m
2
. The de ini ion o ha açade would ap-
pa en ly su ice o ob ain a 3D simula ion model (Figu e 3).
(a)
(b)
Figu e 3.
2017 LiDAR cloud densi y: 14 poin s/m
2
. UNAV campus: (
a
) 3D de ail o he cen al
campus building; (b) no h açade o he same building.
To achie e he objec i es o his esea ch, o he da a o in e es included he es ima ed
app oxima e olume o he campuses’ o es biomass; de ailed es s o ee masses we e
acco dingly conduc ed. I was he eby possible o e i y ano he undamen al cha ac e is ic
o hese new senso s, used in 2017. They enable cap u e o he e es ial elie , de oid
Buildings 2022,12, 380 6 o 32
o any a i icial and/o na u al elemen o he han he g ound (DTM—Digi al Te ain
Model), and o he ea h’s su ace wi h all buil o na u al bodies on i (DSM—Digi al
Su ace Model). The use o speci ic wa eleng hs enables pene a ion be ween ee masses,
cap u ing he lowe g ound le el, which acili a es he heigh measu emen o hose masses.
In his ega d, an example o he cap u e o plan masses a he UNAV uni e si y campus
in Pamplona is shown below. Cloud cu s we e conduc ed in wooded a eas. In he case o
he densely popula ed ege a ion zone, he scanne ’s abili y o pene a e he ee mass and
eco d he g ound le el is obse ed [
14
,
42
], (Figu e 4). Al hough he ees’ compac ness
makes i di icul o ully eco d he espec i e mass, he in o ma ion cap u ed allows o
app oxima e measu emen s o he heigh and olume o he ege a ion, wi h a p ecision
ha can be es ima ed o he nea es decime e .
Buildings 2022, 12, x FOR PEER REVIEW 6 o 33
Figu e 3. 2017 LiDAR cloud densi y: 14 poin s/m2. UNAV campus: (a) 3D de ail o he cen al cam-
pus building; (b) no h açade o he same building.
Dimensional checks o he esul will, subsequen ly, be ca ied ou o asce ain which
buildings need o comple e he poin cloud wi h o he LiDAR echniques.
To achie e he objec i es o his esea ch, o he da a o in e es included he es i-
ma ed app oxima e olume o he campuses’ o es biomass; de ailed es s o ee masses
we e acco dingly conduc ed. I was he eby possible o e i y ano he undamen al cha -
ac e is ic o hese new senso s, used in 2017. They enable cap u e o he e es ial elie ,
de oid o any a i icial and/o na u al elemen o he han he g ound (DTM—Digi al Te -
ain Model), and o he ea h’s su ace wi h all buil o na u al bodies on i (DSM—Digi al
Su ace Model). The use o speci ic wa eleng hs enables pene a ion be ween ee masses,
cap u ing he lowe g ound le el, which acili a es he heigh measu emen o hose
masses. In his ega d, an example o he cap u e o plan masses a he UNAV uni e si y
campus in Pamplona is shown below. Cloud cu s we e conduc ed in wooded a eas. In he
case o he densely popula ed ege a ion zone, he scanne ’s abili y o pene a e he ee
mass and eco d he g ound le el is obse ed [14,42], (Figu e 4). Al hough he ees’ com-
pac ness makes i di icul o ully eco d he espec i e mass, he in o ma ion cap u ed
allows o app oxima e measu emen s o he heigh and olume o he ege a ion, wi h a
p ecision ha can be es ima ed o he nea es decime e .
(a)
(b)
Figu e 4. UNAV campus, ege a ion s ip example: (a) 2017 LiDAR Cloud Plan, densi y: 14
poin s/m2; (b) p o ile o ee mass and le el o g ound unde ha mass.
A e analyzing he possibili ies o he LiDAR cloud o he UNAV campus in Pam-
plona, wi h a densi y o 14 p/m2, he LiDAR clouds cu en ly a ailable o he e i o y o
Gipuzkoa p o ince, whe e he UPV/EHU campus is loca ed in DSS, a e analyzed. In ha
p o ince, LiDAR clouds cap u ed in 2012 and 2017 a e cu en ly a ailable (Figu e 5).
Figu e 4.
UNAV campus, ege a ion s ip example: (
a
) 2017 LiDAR Cloud Plan, densi y:
14 poin s/m2; (b) p o ile o ee mass and le el o g ound unde ha mass.
A e analyzing he possibili ies o he LiDAR cloud o he UNAV campus in Pam-
plona, wi h a densi y o 14 p/m
2
, he LiDAR clouds cu en ly a ailable o he e i o y o
Gipuzkoa p o ince, whe e he UPV/EHU campus is loca ed in DSS, a e analyzed. In ha
p o ince, LiDAR clouds cap u ed in 2012 and 2017 a e cu en ly a ailable (Figu e 5).
The 2012 LiDAR ligh p esen s meshes o 2
×
2 km wi h a densi y o 1 poin /m
2
,
while he 2017 LiDAR ligh p esen s 500
×
500 m meshes wi h a densi y o 2.2 poin s/m
2
.
As in he UNAV’s case, a pa ial cloud was c ea ed o he en i e UPV/EHU campus, o
bo h he 2012 and 2017 clouds. Pa ial sec ions o he campus buildings and ees we e
likewise made o compa e he accu acy and use ulness o he clouds.
Al hough speci ic measu emen s o hese LiDAR clouds will be p esen ed in he
Sec ion 4wi h he analysis o hese wo examples om he UPV/EHU campus, se e al
limi a ions can be app ecia ed. In he 2012 cloud, o al heigh o buildings could be ob ained;
howe e , he olumes o buildings a e no in ui ed, no a e e ical s ipes ma ked. In
addi ion, he p o ile o ee masses p esen s excessi ely isola ed poin s, and i canno be
de e mined whe he hey a e masses o speci ic ees. In he 2017 cloud o he same campus,
building heigh s a e co ec ly app ecia ed, olumes a e ma ked wi h e ical s ipes and
he e is g ea e de ini ion o ee masses. E en wi h his de ini ion o 2.2 poin s/m
2
,
building açades could no be modeled no could biomass olumes be calcula ed, unlike
wha was seen in he cloud o he UNAV campus.
Buildings 2022,12, 380 7 o 32
Buildings 2022, 12, x FOR PEER REVIEW 7 o 33
Figu e 5. 2012 ligh LiDAR poin cloud o he UPV/EHU uni e si y campus in DSS. Cloud den-
si y: 1 poin /m2.
The 2012 LiDAR ligh p esen s meshes o 2 × 2 km wi h a densi y o 1 poin /m2, while
he 2017 LiDAR ligh p esen s 500 × 500 m meshes wi h a densi y o 2.2 poin s/m2. As in
he UNAV’s case, a pa ial cloud was c ea ed o he en i e UPV/EHU campus, o bo h
he 2012 and 2017 clouds. Pa ial sec ions o he campus buildings and ees we e likewise
made o compa e he accu acy and use ulness o he clouds.
Al hough speci ic measu emen s o hese LiDAR clouds will be p esen ed in he e-
sul s sec ion wi h he analysis o hese wo examples om he UPV/EHU campus, se e al
limi a ions can be app ecia ed. In he 2012 cloud, o al heigh o buildings could be ob-
ained; howe e , he olumes o buildings a e no in ui ed, no a e e ical s ipes ma ked.
In addi ion, he p o ile o ee masses p esen s excessi ely isola ed poin s, and i canno
be de e mined whe he hey a e masses o speci ic ees. In he 2017 cloud o he same
campus, building heigh s a e co ec ly app ecia ed, olumes a e ma ked wi h e ical
s ipes and he e is g ea e de ini ion o ee masses. E en wi h his de ini ion o 2.2
poin s/m2, building açades could no be modeled no could biomass olumes be calcu-
la ed, unlike wha was seen in he cloud o he UNAV campus.
The p ecision o hese clouds will condi ion subsequen da a collec ion a he wo
campuses o comple e he poin cloud ha allows he 3D simula ion model o be achie ed.
3.2. Da a Collec ion o Comple e he LiDAR, MLS and TLS Clouds
TLS and MLS echnologies make i possible o ob ain highly accu a e poin clouds.
Howe e , managing hose echnologies o cap u e u ban en i onmen s o a ce ain size
equi es in-dep h s udy o ensu e ha hey can be e ec i e, sus ainable and ela i ely
cheap. MLS scanne s a e gene ally much mo e expensi e han TLS scanne s. Howe e ,
depending on he u ban en i onmen , hey can educe execu ion imes and he e o e be a
mo e e icien op ion. In any case, he scanning o such a eas mus be planned e y well
so ha he poin clouds a e no excessi ely dense and can be handled by s anda d ha d-
wa e.
Rega ding he MLS LiDAR op ions, some wo k has op ed o models, such as he
Leica Pegasus, which allows eali y o be cap u ed om a ehicle, ain o ship. I is an
expensi e op ion ha equi es e y speci ic cap u e condi ions o la ge a eas in ci ies,
hough i is e y use ul o cap u ing linea in as uc u es. Was no conside ed due o he
cha ac e is ics o he wo campuses. One op ion ha was es ed is he Leica BLK2GO
handheld scanne , which cap u es mo ing images and poin clouds in eal ime, using
SLAM (Simul aneous Localiza ion and Mapping) echnology o eco d hei cou se
h ough space [56,57]. Tha scanne combines dual-axis LiDAR, a 4.3 Mpx 360° pano amic
Figu e 5.
2012 ligh LiDAR poin cloud o he UPV/EHU uni e si y campus in DSS. Cloud densi y:
1 poin /m2.
The p ecision o hese clouds will condi ion subsequen da a collec ion a he wo
campuses o comple e he poin cloud ha allows he 3D simula ion model o be achie ed.
3.2. Da a Collec ion o Comple e he LiDAR, MLS and TLS Clouds
TLS and MLS echnologies make i possible o ob ain highly accu a e poin clouds.
Howe e , managing hose echnologies o cap u e u ban en i onmen s o a ce ain size
equi es in-dep h s udy o ensu e ha hey can be e ec i e, sus ainable and ela i ely
cheap. MLS scanne s a e gene ally much mo e expensi e han TLS scanne s. Howe e ,
depending on he u ban en i onmen , hey can educe execu ion imes and he e o e be a
mo e e icien op ion. In any case, he scanning o such a eas mus be planned e y well so
ha he poin clouds a e no excessi ely dense and can be handled by s anda d ha dwa e.
Rega ding he MLS LiDAR op ions, some wo k has op ed o models, such as he
Leica Pegasus, which allows eali y o be cap u ed om a ehicle, ain o ship. I is an
expensi e op ion ha equi es e y speci ic cap u e condi ions o la ge a eas in ci ies,
hough i is e y use ul o cap u ing linea in as uc u es. Was no conside ed due o
he cha ac e is ics o he wo campuses. One op ion ha was es ed is he Leica BLK2GO
handheld scanne , which cap u es mo ing images and poin clouds in eal ime, using
SLAM (Simul aneous Localiza ion and Mapping) echnology o eco d hei cou se h ough
space [
56
,
57
]. Tha scanne combines dual-axis LiDAR, a 4.3 Mpx 360
◦
pano amic iewing
sys em, a 12 Mpx high- esolu ion came a o de ailed pho os and an ine ial measu emen
uni ha enables sel -na iga ion, cap u ing 420,000 p s/s wi h a cap u e ange o 0–25 m.
Sys em pe o mance based on SLAM echnology o e s 6–15 mm ela i e accu acy and
20 mm absolu e posi ioning accu acy o maximum ange [
5
]. Conside ing he size and
cha ac e is ics o he campuses, he scanning p ocess wi h his de ice was uled ou bo h
due o execu ion imes and he excessi e amoun o in o ma ion ha would be cap u ed.
Bea ing in mind he accu acy o he LiDAR clouds discussed in he p e ious poin ,
TLS was deemed he as es , mos e icien and leas p oblema ic op ion o comple e he
clouds, al hough i is ue ha speci ic loca ions can be complemen ed wi h cap u es
made using UAVs. The geome ic da a cap u e echnique using e es ial lase scanning
allows his cap u e o be pe o med quickly and expedi iously, cap u ing a la ge amoun o
in o ma ion a e y high speed, om medium and long dis ances and wi h a high deg ee o
accu acy. The gene a ed high-densi y poin cloud can be supplemen ed by 360
◦
pano amic
pho og aphy aken a each scan posi ion. The poin cloud wi h he o e lapping 360 image,
makes i possible o con igu e a h ee-dimensional isual en i onmen whe ein i is easible
o make millime ic measu emen s and de elop i ual isi s. Some scanne s also ha e
Buildings 2022,12, 380 8 o 32
a buil -in he mog aphic came a ha can disce n he empe a u e o each o he millions
o poin s cap u ed when scanning. The empe a u e o he acades is o special in e es
in his ype o p ojec in which ene gy imp o emen is p oposed by means o passi e
solu ions such as ene gy-minded e o m o building açades. The i e me hodological
s ages ollowed when scanning he wo campuses wi h TLS will be desc ibed nex .
3.2.1. Su ey o Con ol Poin s in UTM Coo dina es Using a To al S a ion
The main objec i e o he con ol poin sys em is o ob ain a h ee-dimensional dig-
i al model o he geo- e e enced su ey in absolu e UTM coo dina es. Ob aining a geo-
e e enced model is no an essen ial equi emen in cases whe e he da a collec ion p o-
cedu e is accomplished by lase scanning, since a local coo dina e sys em can be used.
Howe e , his in o ma ion’s implemen a ion in he pho og amme ic p ocessing ensu es
g ea e accu acy o he h ee-dimensional digi al model. Mo eo e , he con ol poin s
gua an ee he igo and accu acy o he da a p ocessing and acili a es he union be ween
di e en cap u es. The ma e ializa ion o hose poin s is pe o med using adhesi e a ge s
o ixa ion on he di e en suppo s, in he o m o igid pla es o a iable size. The con ol
poin s, loca ed on he g ound, a e pe manen ly e e enced by opog aphic nails o hei
main enance du ing execu ion o he wo k.
The checkpoin sys em layou ollows he ollowing c i e ia:
•
Link checkpoin s, o joining poin clouds co esponding o he di e en lase scans.
Loca ed on e ical walls o he açade and mean o co e he maximum possible
wid h, bo h e ically and ho izon ally. Some o hese con ol poin s, i loca ed on
e ical oo aces, can help acili a e he union be ween he da a cap u ed by lase
scanne and he UAV-assis ed pho og aphic cap u e;
•
Checkpoin s on he oo , he opog aphical a ge s ha we usually place o gi e mo e
p ecision o he pho og amme ic wo k o he UAV in ela ion o he wo k o he lase
scanne . We placed hem a he ends o he oo s a di e en heigh s (especially on
he campus buildings ha had la oo s wi h se e al olumes o di e en heigh s).
Tha way some a ge s a e cap u ed by he 3D lase scanne and by he UAV, and his
acili a es he union be ween poin s clouds;
•
Checkpoin s on he g ound, common o bo h da a collec ion p ocedu es o la e
in eg a ion. Si ua ed in such a way ha hey a e eco ded by bo h lase scanning and
UAV-assis ed pho og aphic cap u e.
3.2.2. Scanning Plan
Fo he scanning p ocess o be e ec i e, i is ecommended ha a p io s udy be
conduc ed o he scan posi ions o he se o be cap u ed. Wi h espec o uni e si y
campuses, he cap u e ocused on wo impo an aspec s:
•
The ex e io su ey o each campus’s buildings, bo h in cloud o ma and in a
360◦image
, o ob ain a mul i ude o da a so ha he 3D simula ion model could
be c ea ed om he o ice wi hou ha ing o en u e in o he ield;
•
Regis a ion o he buildings’ ex e nal en i onmen , whe e he main aim ocused on
cap u ing he g een a eas wi h mo e o less ees, o eco d and measu e he amoun
o a ailable o es Biomass.
Because he si ua ion gene a ed by he COVID-19 heal h c isis mean ha mo emen
was e y limi ed, op imizing he ieldwo k was i ally impo an , i mean o educe he
scanning p ocess o he sho es ime possible. To be e icien , i was essen ial o s udy he
campuses’ ca og aphy be o e p oposing a scanning plan. The UNAV campus in Pamplona,
has e ain wi h la ge slopes in some a eas and ha his needs o be aken in o accoun
when p epa ing he scan posi ion plan. I a plan is d awn up wi hou aking he slopes
in o accoun , as i he e ain we e la , he dis ances be ween scan posi ions a e ho izon al
p ojec ions o he ac ual dis ance. The esea ch ocus in his a icle is on educing he
numbe o scan poin s as much as possible, he e o e, he need o ake he slopes in o
accoun in he scan plan is an impo an one.
Buildings 2022,12, 380 9 o 32
As men ioned abo e, he wo campuses ha e e y di e en cha ac e is ics ha a ec
he scanning plan. The Donos ia campus is a ela i ely la u ban campus wi hou la ge
a eas o ees; i is he e o e p ac ically possible o a ange i s scanning plan using an
o hopho o. The wo k was di ided in o h ee di e en a eas ha co e he en i e UPV/EHU
campus. Howe e , he Pamplona campus is o e whelmingly complex. The ex en o
he campus and he la ge numbe o buildings a e a challenge ha canno be p ac ically
co e ed in a sho ime by a e es ial lase scanne . In many cases, he e ain’s une enness
exceeds he heigh o buildings, which a e loca ed a e y di e en heigh s and e y a
apa . Fu he mo e, he medium-heigh ege a ion and abo e all he la ge ees in many
cases p e en he cap u e o many building açades’ geome y. I i we e no o he high
quali y o he 2017 LiDAR clouds, his would be an o e whelming ask in a sho pe iod o
ime, e en using a scanne as e sa ile as he RTC 360. On he UNAV campus, simpli ied
CAD planime y was used o conduc a p io s udy o possible scan posi ions. Conside ing
he high numbe o scan posi ions and he la ge a ea co e ed by he cloud ake, i was
decided o di ide he wo k in o se en zones. Howe e , his depends a g ea deal on he
powe o he compu e ha will be used when p ocessing he clouds.
3.2.3. Lase Scanning, P e-P ocessed in he Field wi h Mobile De ices
The building açades on he di e en campuses we e scanned, as well as he ex e io
en i onmen s, wi h special a en ion gi en o g een a eas and ege a ion. In he su ey o
he buildings he main objec i e was o measu e he açades’ dimensions, di e en ia ing he
sizes o window openings and opaque su aces. A minimum o h ee scans we e conduc ed
o each açade o he campus buildings. Diagonal scans we e also pe o med o cap u e he
in e nal aces o he açades. Depending on he dis ance be ween buildings, he emaining
scans we e dis ibu ed on he g ound. A colo poin cloud ea men was conduc ed, since
a 360
◦
pano amic iew was also cap u ed a each scan poin . Two e es ial lase scanne s
we e used: an RTC 360 and a BLK 360, bo h om Leica Geosys ems (Figu e 6). They s and
ou due o h ee cha ac e is ics: hey a e ex emely ligh , i is no necessa y o spend ime
le elling and hey cap u e 360
◦
sphe ical images in HDR in a sho ime, gene a ing a colo
poin cloud [
23
]. We ha e wo ked wi h scanne s o o he b ands, and we a e awa e ha
in such cases cap u e wi h a 360 HDR image unde 8 min is complica ed. The wo Leica
scanne s used in his s udy enabled he p ojec objec i es o be achie ed. The cha ac e is ics
and ea u es o he de ices a e e y impo an in enabling us o ob ain he esul s men ioned
in he a icle.
Buildings 2022, 12, x FOR PEER REVIEW 10 o 33
be achie ed. The cha ac e is ics and ea u es o he de ices a e e y impo an in enabling
us o ob ain he esul s men ioned in he a icle.
The i s one was he BLK360, which has a egis y ange o 60 m and ga he s 360,000
poin s/second. I has a pai o special ea u es ha make i e y in e es ing. One is ha i
has a 360 he mal came a ha is e y use ul o sus ainabili y and ene gy e iciency issues.
The o he is ha i is e y small and only weighs 1 kg. I ac ed as a complemen in some
speci ic asks o he second scanne , which was he main de ice used in ield cap u e. The
second de ice, he RTC 360, has a egis y ange o up o 130 m, which enables i o co e
la ge a eas. I is especially in e es ing om an ene gy pe spec i e, because i has VIS ech-
nology ha enables i o au oma ically egis e de ice displacemen wi hou a ge s om
one scanning poin o ano he , so ha he pa ial poin s clouds a e egis e ed in a special
loca ion ela ed o he o he scans. The de ice’s measu ing a e is up o 2 million
poin s/second, and o high esolu ion, (3 mm@10 m), i can scan in 1:42 min wi hou HDR.
Bea ing in mind ha medium esolu ion is o en enough in many cases, a poin cloud wi h
360 image can be ob ained in a scan o less han 2 min. We can sa ely say ha his is “li le
ime” in compa ison o o he scanne s. In ac , we ha e checked and ound ha he 360-
image cap u ed wi h o he 3D lase scanne s in 8 min is o poo e quali y han he one
cap u ed in one minu e wi h hese scanne s.
(a) (b)
Figu e 6. Scanne s used: (a) RTC 360; (b) BLK 360.
Unlike o he scanne s, he BLK and he RTC enable p e-p ocessing wo k o be pe -
o med on si e, ia a mobile de ice ( able o mobile phone), hanks o he Leica Cyclone
FIELD 360 applica ion. While he scanne is cap u ing poin s, we can check he esul s
ob ained and mo e o wa d wi h he p ocessing wo k, uni ing he poin s clouds o each
scan posi ion. These scanne s ansmi a Wi-Fi ne wo k ha enables he mobile de ice o
be linked o he scanne , so ha all he scanne da a can be ans e ed in eal ime o he
Cyclone FIELD 360 applica ion. Any able o mobile phone ha uses iOS o And oid can
be used o his pu pose. Rep esen s an ad ance ha u he s eamlines wo k, besides
enabling he campus digi aliza ion wo k o be e alua ed, op imized and alida ed o ob-
ain he poin s cloud in 3D. I also enables he basics o he DSR me hod o be complied
wi h: implemen , e alua e and op imize.
The scan posi ions p e-p ocessing app has se e al wo k abs. In he “map” o ma
he clouds a e joined in a plan ollowing he “cloud- o-cloud” me hod, and he accu acy
o he union in plan, sec ion and pe spec i e can be consul ed. The union mus always be
conduc ed be ween wo nea by poin clouds, which will be shown in wo di e en colo s
o acili a e he p ocess (usually o ange and blue-cyan). The “360” sec ion allows imme -
sion in each scan poin , o iew de ails. Las ly, a speci ic cloud o he assembled se o
clouds can be iewed in 3D (Figu e 7).
Figu e 6. Scanne s used: (a) RTC 360; (b) BLK 360.
The i s one was he BLK360, which has a egis y ange o 60 m and ga he s
360,000 poin s/second
. I has a pai o special ea u es ha make i e y in e es ing. One is
ha i has a 360 he mal came a ha is e y use ul o sus ainabili y and ene gy e iciency
issues. The o he is ha i is e y small and only weighs 1 kg. I ac ed as a complemen in
Buildings 2022,12, 380 16 o 32
Table 4. Con .
Wo k Phase So wa e Scan Da a Impo Impo Fo ma s Expo Fo ma s
Resul s
isualiza ion
T uView * [72] Only Leica LGS
Almos any o ma : cloud o poin s
and geome ies. Accu acy epo s.
360 images.
Je s eam Viewe * [73] Only Leica LGS Poin in o ma ion,
CAD o ma s ( ee)
ReCap Au odesk [71] Independen XYZ, E57 Au ocad (DWG), E57 PTS
RCP/RCS
* Leica.
4. Resul s
Al hough pa o he esul s we e shown in he p e ious sec ion o enable be e and
mo e g aphic unde s anding o he me hodological p ocess, his sec ion ocuses on he
pa ial esul s ha will allow he 3D simula ion model o be ob ained. Some quan i a i e
da a o he LiDAR cloud, he e o om he union o he TLS clouds and he UAV ligh
ope a ion will be shown. In addi ion, g aphic esul s o he cloud will be explained, so
ha he way da a is ex ac ed o gene a e he 3D model can be checked. Finally, some
alues om he en i onmen al assessmen will be p esen ed, hough hese a e no he
di ec objec i e o his publica ion.
4.1. ALS LiDAR Clouds
In his sec ion, he esul s o he ALS LiDAR poin cloud p oduced using he LiDAR o
public se ices in ol ed in he p ojec will be analyzed. Two main aspec s will be s udied:
he cloud’s sui abili y o modeling he açades o campus buildings, conside ing ha he
buildings’ olumes, he windows and opaque pa s o açades mus be de ined; and he
sui abili y o calcula ing biomass olumes by making sec ions o ees masses in he cloud.
As o he campus buildings, wen y buildings we e e alua ed a he UPV/EHU
campus in DSS, while 31 we e e alua ed a he UNAV campus in Pamplona [
12
]. Be o e
de e mining he scanning plan s a egy o TLS, he 51 buildings had o be analyzed by
making pa ial sec ions o he ALS cloud. Below is an example o measu emen s made in a
building on he UNAV campus, whe e he cloud has a densi y o 14 p s/m2(Figu e 11).
Buildings 2022, 12, x FOR PEER REVIEW 17 o 33
alues om he en i onmen al assessmen will be p esen ed, hough hese a e no he
di ec objec i e o his publica ion.
4.1. ALS LiDAR Clouds
In his sec ion, he esul s o he ALS LiDAR poin cloud p oduced using he LiDAR
o public se ices in ol ed in he p ojec will be analyzed. Two main aspec s will be s ud-
ied: he cloud’s sui abili y o modeling he açades o campus buildings, conside ing ha
he buildings’ olumes, he windows and opaque pa s o açades mus be de ined; and
he sui abili y o calcula ing biomass olumes by making sec ions o ees masses in he
cloud.
As o he campus buildings, wen y buildings we e e alua ed a he UPV/EHU cam-
pus in DSS, while 31 we e e alua ed a he UNAV campus in Pamplona [12]. Be o e de-
e mining he scanning plan s a egy o TLS, he 51 buildings had o be analyzed by mak-
ing pa ial sec ions o he ALS cloud. Below is an example o measu emen s made in a
building on he UNAV campus, whe e he cloud has a densi y o 14 p s/m
2
(Figu e 11).
(a)
(b)
Figu e 11. 2017 LiDAR cloud, densi y: 14 poin s/m
2
. UNAV campus: (a) measu emen s on he poin
cloud o he campus’s cen al building; (b) g aphic su ey o he same building açade.
As an example, a se ies o basic measu emen s we e made o calcula e building
heigh , açade su ace and openings on he no h açade o he cen al building a he
UNAV uni e si y campus in Pamplona. The poin densi y alue o 14 p/m
2
allows ap-
p oxima e measu emen s o pa o he building elemen s o be ob ained, whose accu acy
may some imes be su icien o he 3D simula ion model. The esul s ob ained in he ex-
ample a e shown below (Table 5).
Table 5. Example o açade measu emen able in he ALS LiDAR cloud.
Facade
Leng h Heigh o Ea e Facade Su ace Window
Dimensions F0/F1
Window
Dimensions F2/F3
51.90 m. 15.50 m. 804.45 m
2
1.55 × 1.35 m. 1.55 × 2.30 m.
I was hus possible o ob ain measu emen s o all buildings on he UNAV campus;
i any addi ional measu emen was necessa y, supplemen a y measu emen s om he
Figu e 11.
2017 LiDAR cloud, densi y: 14 poin s/m
2
. UNAV campus: (
a
) measu emen s on he poin
cloud o he campus’s cen al building; (b) g aphic su ey o he same building açade.
Buildings 2022,12, 380 17 o 32
As an example, a se ies o basic measu emen s we e made o calcula e building
heigh , açade su ace and openings on he no h açade o he cen al building a he
UNAV uni e si y campus in Pamplona. The poin densi y alue o 14 p/m
2
allows
app oxima e measu emen s o pa o he building elemen s o be ob ained, whose accu acy
may some imes be su icien o he 3D simula ion model. The esul s ob ained in he
example a e shown below (Table 5).
Table 5. Example o açade measu emen able in he ALS LiDAR cloud.
Facade
Leng h Heigh o Ea e Facade Su ace Window
Dimensions F0/F1
Window
Dimensions F2/F3
51.90 m. 15.50 m. 804.45 m21.55 ×1.35 m. 1.55 ×2.30 m.
I was hus possible o ob ain measu emen s o all buildings on he UNAV campus; i
any addi ional measu emen was necessa y, supplemen a y measu emen s om he cloud
ob ained wi h TLS echniques was used. On he UPV/EHU campus, he LiDAR clouds in
Gipuzkoa p o ince we e no able o ob ain he same esul s. I was only possible o ob ain
measu emen s o açades (heigh and wid h), bu no o window opening sizes. To conduc
ha , a mo e exhaus i e cap u e had o be pe o med wi h TLS. The da a used o model
buildings on he UPV/EHU campus in DSS was ex ac ed di ec ly om he LiDAR clouds
ob ained wi h TLS echniques.
The esul s we e also analyzed o ob ain he campuses’ app oxima e o es biomass
olume. The measu emen o an isola ed ee o ege a ion elemen will be used he e as an
example, s a ing wi h analysis o he sui abili y o he LiDAR clouds a he UPV/EHU
campus. In Figu e 6, he 2012 cloud ba ely shows poin s o ege a ion o soil. The 2017
cloud o he wo campuses is analyzed in compa ison (Figu es 12 and 13).
Buildings 2022, 12, x FOR PEER REVIEW 18 o 33
cloud ob ained wi h TLS echniques was used. On he UPV/EHU campus, he LiDAR
clouds in Gipuzkoa p o ince we e no able o ob ain he same esul s. I was only possible
o ob ain measu emen s o açades (heigh and wid h), bu no o window opening sizes.
To conduc ha , a mo e exhaus i e cap u e had o be pe o med wi h TLS. The da a used
o model buildings on he UPV/EHU campus in DSS was ex ac ed di ec ly om he Li-
DAR clouds ob ained wi h TLS echniques.
The esul s we e also analyzed o ob ain he campuses’ app oxima e o es biomass
olume. The measu emen o an isola ed ee o ege a ion elemen will be used he e as
an example, s a ing wi h analysis o he sui abili y o he LiDAR clouds a he UPV/EHU
campus. In Figu e 6, he 2012 cloud ba ely shows poin s o ege a ion o soil. The 2017
cloud o he wo campuses is analyzed in compa ison (Figu es 12 and 13).
Figu e 12. Isola ed ege a ion elemen . 2017 LiDAR cloud, densi y: 2.2 poin s/m
2
. UPV/EHU cam-
pus in DSS.
In he cloud in Figu e 12, he e ain’s con igu a ion can be obse ed, hough i is
di icul o measu e ee olume. A e making se e al measu emen s in ha LiDAR
cloud, he es ima e ob ained has a maximum p ecision o 30 cm in he XY axes and 20 cm
in he Z axis. In con as , i was e i ied ha in he LiDAR cloud o he UNAV campus in
Pamplona, he ee masses ha e enough p ecision o make measu emen s. In Figu e 13
he ee’s heigh is exac ly 22.45 m.
Figu e 13. 2017 ligh LiDAR poin cloud, densi y: 14 poin s/m
2
. Isola ed ege a ion elemen whe e
he comple e sec ion o i s mass is obse ed, as well as le el o he g ound.
Many publica ions p esen mul iple ways o calcula e ee olumes [74]. Conside ing
he equi emen s o he NEST e alua ion so wa e, calcula ions ha e been pe o med in
wo ways: o isola ed ees such as he one in he example, hei olume is assimila ed o
Figu e 12.
Isola ed ege a ion elemen . 2017 LiDAR cloud, densi y: 2.2 poin s/m
2
. UPV/EHU
campus in DSS.
In he cloud in Figu e 12, he e ain’s con igu a ion can be obse ed, hough i is
di icul o measu e ee olume. A e making se e al measu emen s in ha LiDAR cloud,
he es ima e ob ained has a maximum p ecision o 30 cm in he XY axes and 20 cm in he Z
axis. In con as , i was e i ied ha in he LiDAR cloud o he UNAV campus in Pamplona,
he ee masses ha e enough p ecision o make measu emen s. In Figu e 13 he ee’s heigh
is exac ly 22.45 m.
Many publica ions p esen mul iple ways o calcula e ee olumes [
74
]. Conside ing
he equi emen s o he NEST e alua ion so wa e, calcula ions ha e been pe o med in
wo ways: o isola ed ees such as he one in he example, hei olume is assimila ed
o a cone, cylinde o sphe e [
75
]. In Figu e 13b i is assimila ed o a cone. The ee’s
o al heigh and he base o he b anches a e measu ed; he olume o he cone is hen
calcula ed. Fo con inuous masses, such as in Figu e 4, pa ial sec ions o he cloud a e
used, and he app oxima e olume is ex ac ed conside ing he con ou s o he mass. To
Buildings 2022,12, 380 18 o 32
calcula e he inal olume o he campus’s biomass, ha ee da a mus be comple ed wi h
he olumes o g een a eas associa ed o he e ain’s g een su aces. In he case o he
UNAV campus, i s o al a ea was calcula ed as being 1,547,278 m
2
, wi h a g een space
su ace a ea o
1,082,210 m2
, accoun ing o 70% o he o al. A he UPV/EHU campus
in DSS, he o al a ea o he campus was es ima ed o be 565,140 m
2
, wi h a g een space
a ea o 168,816 m
2
, accoun ing o app oxima ely 30% o he o al. Nume ous publica ions
ha e explained di e en ways o calcula ing o es biomass [
75
–
77
]. Wi h his da a i was
possible o eed he NEST model o ca y ou he e alua ion.
Buildings 2022, 12, x FOR PEER REVIEW 18 o 33
cloud ob ained wi h TLS echniques was used. On he UPV/EHU campus, he LiDAR
clouds in Gipuzkoa p o ince we e no able o ob ain he same esul s. I was only possible
o ob ain measu emen s o açades (heigh and wid h), bu no o window opening sizes.
To conduc ha , a mo e exhaus i e cap u e had o be pe o med wi h TLS. The da a used
o model buildings on he UPV/EHU campus in DSS was ex ac ed di ec ly om he Li-
DAR clouds ob ained wi h TLS echniques.
The esul s we e also analyzed o ob ain he campuses’ app oxima e o es biomass
olume. The measu emen o an isola ed ee o ege a ion elemen will be used he e as
an example, s a ing wi h analysis o he sui abili y o he LiDAR clouds a he UPV/EHU
campus. In Figu e 6, he 2012 cloud ba ely shows poin s o ege a ion o soil. The 2017
cloud o he wo campuses is analyzed in compa ison (Figu es 12 and 13).
Figu e 12. Isola ed ege a ion elemen . 2017 LiDAR cloud, densi y: 2.2 poin s/m
2
. UPV/EHU cam-
pus in DSS.
In he cloud in Figu e 12, he e ain’s con igu a ion can be obse ed, hough i is
di icul o measu e ee olume. A e making se e al measu emen s in ha LiDAR
cloud, he es ima e ob ained has a maximum p ecision o 30 cm in he XY axes and 20 cm
in he Z axis. In con as , i was e i ied ha in he LiDAR cloud o he UNAV campus in
Pamplona, he ee masses ha e enough p ecision o make measu emen s. In Figu e 13
he ee’s heigh is exac ly 22.45 m.
Figu e 13. 2017 ligh LiDAR poin cloud, densi y: 14 poin s/m
2
. Isola ed ege a ion elemen whe e
he comple e sec ion o i s mass is obse ed, as well as le el o he g ound.
Many publica ions p esen mul iple ways o calcula e ee olumes [74]. Conside ing
he equi emen s o he NEST e alua ion so wa e, calcula ions ha e been pe o med in
wo ways: o isola ed ees such as he one in he example, hei olume is assimila ed o
Figu e 13.
2017 ligh LiDAR poin cloud, densi y: 14 poin s/m
2
. Isola ed ege a ion elemen whe e
he comple e sec ion o i s mass is obse ed, as well as le el o he g ound.
4.2. TLS LiDAR Clouds
The measu emen s ha could no be ob ained om he ALS LiDAR cloud we e made
om he cloud supplemen ed by TLS echniques. All kinds o geome ic da a (leng hs,
angles, a eas, e c.) and he mal da a o he poin s in he cloud can be ex ac ed om ha
cloud using lase scanning. A he end o his sec ion, in he isualiza ion pa o he
esul ing clouds, some examples o quan i a i e esul s in cloud measu emen s can be seen.
Al hough we ha e shown ha he ield wo k in 7 days enabled enough esul s o be
ob ained o en e all he necessa y da a in NEST, we shall show how he p e ious es ima e
o 7 days was calcula ed. The basis o e e y hing is o d aw up di e en scanning plans,
placing he scan posi ions a di e en dis ances. The lesse he dis ance, he mo e scan
poin s he e a e, and he e o e, mo e scanning ime. The scanning ime can be calcula ed
wi h he da a we show in esponse 2.6. We could es ima e a mean scanning ime o 2 min
( he maximum would be 2:42 wi h an HDR image in colo ).
3D lase scanne s p esen da a a esolu ions o p ecisions exp essed a 10 m. In he
case o he RTC360, wi h a measu emen a e o up o 2 million poin s/second, he da a is
as ollows: Fo high esolu ion, (3 mm@10 m); o medium esolu ion, (6 mm@10 m); o
low esolu ion, (12 mm@10 m). I we wan o make su e o hese esolu ions, he logical
hing o do is es ablish a scanning plan wi h he scan poin s a 10 m. Howe e , such a
decision would mean ha he scan wi h TLS would ake much longe . The UNAV campus
is 113 Ha., 70% a e g een spaces, 10% is occupied by buildings and he o he 20% consis s
o ca pa ks, oads, e c. Th ee calcula ion scena ios we e es ablished:
In (I) we se ou o es ima e a scanning plan wi h a benchma k dis ance o 10 m.
100 scan posi ions/Ha
. we e es ima ed in he g een spaces, making a o al o 7910 scan/pos.
31 buildings associa ed wi h he campus and i s ac i i ies had o be modelled and simula ed.
Only he ex e io geome y o he acades had o be cap u ed. In all, 775 scans we e
es ima ed o he acade pe ime e s (an a e age o 25 scans pe building). The o he a eas,
(pa king, oads, e c.) we e no as impo an o he en i onmen al e alua ion o he model,
al hough hey ake up a lo o su ace a ea, and so a o al o 678 scan/pos was es ima ed.
The o al scan posi ions was 9363 ha , wi h a scanning ime o 2:42 min, made o a o al o
Buildings 2022,12, 380 19 o 32
52 days o ield wo k wi hou including he displacemen s o he scanne om one posi ion
o ano he .
In scena io (II) we es ima ed a scanning plan a 33 m in he g een spaces, wi h an es i-
ma e o 16 scan posi ions pe Ha. wi h a o al o 1265 scan/pos. Scan posi ions we e planned
a 30 m in he building pe ime e s, bu wi h a minimum o 3 scans pe acade.
258 scan/pos
we e calcula ed o he cha ac e is ics o he buildings. In all,
339 scan/pos
we e es ima ed
o he es . The o al numbe o scan posi ions was 1862, which wi h a scanning ime o
2:42 min makes a o al o 10.5 days o wo k wi hou including he displacemen s o he
scanne om one posi ion o ano he . This would mean o e 2 weeks’ wo k.
In scena io (III) we es ima ed a scanning plan in he g een spaces a dis ances unde
50 m
, wi h an es ima e o 8 scan posi ions pe Ha. wi h a o al o 632 scan/pos. The
p e ious plan o 30 m wi h 258 scan/pos was main ained o he building pe ime e s. In
all, 169 scan/pos we e es ima ed o he es . The o al numbe o scan posi ions was 1059,
which wi h a scanning ime o 2:42 min makes a o al o 6 days ield wo k. Displacemen s
o he scanne om one scan/posi ion o ano he (a minimum es ima e o 30 s) adds one
mo e day. This makes o a o al o 7 days. The es ima es calcula ed o wo king days o
8 h/day
, al hough in ea ly Ap il he e was mo e han 12 h o na u al ligh a day, which
ga e a deg ee o ma gin o con ingencies in he 7 days.
A e p esen ing he es ima ion o he educ ion o scan p ocess in he ield, nex
sec ion shows how he cloud p ocessing esul s ob ained wi h TLS can be imp o ed. Since
he LiDAR clouds al eady ha e basic in o ma ion on which o add he TLS clouds, he e o
and o e lap o he esul ing clouds is no he same as when scanning a single building using
only TLS echniques. The main accu acy pa ame e s o upg ade a e: se e o , o e lap,
s eng h o link and cloud- o-cloud e o . An example o he di ec esul a e he scanning
and p e-p ocessing phase o Zone 2 o he UPV/EHU campus in DSS is shown (Figu e 14).
Buildings 2022, 12, x FOR PEER REVIEW 20 o 33
In scena io (III) we es ima ed a scanning plan in he g een spaces a dis ances unde
50 m, wi h an es ima e o 8 scan posi ions pe Ha. wi h a o al o 632 scan/pos. The p e i-
ous plan o 30 m wi h 258 scan/pos was main ained o he building pe ime e s. In all, 169
scan/pos we e es ima ed o he es . The o al numbe o scan posi ions was 1059, which
wi h a scanning ime o 2:42 min makes a o al o 6 days ield wo k. Displacemen s o he
scanne om one scan/posi ion o ano he (a minimum es ima e o 30 s) adds one mo e
day. This makes o a o al o 7 days. The es ima es calcula ed o wo king days o 8 h/day,
al hough in ea ly Ap il he e was mo e han 12 h o na u al ligh a day, which ga e a
deg ee o ma gin o con ingencies in he 7 days.
A e p esen ing he es ima ion o he educ ion o scan p ocess in he ield, nex sec-
ion shows how he cloud p ocessing esul s ob ained wi h TLS can be imp o ed. Since
he LiDAR clouds al eady ha e basic in o ma ion on which o add he TLS clouds, he
e o and o e lap o he esul ing clouds is no he same as when scanning a single build-
ing using only TLS echniques. The main accu acy pa ame e s o upg ade a e: se e o ,
o e lap, s eng h o link and cloud- o-cloud e o . An example o he di ec esul a e
he scanning and p e-p ocessing phase o Zone 2 o he UPV/EHU campus in DSS is
shown (Figu e 14).
Figu e 14. Re iew and op imiza ion o he scan o he Donos ia-San Sebas ián campus in plan,
using he Leica Cyclone Regis e , Zone 2.
As de ailed p e iously, he p ocessing phase has se e al s ages sepa a ed in ou abs
in he so wa e. The i s allows he da a collec ed on-si e o be impo ed. In his case, as
p e-p ocessing o p e- egis a ion has been conduc ed, he scan poin s appea linked in
he impo om Cyclone FIELD 360 o Leica Cyclone REGISTER 360. A his poin , he
p ocessing so wa e analyzes he join da a in he ield and assigns a colo o each join
based on i s s eng h and accu acy. G een indica es he highes s eng h and ed he low-
es s eng h; wo o he colo s, yellow and blue, a e in be ween. In he case o Figu e 16,
he inpu da a o he p ocessing ma ks he ollowing esul s: se e o 1 mm, o e lap 33%,
s eng h 34%, cloud- o-cloud e o 1 mm. This da a could be op imized in REGISTER 360
by op imizing he cloud- o-cloud join s, hough he 1 mm assembly e o is mo e han
enough o mee he needs o he 3D model in NEST. Fu he mo e, in en i onmen s wi h
la ge ege a ion, al hough he scan’s accu acy is high (1–3 mm e o ), i may happen ha
he o e lap be ween clouds is no as app op ia e, due o he singula i ies o mo ing
b anches and lea es (Figu e 15).
Figu e 14.
Re iew and op imiza ion o he scan o he Donos ia-San Sebas ián campus in plan, using
he Leica Cyclone Regis e , Zone 2.
As de ailed p e iously, he p ocessing phase has se e al s ages sepa a ed in ou abs
in he so wa e. The i s allows he da a collec ed on-si e o be impo ed. In his case, as
p e-p ocessing o p e- egis a ion has been conduc ed, he scan poin s appea linked in
he impo om Cyclone FIELD 360 o Leica Cyclone REGISTER 360. A his poin , he
p ocessing so wa e analyzes he join da a in he ield and assigns a colo o each join
based on i s s eng h and accu acy. G een indica es he highes s eng h and ed he lowes
s eng h; wo o he colo s, yellow and blue, a e in be ween. In he case o Figu e 16, he
inpu da a o he p ocessing ma ks he ollowing esul s: se e o 1 mm, o e lap 33%,
s eng h 34%, cloud- o-cloud e o 1 mm. This da a could be op imized in REGISTER 360 by
op imizing he cloud- o-cloud join s, hough he 1 mm assembly e o is mo e han enough
Buildings 2022,12, 380 20 o 32
o mee he needs o he 3D model in NEST. Fu he mo e, in en i onmen s wi h la ge
ege a ion, al hough he scan’s accu acy is high (1–3 mm e o ), i may happen ha he
o e lap be ween clouds is no as app op ia e, due o he singula i ies o mo ing b anches
and lea es (Figu e 15).
Buildings 2022, 12, x FOR PEER REVIEW 21 o 33
Figu e 15. TLS LiDAR clouds in he ee-lined a ea o he UPV/EHU DSS campus. Fo es biomass
da a can be ob ained om hese clouds, which o se he lack o ALS LiDAR clouds.
A he UNAV campus, he p ocessing s age was e y simila o ha o he Donos ia
campus. Al hough, since he ALS LiDAR cloud is much mo e accu a e, he scanning
poin s pe m
2
o campus a e much lowe . Conside ing also ha 70% o he campus com-
p ises g een su aces, means ha he e a e no e y eliable e e ences be ween adjoining
clouds. A e he scan da a dump be o e p ocessing, he ini ial se he e o e had much
ewe join s in g een (because o ha lack o s eng h and o e lap). Al hough he e o
was accep able (2 mm), ini ially, he o ce was only 22% and he o e lap 19%. Indeed, in
some a eas he e is a p e ious join ha he so wa e s opped linking so ha i could be
s udied and imp o ed du ing p ocessing (Figu e 16).
Figu e 16. 3D poin cloud. Re iew and op imiza ion o scan o he UPV/EHU Campus, DSS, Zone
1. Desc ip ion o he join s be o e he p ocessing and op imiza ion s age.
As he scan posi ions ha e been c ea ed a longe dis ances han usual o s eamline
he p ocess, he p e- egis a ion wo k conduc ed on si e unde goes a e ision. Since image
16 is in pe spec i e, i is ha de o see he colo o he unions be ween scan poin s. When
he da a ob ained in he ield is impo ed, he so wa e analyzes whe he he poin s o a
cloud o a scan posi ion o e lap enough wi h hose o he p e ious and pos e io clouds.
The so wa e checks and colo s he unions in line wi h h ee concep s: E o , s eng h o
union and o e lap be ween clouds. I hey ha e any undesi able pa ame e s, he may
colo hem in ed, yellow o blue o i may no di ec ly p opose he union because i is
ou side a minimum ange. Wha happens hen, is ha an op imiza ion p ocess com-
mences analyses and imp o es he cloud- o-cloud union (as in Figu e 8), be ween he wo
clouds ha do no ha e a connec ion line in g een (Figu es 14 and 16). I we can imp o e
he o e lap pa ame e s in his op imiza ion p ocess, he line o union changes o g een
and i is accep ed as alid. I we canno change he union o g een in he op imiza ion
p ocess, he e is s ill he op ion o ca ying ou ano he scan on si e he nex day and u -
he s eng hening he cloud o he se . This means ha he ield wo k should be p ocessed
e e y day, and so i is impo ance o use he ools men ioned in his a icle. The de ices,
Figu e 15.
TLS LiDAR clouds in he ee-lined a ea o he UPV/EHU DSS campus. Fo es biomass
da a can be ob ained om hese clouds, which o se he lack o ALS LiDAR clouds.
A he UNAV campus, he p ocessing s age was e y simila o ha o he Donos ia
campus. Al hough, since he ALS LiDAR cloud is much mo e accu a e, he scanning poin s
pe m
2
o campus a e much lowe . Conside ing also ha 70% o he campus comp ises
g een su aces, means ha he e a e no e y eliable e e ences be ween adjoining clouds.
A e he scan da a dump be o e p ocessing, he ini ial se he e o e had much ewe join s
in g een (because o ha lack o s eng h and o e lap). Al hough he e o was accep able
(2 mm), ini ially, he o ce was only 22% and he o e lap 19%. Indeed, in some a eas he e is
a p e ious join ha he so wa e s opped linking so ha i could be s udied and imp o ed
du ing p ocessing (Figu e 16).
Buildings 2022, 12, x FOR PEER REVIEW 21 o 33
Figu e 15. TLS LiDAR clouds in he ee-lined a ea o he UPV/EHU DSS campus. Fo es biomass
da a can be ob ained om hese clouds, which o se he lack o ALS LiDAR clouds.
A he UNAV campus, he p ocessing s age was e y simila o ha o he Donos ia
campus. Al hough, since he ALS LiDAR cloud is much mo e accu a e, he scanning
poin s pe m
2
o campus a e much lowe . Conside ing also ha 70% o he campus com-
p ises g een su aces, means ha he e a e no e y eliable e e ences be ween adjoining
clouds. A e he scan da a dump be o e p ocessing, he ini ial se he e o e had much
ewe join s in g een (because o ha lack o s eng h and o e lap). Al hough he e o
was accep able (2 mm), ini ially, he o ce was only 22% and he o e lap 19%. Indeed, in
some a eas he e is a p e ious join ha he so wa e s opped linking so ha i could be
s udied and imp o ed du ing p ocessing (Figu e 16).
Figu e 16. 3D poin cloud. Re iew and op imiza ion o scan o he UPV/EHU Campus, DSS, Zone
1. Desc ip ion o he join s be o e he p ocessing and op imiza ion s age.
As he scan posi ions ha e been c ea ed a longe dis ances han usual o s eamline
he p ocess, he p e- egis a ion wo k conduc ed on si e unde goes a e ision. Since image
16 is in pe spec i e, i is ha de o see he colo o he unions be ween scan poin s. When
he da a ob ained in he ield is impo ed, he so wa e analyzes whe he he poin s o a
cloud o a scan posi ion o e lap enough wi h hose o he p e ious and pos e io clouds.
The so wa e checks and colo s he unions in line wi h h ee concep s: E o , s eng h o
union and o e lap be ween clouds. I hey ha e any undesi able pa ame e s, he may
colo hem in ed, yellow o blue o i may no di ec ly p opose he union because i is
ou side a minimum ange. Wha happens hen, is ha an op imiza ion p ocess com-
mences analyses and imp o es he cloud- o-cloud union (as in Figu e 8), be ween he wo
clouds ha do no ha e a connec ion line in g een (Figu es 14 and 16). I we can imp o e
he o e lap pa ame e s in his op imiza ion p ocess, he line o union changes o g een
and i is accep ed as alid. I we canno change he union o g een in he op imiza ion
p ocess, he e is s ill he op ion o ca ying ou ano he scan on si e he nex day and u -
he s eng hening he cloud o he se . This means ha he ield wo k should be p ocessed
e e y day, and so i is impo ance o use he ools men ioned in his a icle. The de ices,
Figu e 16.
3D poin cloud. Re iew and op imiza ion o scan o he UPV/EHU Campus, DSS, Zone 1.
Desc ip ion o he join s be o e he p ocessing and op imiza ion s age.
As he scan posi ions ha e been c ea ed a longe dis ances han usual o s eamline he
p ocess, he p e- egis a ion wo k conduc ed on si e unde goes a e ision. Since
image 16
is in pe spec i e, i is ha de o see he colo o he unions be ween scan poin s. When
he da a ob ained in he ield is impo ed, he so wa e analyzes whe he he poin s o a
cloud o a scan posi ion o e lap enough wi h hose o he p e ious and pos e io clouds.
The so wa e checks and colo s he unions in line wi h h ee concep s: E o , s eng h o
union and o e lap be ween clouds. I hey ha e any undesi able pa ame e s, he may
colo hem in ed, yellow o blue o i may no di ec ly p opose he union because i is
ou side a minimum ange. Wha happens hen, is ha an op imiza ion p ocess commences
analyses and imp o es he cloud- o-cloud union (as in Figu e 8), be ween he wo clouds
ha do no ha e a connec ion line in g een (Figu es 14 and 16). I we can imp o e he
o e lap pa ame e s in his op imiza ion p ocess, he line o union changes o g een and i is
Buildings 2022,12, 380 21 o 32
accep ed as alid. I we canno change he union o g een in he op imiza ion p ocess, he e
is s ill he op ion o ca ying ou ano he scan on si e he nex day and u he s eng hening
he cloud o he se . This means ha he ield wo k should be p ocessed e e y day, and
so i is impo ance o use he ools men ioned in his a icle. The de ices, so wa e and
applica ions enable p e-p ocessing ha g ea ly educes he amoun o daily p ocessing
wo k; hey also make i easie o check ha he on-si e cap u e is sa is ac o y.
Once he p ocessing s age is inished, o iew he esul s and ex ac he necessa y
in o ma ion he mos app op ia e ile o ma is LGS. The ee Leica Je s eam Viewe
so wa e enables he iewing and consul a ion o da a om he digi al model comp ising
he se o poin clouds and he 360
◦
images o each o he scan posi ions. This applica ion
allows a isual and me ic inspec ion o be ca ied ou i ually along he ou e. Wi h his
esou ce you can accomplish he analysis, e i ica ion and da a ex ac ion asks needed
o he modeling and simula ion p ocess in NEST, such as dis ances, a eas, angles o e en
su ace empe a u es (Figu e 17).
Buildings 2022, 12, x FOR PEER REVIEW 22 o 33
so wa e and applica ions enable p e-p ocessing ha g ea ly educes he amoun o daily
p ocessing wo k; hey also make i easie o check ha he on-si e cap u e is sa is ac o y.
Once he p ocessing s age is inished, o iew he esul s and ex ac he necessa y
in o ma ion he mos app op ia e ile o ma is LGS. The ee Leica Je s eam Viewe so -
wa e enables he iewing and consul a ion o da a om he digi al model comp ising he
se o poin clouds and he 360° images o each o he scan posi ions. This applica ion
allows a isual and me ic inspec ion o be ca ied ou i ually along he ou e. Wi h his
esou ce you can accomplish he analysis, e i ica ion and da a ex ac ion asks needed
o he modeling and simula ion p ocess in NEST, such as dis ances, a eas, angles o e en
su ace empe a u es (Figu e 17).
(a)
(b)
Figu e 17. Visualiza ion o esul s o he TLS LiDAR clouds in Je s eam Viewe , ob aining quan i-
a i e da a o he model: (a) 3D colo ed cloud isualiza ion, geome ic da a, Gipuzkoa School o
Enginee ing building, UPV/EHU campus in DSS; (b) sou hwes açade o he same building. Visu-
aliza ion om he 360° image o he building, geome ic da a (g een and g ay) and he mal da a
( ed).
Be o e s a ing o su ey he campuses wi h TLS, he easibili y was analyzed in
e ms o esou ces and ime needed o comple e he ALS LiDAR cloud. TLS un imes we e
calcula ed o compa ison o UAV ones, wi h au oma ed pho og amme y. In he case o
Figu e 17.
Visualiza ion o esul s o he TLS LiDAR clouds in Je s eam Viewe , ob aining quan-
i a i e da a o he model: (
a
) 3D colo ed cloud isualiza ion, geome ic da a, Gipuzkoa School
o Enginee ing building, UPV/EHU campus in DSS; (
b
) sou hwes açade o he same building.
Visualiza ion om he 360
◦
image o he building, geome ic da a (g een and g ay) and he mal
da a ( ed).
Buildings 2022,12, 380 22 o 32
Be o e s a ing o su ey he campuses wi h TLS, he easibili y was analyzed in e ms
o esou ces and ime needed o comple e he ALS LiDAR cloud. TLS un imes we e
calcula ed o compa ison o UAV ones, wi h au oma ed pho og amme y. In he case o
he UPV/EHU campus in DSS, he con igu a ion o he u ban en i onmen , i s dimensions
and he buildings’ closeness enable e y as complemen a y cap u e wi h TLS, o which
addi ional wo k o h ee days was es ima ed, in h ee zones. Howe e , he UNAV campus
in Pamplona p esen s g oups o buildings, hough wi h la ge dis ances be ween some o
hem. Cap u ing he en i e campus wi h TLS would ha e aken se e al weeks, wi h an
o e whelming amoun o in o ma ion. Howe e , as he ALS LiDAR cloud, wi h a densi y
o 14 p s/m
2
, p o ided e y accep able measu emen esul s, he complemen a y su ey
wi h TLS was es ima ed, a e p e ious scanning plans, o ake se en days, one day o
each zone. A e he TLS calcula ions, he esou ces we e es ima ed wi h he UAV o decide
which would be he as es and mos e icien combina ion.
4.3. Cap u e by UAV and S M Pho og amme ic P ocessing
The combina ion o da a collec ion by UAV and he S M au oma ed pho og amme y
me hod acili a es documen a ion due o he simplici y o he p ocess and he wid h o he
wo king ange [
78
,
79
]. The s udy ocuses on an assessmen o he ope a ion and e iciency
o he cap u e o hese la ge u ban a eas [
14
]. I conce ns assessmen o whe he i could
be as e and mo e e icien han addi ional cap u es wi h TLS o comple e he ALS LiDAR
cloud o he campuses. A he UPV/EHU campus, ligh es ic ions due o he loca ion in
con olled ai space made ope a ions di icul . Fo ha eason, i was ul ima ely decided o
no comple e he clouds wi h UAV-assis ed pho og amme y.
The su ey o ecas was pe o med a he UNAV campus in Pamplona. The p o-
g ammable applica ion (DJI-GS P o) allows con igu a ion o he au oma ic ligh mission
h ough na iga ion based on sa elli e posi ioning (GNSS). The UAV used is a DJI Phan om
4 P o model, equipped wi h a 1” CMOS senso ha educes adial dis o ion and imp o es
he me ic quali y o he S M es i u ion me hod [
80
]. A g id comp ising 147 squa e sec o s
measu ing 100 m on each side and an a ea o 1 Ha was c ea ed, es ablishing a low-al i ude
ligh pa ame e (37.3 m) o gua an ee he model’s accu acy [
5
]. This al i ude o e s a GSD
ac o o 1 cm/pixel wi h an o e lap be ween pho os, on and side, o 75%. The combined
use o nadi sho s (gimbal il —90
◦
) and oblique sho s (gimbal il —45
◦
) was de e mined.
One nadi and ou oblique sho s we e planned pe sec o , he la e aligned wi h he ou
ajec o ies ha join he e ices o each sec o o i s cen e (Figu e 18).
The esul s o he esou ce easibili y s udy a e p esen ed in Table 6.
Table 6. Summa y o UAV pho og amme y cap u e da a.
Sec o s Sec o s
O e lap
Fligh
Heigh
G ound
Sample
Dis ance GSD
Nadi
Sho
(−90◦)
Oblique
Sho
(−65◦)
F on al
O e lap
La e al
O e lap
Fligh
Time Ba e ies Pho os
Sec o ype
1 ha - 37.3 m 1 cm/pixel 1 4 75% 75% 100 min 5 se s 450
Se o sec o s
147 ha
10 m.
(10%) 37.3 m 1 cm/pixel 147 588 75% 75% 245 h 735 se s 66,150
Cap u ing da a om he en i e campus would equi e 245 ligh hou s, which, added
o he p epa a ion wo k, could mean a ield ask o mo e han one mon h. As p e iously
jus i ied, he addi ional wo ks o he ALS LiDAR cloud wi h TLS echniques was es ima ed
o ake se en days, so he UAV was used o complemen he TLS da a in hose sec o s o
he 147 planned sec o s whe e he e a e no buildings.
Buildings 2022,12, 380 23 o 32
Buildings 2022, 12, x FOR PEER REVIEW 23 o 33
he UPV/EHU campus in DSS, he con igu a ion o he u ban en i onmen , i s dimensions
and he buildings’ closeness enable e y as complemen a y cap u e wi h TLS, o which
addi ional wo k o h ee days was es ima ed, in h ee zones. Howe e , he UNAV campus
in Pamplona p esen s g oups o buildings, hough wi h la ge dis ances be ween some o
hem. Cap u ing he en i e campus wi h TLS would ha e aken se e al weeks, wi h an
o e whelming amoun o in o ma ion. Howe e , as he ALS LiDAR cloud, wi h a densi y
o 14 p s/m2, p o ided e y accep able measu emen esul s, he complemen a y su ey
wi h TLS was es ima ed, a e p e ious scanning plans, o ake se en days, one day o
each zone. A e he TLS calcula ions, he esou ces we e es ima ed wi h he UAV o de-
cide which would be he as es and mos e icien combina ion.
4.3. Cap u e by UAV and S M Pho og amme ic P ocessing
The combina ion o da a collec ion by UAV and he S M au oma ed pho og amme y
me hod acili a es documen a ion due o he simplici y o he p ocess and he wid h o he
wo king ange [78,79]. The s udy ocuses on an assessmen o he ope a ion and e iciency
o he cap u e o hese la ge u ban a eas [14]. I conce ns assessmen o whe he i could
be as e and mo e e icien han addi ional cap u es wi h TLS o comple e he ALS LiDAR
cloud o he campuses. A he UPV/EHU campus, ligh es ic ions due o he loca ion in
con olled ai space made ope a ions di icul . Fo ha eason, i was ul ima ely decided
o no comple e he clouds wi h UAV-assis ed pho og amme y.
The su ey o ecas was pe o med a he UNAV campus in Pamplona. The p og am-
mable applica ion (DJI-GS P o) allows con igu a ion o he au oma ic ligh mission
h ough na iga ion based on sa elli e posi ioning (GNSS). The UAV used is a DJI Phan om
4 P o model, equipped wi h a 1” CMOS senso ha educes adial dis o ion and imp o es
he me ic quali y o he S M es i u ion me hod [80]. A g id comp ising 147 squa e sec o s
measu ing 100 m on each side and an a ea o 1 Ha was c ea ed, es ablishing a low-al i ude
ligh pa ame e (37.3 m) o gua an ee he model’s accu acy [5]. This al i ude o e s a GSD
ac o o 1 cm/pixel wi h an o e lap be ween pho os, on and side, o 75%. The combined
use o nadi sho s (gimbal il —90°) and oblique sho s (gimbal il —45°) was de e mined.
One nadi and ou oblique sho s we e planned pe sec o , he la e aligned wi h he ou
ajec o ies ha join he e ices o each sec o o i s cen e (Figu e 18).
(a) (b)
Figu e 18.
(
a
) Wo k planning p e ision in 147 sec o s; (
b
) planning o ligh di ec ions in one o he
sec o s, om op o bo om: 1. Shoo ing o e lap, 2. Nadi al shoo ing/G.P.A.:
−
90
◦
/C.A.: 0
◦
, 3.
Oblique shoo ing/G.P.A.:
−
45
◦
/C.A.: 45
◦
, 4. Oblique shoo ing/G.P.A.:
−
45
◦
/C.A.: 135
◦
, 5. Oblique
shoo ing/G.P.A.:
−
45
◦
/C.A.: 225
◦
, 6. Oblique shoo ing/G.P.A.:
−
45
◦
/C.A.: 315
◦
, 7. O hopho o,
(G.P.A: Gimbal Pi ch Angle; C.A: Cou se Angle).
4.4. Modeling
The poin cloud o he ensemble will be used o pe o m he simula ion 3D model o
each campus in NEST (Figu e 19).
Buildings 2022, 12, x FOR PEER REVIEW 24 o 33
Figu e 18. (a) Wo k planning p e ision in 147 sec o s; (b) planning o ligh di ec ions in one o he
sec o s, om op o bo om: 1. Shoo ing o e lap, 2. Nadi al shoo ing/G.P.A.: −90°/C.A.: 0°, 3. Oblique
shoo ing/G.P.A.: −45°/C.A.: 45°, 4. Oblique shoo ing/G.P.A.: −45°/C.A.: 135°, 5. Oblique shoo -
ing/G.P.A.: −45°/C.A.: 225°, 6. Oblique shoo ing/G.P.A.: −45°/C.A.: 315°, 7. O hopho o, (G.P.A: Gim-
bal Pi ch Angle; C.A: Cou se Angle).
The esul s o he esou ce easibili y s udy a e p esen ed in Table 6.
Table 6. Summa y o UAV pho og amme y cap u e da a.
Sec o s Sec o s
O e lap
Fligh
Heigh
G ound
Sample
Dis ance GSD
Nadi
Sho
(−90°)
Oblique
Sho
(−65°)
F on al
O e lap
La e al
O e lap
Fligh
Time Ba e ies Pho os
Sec o ype
1 ha - 37.3 m 1 cm/pixel 1 4 75% 75% 100 min 5 se s 450
Se o sec o s
147 ha
10 m.
(10%) 37.3 m 1 cm/pixel 147 588 75% 75% 245 h 735 se s 66,150
Cap u ing da a om he en i e campus would equi e 245 ligh hou s, which, added
o he p epa a ion wo k, could mean a ield ask o mo e han one mon h. As p e iously
jus i ied, he addi ional wo ks o he ALS LiDAR cloud wi h TLS echniques was es i-
ma ed o ake se en days, so he UAV was used o complemen he TLS da a in hose
sec o s o he 147 planned sec o s whe e he e a e no buildings.
4.4. Modeling
The poin cloud o he ensemble will be used o pe o m he simula ion 3D model o
each campus in NEST (Figu e 19).
Figu e 19. Final 3D model o he UNAV campus in Pamplona o simula ion in NEST, (see Appen-
dix A).
The esul ing inal cloud can also be segmen ed o model isola ed buildings in g ea e
de ail and make speci ic simula ions o he campuses (Figu e 20).
Figu e 19.
Final 3D model o he UNAV campus in Pamplona o simula ion in NEST, (see
Appendix A).
The esul ing inal cloud can also be segmen ed o model isola ed buildings in g ea e
de ail and make speci ic simula ions o he campuses (Figu e 20).
Buildings 2022,12, 380 24 o 32
Buildings 2022, 12, x FOR PEER REVIEW 25 o 33
Figu e 20. BIM model in Au odesk Re i so wa e. School o Enginee ing building. Campus
UPV/EHU DSS, Zone 2 o he scan.
4.5. NEST—En i onmen al Assessmen Resul s
As p e iously s a ed, i is no he pu pose o his a icle o analyze he esul s o he
assessmen , which has al eady been deal wi h in o he speci ic a icles [11,12]. In any case,
o ound up he low o he p ojec , we el i was app op ia e o show a e y educed
sample o he esul s gi en by NEST. An image o he simula ion model and a summa y
able o he wo campuses will be p esen ed (Figu e 21 and Table 7).
Figu e 21. NEST model wi h simula ion o CO2 impac s. UPV/EHU DSS campus.
Table 7. Resul s ob ained in he NEST simula ion o he imp o emen scena ios o he yea s 2030
and 2050.
Impac
Indica o Sec o Li e-Cycle
S age *
Baseline
Scena io 2030 2050
UNAV UPV/EHU UNAV UPV/EHU UNAV UPV/EHU
PE
(MJ/yea )
Buildings (BP) A1–3, A4–5, B2,
B4, C1–4 7.5 × 106 1.0 × 107 1.7 × 107 1.5 × 107 3.4 × 107 1.7 × 107
Buildings (BU) B6 1.3 × 108 3.8 × 107 1.1 × 108 3.7 × 107 9.3 × 107 3.5 × 107
Public ligh ing (PL) B6 2.2 × 106 3.1 × 106 1.7 × 106 2.5 × 106 1.1 × 106 1.6 × 106
Mobili y A1–3, B6, C1–4 1.8 × 103 3.1 × 103 1.8 × 103 3.1 × 103 1.8 × 103 3.1 × 103
GWP
(kgeqCO2/yea )
Buildings (BP) A1–3, A4–5, B2,
B4, C1–4 3.5 × 105 4.1 × 105 8.1 × 105 6.3 × 105 9.7 × 105 7.2 × 105
Buildings (BU) B6 6.3 × 106 1.9 × 106 5.0 × 106 1.8 × 106 4.4 × 106 1.7 × 106
Public ligh ing (PL) B6 2.3 × 105 3.3 × 105 1.9 × 105 2.7 × 105 1.2 × 105 1.7 × 105
Mobili y A1–3, B6, C1–4 1.0 × 102 1.4 × 102 1.0 × 102 1.4 × 102 1.0 × 102 1.4 × 102
Ene gy
consump ion
(kWh/yea )
Na u al gas (NG) B6 1.7 × 107 1.5 × 106 9.4 × 106 1.2 × 106 5.7 × 106 6.9 × 105
Elec ici y (E) B6 9.7 × 106 5.9 × 106 1.0 × 107 5.7 × 106 1.1 × 107 5.5 × 106
Biomass (B) B6 7.0 × 103 0.0 6.3 × 105 4.5 × 104 1.0 × 106 9.8 × 104
Renewable en-
e gy
p oduc ion
(kWh)
The mal sola (TS) 3.6 × 104 7.1 × 104 2.4 × 105 1.9 × 105 6.6 × 105 4.7 × 105
Pho o ol aic (P) 0.0 4.7 × 105 1.8 × 106 7.8 × 105 3.7 × 106 1.6 × 106
* See Table 2.
Figu e 20.
BIM model in Au odesk Re i so wa e. School o Enginee ing building. Campus
UPV/EHU DSS, Zone 2 o he scan.
4.5. NEST—En i onmen al Assessmen Resul s
As p e iously s a ed, i is no he pu pose o his a icle o analyze he esul s o he
assessmen , which has al eady been deal wi h in o he speci ic a icles [
11
,
12
]. In any case,
o ound up he low o he p ojec , we el i was app op ia e o show a e y educed
sample o he esul s gi en by NEST. An image o he simula ion model and a summa y
able o he wo campuses will be p esen ed (Figu e 21 and Table 7).
Buildings 2022, 12, x FOR PEER REVIEW 25 o 33
Figu e 20. BIM model in Au odesk Re i so wa e. School o Enginee ing building. Campus
UPV/EHU DSS, Zone 2 o he scan.
4.5. NEST—En i onmen al Assessmen Resul s
As p e iously s a ed, i is no he pu pose o his a icle o analyze he esul s o he
assessmen , which has al eady been deal wi h in o he speci ic a icles [11,12]. In any case,
o ound up he low o he p ojec , we el i was app op ia e o show a e y educed
sample o he esul s gi en by NEST. An image o he simula ion model and a summa y
able o he wo campuses will be p esen ed (Figu e 21 and Table 7).
Figu e 21. NEST model wi h simula ion o CO2 impac s. UPV/EHU DSS campus.
Table 7. Resul s ob ained in he NEST simula ion o he imp o emen scena ios o he yea s 2030
and 2050.
Impac
Indica o Sec o Li e-Cycle
S age *
Baseline
Scena io 2030 2050
UNAV UPV/EHU UNAV UPV/EHU UNAV UPV/EHU
PE
(MJ/yea )
Buildings (BP) A1–3, A4–5, B2,
B4, C1–4 7.5 × 106 1.0 × 107 1.7 × 107 1.5 × 107 3.4 × 107 1.7 × 107
Buildings (BU) B6 1.3 × 108 3.8 × 107 1.1 × 108 3.7 × 107 9.3 × 107 3.5 × 107
Public ligh ing (PL) B6 2.2 × 106 3.1 × 106 1.7 × 106 2.5 × 106 1.1 × 106 1.6 × 106
Mobili y A1–3, B6, C1–4 1.8 × 103 3.1 × 103 1.8 × 103 3.1 × 103 1.8 × 103 3.1 × 103
GWP
(kgeqCO2/yea )
Buildings (BP) A1–3, A4–5, B2,
B4, C1–4 3.5 × 105 4.1 × 105 8.1 × 105 6.3 × 105 9.7 × 105 7.2 × 105
Buildings (BU) B6 6.3 × 106 1.9 × 106 5.0 × 106 1.8 × 106 4.4 × 106 1.7 × 106
Public ligh ing (PL) B6 2.3 × 105 3.3 × 105 1.9 × 105 2.7 × 105 1.2 × 105 1.7 × 105
Mobili y A1–3, B6, C1–4 1.0 × 102 1.4 × 102 1.0 × 102 1.4 × 102 1.0 × 102 1.4 × 102
Ene gy
consump ion
(kWh/yea )
Na u al gas (NG) B6 1.7 × 107 1.5 × 106 9.4 × 106 1.2 × 106 5.7 × 106 6.9 × 105
Elec ici y (E) B6 9.7 × 106 5.9 × 106 1.0 × 107 5.7 × 106 1.1 × 107 5.5 × 106
Biomass (B) B6 7.0 × 103 0.0 6.3 × 105 4.5 × 104 1.0 × 106 9.8 × 104
Renewable en-
e gy
p oduc ion
(kWh)
The mal sola (TS) 3.6 × 104 7.1 × 104 2.4 × 105 1.9 × 105 6.6 × 105 4.7 × 105
Pho o ol aic (P) 0.0 4.7 × 105 1.8 × 106 7.8 × 105 3.7 × 106 1.6 × 106
* See Table 2.
Figu e 21. NEST model wi h simula ion o CO2impac s. UPV/EHU DSS campus.
Table 7.
Resul s ob ained in he NEST simula ion o he imp o emen scena ios o he yea s 2030
and 2050.
Impac
Indica o Sec o Li e-Cycle
S age *
Baseline
Scena io 2030 2050
UNAV UPV/EHU UNAV UPV/EHU UNAV UPV/EHU
PE
(MJ/yea )
Buildings (BP) A1–3, A4–5, B2,
B4, C1–4
7.5
×
10
61.0 ×107
1.7
×
10
71.5 ×107
3.4
×
10
71.7 ×107
Buildings (BU) B6
1.3
×
10
83.8 ×107
1.1
×
10
83.7 ×107
9.3
×
10
73.5 ×107
Public ligh ing (PL) B6
2.2
×
10
63.1 ×106
1.7
×
10
62.5 ×106
1.1
×
10
61.6 ×106
Mobili y A1–3, B6, C1–4
1.8
×
10
33.1 ×103
1.8
×
10
33.1 ×103
1.8
×
10
33.1 ×103
GWP
(kgeqCO2/yea )
Buildings (BP) A1–3, A4–5, B2,
B4, C1–4
3.5
×
10
54.1 ×105
8.1
×
10
56.3 ×105
9.7
×
10
57.2 ×105
Buildings (BU) B6
6.3
×
10
61.9 ×106
5.0
×
10
61.8 ×106
4.4
×
10
61.7 ×106
Public ligh ing (PL) B6
2.3
×
10
53.3 ×105
1.9
×
10
52.7 ×105
1.2
×
10
51.7 ×105
Mobili y A1–3, B6, C1–4
1.0
×
10
21.4 ×102
1.0
×
10
21.4 ×102
1.0
×
10
21.4 ×102
Ene gy
consump ion
(kWh/yea )
Na u al gas (NG) B6
1.7
×
10
71.5 ×106
9.4
×
10
61.2 ×106
5.7
×
10
66.9 ×105
Elec ici y (E) B6
9.7
×
10
65.9 ×106
1.0
×
10
75.7 ×106
1.1
×
10
75.5 ×106
Biomass (B) B6
7.0
×
10
30.0
6.3
×
10
54.5 ×104
1.0
×
10
69.8 ×104
Renewable ene gy
p oduc ion (kWh)
The mal sola (TS)
3.6
×
10
47.1 ×104
2.4
×
10
51.9 ×105
6.6
×
10
54.7 ×105
Pho o ol aic (P) 0.0 4.7 ×105
1.8
×
10
67.8 ×105
3.7
×
10
61.6 ×106
* See Table 2.
Buildings 2022,12, 380 25 o 32
5. Discussion
The s a ing hypo hesis ocused on he possibili ies o ob aining apid digi iza ion o
wo uni e si y campuses ha occupy a la ge a ea o he ci ies whe e hey a e loca ed. The
lockdown si ua ion caused by COVID-19 did no allow long s ays in he ield o collec
da a, and a combina ion o echniques we e s udied so ha he su ey could be pe o med
in jus a ew days.
A e analyzing he esul s, i can be s a ed ha i was possible o cap u e bo h uni-
e si y campuses in 3D, while ield wo k ime was educed o jus 10 days (7 + 3), wi h a
sa is ac o y ou come ha was economical and e icien . I was possible o comple e he
ieldwo k in en days, hanks o a combina ion o esou ces and echniques. A e s udying
di e en echnologies and esou ces, he wo k was accomplished by using ALS LiDAR
poin clouds cap u ed by public se ices and by cap u ing complemen a y poin clouds
wi h g ea e densi y and accu acy by means o TLS; inally, UAV-assis ed au oma ed pho-
og amme ic echniques we e used o complemen p e ious clouds. The s udy o each o
hese echnologies used o achie e he objec i es o his wo k enabled discussion o se e al
speci ic ad an ages and disad an ages o he speci ic si ua ion a ising in his a icle.
The ALS LiDAR clouds ob ained by public se ices a e ee and hus did no in ol e
any in es men o esou ces, inances o ime. Al hough, i was disce ned ha hose
LiDAR clouds o he wo ci ies in ques ion ha e e y di e en quali ies, which a ec ed
he app oach o he wo k and i s de elopmen . The LiDAR cloud o he UNAV campus in
Pamplona, wi h a densi y o 14 poin s/m
2
, allows measu emen s o be made wi h su icien
p ecision o do he simula ion model in NEST. Howe e , he LiDAR cloud o he UPV/EHU
campus in he ci y o Donos ia-San Sebas ián, wi h a maximum densi y o 2.2 poin s/m
2
,
has many limi a ions. Rega ding he measu emen o pa s o building elemen s, i would
be limi ed o dimensioning he o al heigh o he buil olume, wi hou being able o make
o he kinds o measu emen s such as heigh o s anda d loo plan, opening sizes and e en,
in some cases, size o he açade su aces. Wi h espec o plan masses, i is also e i ied
ha he cloud’s low densi y does no allow es ima es o mass olume and in some cases,
no e en basic measu emen s o he plan elemen ’s heigh . Tha is why i was decided o
complemen he LiDAR da a wi h massi e poin cap u e based on TLS, since ha echnique
allows cap u e in he sho es possible ime. Conside ing he cha ac e is ics o he LiDAR
clouds ound, he numbe o scanning poin s o each campus was mo e o less in ensi ied.
We ha e used his p ocedu e o y ou scan poin s a longe dis ances han usual (30 m,
50 m
and
80 m
, wi hou exceeding he cap u e dis ance limi p esen ed by he lase scanne ,
which is 130 m). I should be bo ne in mind ha his wo k wi h TLS se s ou o complemen
he LiDAR cloud o he Go e nmen o Na a a. Wi h hese es s we ound ha mos scan
poin s a dis ances o 30 m gi e accep able esul s (unions in g een) and ha he maximum
ange o he scanne a high esolu ion is ac ually 60 m, e en when he capaci y is 130 m.
Rega ding he complemen a y wo k o he poin cloud wi h TLS echniques, he
UPV/EHU campus has a highe scan posi ion densi y han Pamplona, since i s 2017
ALS LiDAR cloud is much less p ecise han ha o he UNAV campus. The wo k wi h
TLS showed some di e ences a each campus. Conside ing he maximum ange o he
scanne (130 m), a minimum o h ee scans pe building açade could ini ially be conside ed.
Howe e , due o he cha ac e is ics o each campus, he p e- egis a ion and p ocessing
wo k had di e en disad an ages. A he UPV/EHU campus, when joining he adjacen
clouds, he p e- egis a ion was s ong enough because i was a la a ea wi h high building
densi y and li le ege a ion. Be e cloud s eng h and o e lap esul s we e ob ained, e en
hough he scanne used he same millime ic p ecision. Howe e , a UNAV, he campus’s
size makes he buildings mo e dis an , so when scanning a building i is no easy o eco d
many poin s om ano he adjacen building o s eng hen he join s. In addi ion, he dense
ege a ion and he une enness o he s eep e ain make he o e lap and s eng h o he
inal clouds lowe . In any case, in Pamplona, his aspec was supplied wi h he good quali y
o he 2017 ALS LiDAR clouds.
Buildings 2022,12, 380 32 o 32
65.
ISO 14040; En i onmen al Managemen —Li e Cycle Assessmen –P inciples and F amewo k; In e na ional O ganiza ion o
S anda diza ion: Gene a, Swi ze land, 2018.
66. Su ey O ice, e sion 5.30; Spec a Geospa ial: Wes mins e , CO, USA; T imble: Sunny ale, CA, USA, 2020.
67. Cyclone FIELD 360, Leica Geosys ems: S . Gallen, Swi ze land, 2020.
68. DJI-GS P o, e sion 2.0.15; (iOS 9.2); DJI: Shenzhen, China, 2018.
69. Cyclone REGISTER 360, Leica Geosys ems: S . Gallen, Swi ze land, 2020.
70. Poin Cab, Poin Cab GmbH: We nau, Ge many, 2020.
71. ReCap P o, (Windows); Au odesk: Mill Valley, CA, USA, 2020.
72. T uView, Leica Geosys ems: S . Gallen, Swi ze land, 2020.
73. Je s eam Viewe , Leica Geosys ems: S . Gallen, Swi ze land, 2020.
74.
You, H.; Li, S.; Xu, Y.; He, Z.; Wang, D. T ee Ex ac ion om Ai bo ne Lase Scanning Da a in U ban A eas. Remo e Sens.
2021,13, 3428. [C ossRe ]
75.
Robles, J.L.M.; Sanz, J.O.; Rod íguez, S.M.; Lau naga, F.G.; Sas e, L.F.S.; Na a o, S.H.; Pé ez, Z.C.; Sanz, L.O. De e minación
de biomasa en pa celas de cul i os he báceos median e cáma as óp icas ele adas po medio de ehículos aé eos no ipulados
(UAV). CIAIQ 2016,4, 95–103.
76.
G ego io, M.; Rica do, R.; Peinado, M. P oducción de Biomasa y Fijación de CO
2
po los Bosques Españoles; INIA-Ins i u o Nacional de
In es igación y Tecnología Ag a ia y Alimen a ia: Mad id, Spain, 2005; p. 13.
77. Chou, S.; Espele a, E. Ecuación pa a es ima la biomasa a bó ea en los bosques opicales de Cos a Rica. Tecnol. En Ma cha 2013,
26, 41–54. [C ossRe ]
78.
Sun, Z.; Cao, Y. Da a p ocessing wo k lows om low-cos digi al su ey o a ious applica ions: Th ee case s udies o Chinese
his o ic a chi ec u e. ISPRS In . A ch. Pho og amm. Remo e Sens. Spa . In . Sci. 2015,5, 409–416. [C ossRe ]
79.
Gaiani, M.; Remondino, F.; Apollonio, F.I.; Ballabeni, A. An Ad anced P e-P ocessing Pipeline o Imp o e Au oma ed Pho og am-
me ic Recons uc ions o A chi ec u al Scenes. Remo e Sens. 2016,8, 178. [C ossRe ]
80.
Sun, Z.; Zhang, Y. Using D ones and 3D Modeling o Su ey Tibe an A chi ec u al He i age: A Case S udy wi h he Mul i-Doo
S upa. Sus ainabili y 2018,10, 2259. [C ossRe ]
81.
Xingzhi, W.; Yuchen, W.; Fei, T.; Ang, L. New Pa adigm o Da a-D i en Sma Cus omisa ion h ough Digi al Twin. J. Manu . Sys .
2021,58, 270–280.
82.
Lee, A.; Lee, K.-W.; Kim, K.-H.; Shin, S.-W. A Geospa ial Pla o m o Manage La ge-Scale Indi idual Mobili y o an U ban Digi al
Twin Pla o m. Remo e Sens. 2022,14, 723. [C ossRe ]
83.
Wa ke, V.; Kuma , S.; Bongale, A.; Ko echa, K. Sus ainable De elopmen o Sma Manu ac u ing D i en by he Digi al Twin
F amewo k: A S a is ical Analysis. Sus ainabili y 2021,13, 10139. [C ossRe ]
84.
Demys i ying A i icial In elligence based Digi al Twins in Manu ac u ing-A Bibliome ic Analysis o T ends and Techniques.
A ailable online: h ps://www. esea chga e.ne /p o ile/Sa ish-Kuma -V-C/publica ion/346057367_Demys i ying_A i icial_
In elligence_based_Digi al_Twins_in_Manu ac u ing-A_Bibliome ic_Analysis_o _T ends_and_Techniques/links/5 b9160 2
99b 104c 66d615/Demys i ying-A i icial-In elligence-based-Digi al-Twins-in-Manu ac u ing-A-Bibliome ic-Analysis-o -
T ends-and-Techniques.pd (accessed on 14 Feb ua y 2022).
85.
Iba ola, E.; O egi, X.; Leon, I.; Ma ie a, C. E aluación de los aspec os Ene gé icos y Medioambien ales del campus de Vi o-
ia/Gas eiz. In P oceedings o he RED-U, Bilbao, Spain, 13–14 No embe 2017; pp. 235–245.
