P ocedia Compu e Science 77 ( 2015 ) 33 – 39
A ailable online a www.sciencedi ec .com
1877-0509 © 2015 The Au ho s. Published by Else ie B.V. This is an open access a icle unde he CC BY-NC-ND license
(h p://c ea i ecommons.o g/licenses/by-nc-nd/4.0/).
Pee - e iew unde esponsibili y o he Socio echnical Sys ems Enginee ing Ins i u e o Vidzeme Uni e si y o Applied Sciences
doi: 10.1016/j.p ocs.2015.12.356
ScienceDi ec
ICTE in Regional De elopmen
Augmen ed eali y o aining spa ial skills
Jo ge Ma ín-Gu ié eza
*
, Manuel Con e ob, Ma iano Alcañizb
aDp o. Tecnicas y P oyec os en Ingenie ía y A qui ec u a, Uni e sidad de La Laguna, La Laguna, Spain
bIns i u o en Bioingenie ía y Tecnología O ien ada al Se Humano, Uni e sidad Poli écnica de Valencia,Valencia, Spain
Abs ac
La Laguna Uni e si y has been o e ing cou ses o he de elopmen o spa ial skills since 2004. Each yea since ha ime spa ial
abili y o enginee ing s uden s has been measu ed be o e and a e he cou ses o check p og ess a e each aining session. We
ha e de eloped a spa ial skills aining cou se based on augmen ed eali y and g aphic enginee ing con en s, and designed he
AR_Dehaes ool, which is based on i s own lib a y he uses compu e ision echniques o inco po a ing isual objec s h ough
iducial ma ke s in he eal wo ld.
© 2016 The Au ho s. Published by Else ie B.V.
Pee - e iew unde esponsibili y o he Socio echnical Sys ems Enginee ing Ins i u e o Vidzeme Uni e si y o Applied Sciences.
Keywo ds: Spa ial abili y; Augmen ed Reali y; Enginee ing Educa ion; G aphic Enginee ing
1. In oduc ion
Fe guson1 in “Enginee ing and he Mind´s Eye” indica es ha usually an in ui i e connec ion is es ablished
be ween spa ial skills and enginee ing. Besides, mos pape s abou spa ial skills o hei componen s e e o he ac
ha enginee ing, a chi ec u e and mos scien i ic jobs need good le els o spa ial abili y. Spa ial abili y has an
impac on e e y scien i ic and echnical ield so i ’s s ill enjoying s ong de elopmen when i comes o enginee ing,
echnology, a and many o he aspec s o li e e en conside ing i has no been paid as much a en ion as o he
in elligence componen s such as e bal and nume ic abili ies.
Many eache s and esea che s like P. Connolly2 sugges he need o de eloping spa ial abili y om G aphic
Enginee ing subjec s. Ba 3 analysed u u e academic enginee ing plans keeping in mind mode n endencies and
* Co esponding au ho .
E-mail add ess: [email p o ec ed]; mcon e [email protected]; m[email p o ec ed]
© 2015 The Au ho s. Published by Else ie B.V. This is an open access a icle unde he CC BY-NC-ND license
(h p://c ea i ecommons.o g/licenses/by-nc-nd/4.0/).
Pee - e iew unde esponsibili y o he Socio echnical Sys ems Enginee ing Ins i u e o Vidzeme Uni e si y o Applied Sciences
34 Jo ge Ma ín-Gu ié ez e al. / P ocedia Compu e Science 77 ( 2015 ) 33 – 39
highligh ed ha he mos impo an subjec included in he p og am should be de elopmen o spa ial skills.
His o ically, he e has been a g ea deal o in e es in me hods o ins uc ion and echnology ha could po en ially
inc ease he spa ial skills o i s use s,4, 5, 6 Cu en ly, he ise in i ual eali y, be i imme si e, augmen ed, o
desk op, has uelled enewed esea ch in spa ial abili y de elopmen .
O e he las yea s we ha e c ea ed nume ous expe iences in as emedial cou ses o imp o e spa ial abili ies o
enginee ing s uden s a he Uni e si y o La Laguna in Spain. In hese cou ses di e en ools ha e been ied ou :
classic exe cises ( iews) using pen and pape , online mul imedia web-based exe cises, ske ch-based modelling
h ough a callig aphic in e ace,7 use o he Google Ske ch-up modelling applica ion8 and ideogames as a wo k
ool9.
Augmen ed Reali y (AR) echnology eme ges as a nice esou ce o he younge popula ion. This is accomplished
o he la es de elopmen s in echnology and en e ainmen . The cogni i e aspec iden i ies any imp o emen o he
subjec ’s in e nal concep s h ough his lea ning expe ience. In he ollowing sec ion, we desc ibe he con en o a
didac ic oolki (AR_Dehaes) ha p o ides he s uden s wi h a se o a ied exe cises o spa ial abili y aining using
an augmen ed book.
2. Desc ip ion applica ion o AR
Ou applica ion equi es accu a e posi ion and o ien a ion acking in o de o egis e i ual elemen s in he eal
wo ld and so we ha e used a ma ke -based me hod. Fo c ea ing he Magicbook only i een di e en ma ke s ha e
been used. Wi h hese ma ke s all one hund ed exe cises comp ised in he aining can be codi ied. Ma ke s ha e
been combined in pai s so e e y wo o hem a e associa ed o a 3D model and any wo ma ke s in di e en
posi ions can be associa ed o ano he i ual model (see igu e 1.).
Fig. 1. Combina ion wo ma ke s may be associa ed wi h se e al i ual objec s. Abo e a e ou combina ions.
To cap u e he ideo, we ha e used a USB came a (QuickCam P o 9000). This came a has he ollowing ea u es:
cap u ed image size - 1600 x1200 a 30 ps; ocal leng h - 3.7 mm; au oma ic ocus adjus men and Righ Ligh
( echnology wi h he capabili y o adjus he webcam o ake good sho s e en in dim ligh se ings).
Al hough he e a e se e al public lib a ies wi h AR capabili ies, we ha e c ea ed a so wa e lib a y called
HUMANAR in o de o ensu e in eg a ion o Augmen ed Reali y in o ou applica ions and o e come some
d awbacks p esen in some public lib a ies (AR oolKi , MRX oolKi e c.). This lib a y uses compu e ision
echniques o calcula ing he eal came a iewpoin ela i e o a eal wo ld ma ke , which calcula es in eg a ion o
h ee-dimensional objec s codi ied by he came a and cap u ed by i sel in eal ime10.
The g aphical engine used o he augmen a ion o a i ual objec on he scene is B ains o m eS udio11. This is an
ad anced, mul ipla o m, eal ime 3D g aphics p esen a ion ool. In o de o include Augmen ed Reali y
unc ionali y in B ains o m eS udio, a plugin has been de eloped in C++. This plugin inco po a es de ec ion and
acking op ions om he HUMANAR lib a y.
The AR-Dehaes oolki is comp ised o a so wa e applica ion, an explana o y ideo, a no ebook and an
augmen ed book (see igu e 2).
35
Jo ge Ma ín-Gu ié ez e al. / P ocedia Compu e Science 77 ( 2015 ) 33 – 39
x A so wa e applica ion ha con ains he h ee dimensional i ual models. These models will be isualized on he
compu e sc een when iducial ma ke s a e ecei ed by he webcam.
x A sho explana o y ideo ha explains he heo e ical con en s o o hog aphic iews and eehand ske ching.
x A no ebook ha con ains ques ions and exe cises o be sol ed by he s uden s.
x An augmen ed book ha p o ides iducial ma ke s o i ual h ee dimensional objec s.
The s uden s can u n, mo e o b ing he no ebook o he webcam being able o see di e en pe spec i es o he
i ual model and complemen a y in o ma ion o exe cise esolu ion.
Fig. 2. Explana o y ideo, no ebook, augmen ed book and p og am unning
Didac ic ma e ial was c ea ed using Bloom axonomy12 being s uc u ed on i e le els, each one con aining
se e al kinds o exe cises. All le els a e designed o wo hou sessions, excep le el 5 (e alua ion), in which six
exe cises mus be comple ed by s uden s in jus one hou wi hou any i ual model help.
The i s exe cise o each le el and ypology has a physical ges u e ela ed o i , which mus be pe o med by he
s uden in o de o igu e ou i s solu ion and unde s and how o sol e i :
x Le el 1. In his le el, he s uden s ha e o iden i y su aces and e exes in bo h o hog aphic and axonome ic
iews o a h ee-dimensional i ual objec , which is c ea ed on he augmen ed book.
x Le el 2. In his le el, he s uden s ha e o iden i y o hog aphic iews o he i ual h ee dimensional models
om he exe cise book.
x Le el 3 is dedica ed o he iden i ica ion o spa ial ela ionships be ween objec s. This is ca ied ou by means o
“ ecoun ” exe cises, whe e s uden s a e asked o iden i y how many objec s ouching he one selec ed. The e a e
also exe cises abou he selec ion o he minimum numbe o iews o comple e de ini ion o an objec .
x Le el 4 has g ea e di icul y han he p e ious ones. The e a e exe cises whe e he s uden s ha e o ske ch a
missing o hog aphic iew, knowing wo o hog aphic iews o a model. In a second s age, using he i ual
model as he only inpu , hey ha e o ske ch all o hog aphic iews.
x Le el 5 exe cises a e he mos di icul ones o s uden s, because hey equi e a g ea e le el o spa ial abili y.
S uden s a e p o ided wi h h ee o hog aphic iews o each objec , and hey ha e o build he ma ching h ee-
dimensional model in hei minds and hen d aw a eehand pe spec i e o i .
S uden s can isualize he h ee-dimensional model in augmen ed eali y (augmen ed book) and hey can check i
hei eehand ske ches ma ch he h ee-dimensional i ual models hey a e iewing. Comple ion o each le el on
consecu i e days is sugges ed so i should be inished in 5 days o sessions.
36 Jo ge Ma ín-Gu ié ez e al. / P ocedia Compu e Science 77 ( 2015 ) 33 – 39
Table 1: Le els and kinds o exe cises
LEVEL 1
Task 1.1.
Iden i ica ion
o su aces.
O hog aphic
iews.
Task 1.2.Iden i ica ion o su aces.
Axonome ic p ojec ion.
Task
1.3.Iden i ica io
n o e exes.
LEVEL 2
Task 2.1.Iden i ica ion o
o hog aphic iews.
Task 2.2.Iden i ica ion o w ong
o hog aphic iews.
LEVEL 3
Task 3.1. Recoun
Task 3.2. Selec ion o he minimum
numbe o o hog aphic iews o
de ine an objec .
LEVEL 4
Task 4.1. “Missing iew”
exe cises.
Task 4.2. O hog aphic iew
ske ching.
LEVEL 5
37
Jo ge Ma ín-Gu ié ez e al. / P ocedia Compu e Science 77 ( 2015 ) 33 – 39
Task 5.1. Pe spec i e ske ching.
3. S udy and Resul
To check i aining has managed o imp o e he spa ial abili y o s uden s, we in i ed 24 i s yea mechanical
enginee ing s uden s and he cou se ook place on he i s week o he academic yea o a oid ha hey could ob ain
con en s om o he subjec s. We also conside ed a con ol g oup consis ing o 25 s uden s ( i s yea ) in ag icul u al
enginee ing who unde wen no kind o spa ial skills aining. A he beginning and end o he cou se s uden s ha e
pe o med wo es s o measu ing spa ial skills (Table 2). Those es s measu e di e en aspec s o spa ial abili y:
spa ial ela ions and isualisa ion h ough Di e en ial Ap i ude Tes (DAT-5:SR Le el 2)13 and Men al Ro a ion
Tes (MRT)14.
Table 2: Mean alues p e/pos es and gain sco es
P e- es
Pos - Tes
Gain
MRT
Gain
DAT
5:SR
MRT
DAT
5:SR
MRT
DAT
5:SR
Exp.G oup
n = 24
19.67
(7.91)
29.17
(7.29)
27.71
(7.83)
38,46
(7.05)
8.04
(5.31)
9.29
(4.08)
Con ol G oup n=25
17.44
(9.82)
28.40
(10.17)
22.08
(9.94)
33.52
(11.77)
4.64
(4.36)
5.12
(7.13)
An analysis o co a iance (ANCOVA) was used o s udy i . The ANCOVA me hod allows elimina ing he
di e ence o p e es sco es be ween g oups, and hen he adjus ed pos es sco es, e ealing he eal e ec s o he
expe imen al ea men . Hence, i elimina es he possible memo y e ec . This s a is ical p ocedu e also es ed he
in e ac ion be ween g oups (T aining AR and con ol). The dependen a iables, co- a ian s and independen
a iables we e pos - es measu emen s, p e- es measu emen s and ype o g oup espec i ely. Sui abili y o using
his analysis was es ed by i s conduc ing he analysis using a s a is ical model con aining in e ac ion e ms
be ween he co- a ian s (p e- es mean sco es o MRT and DAT-5:SR) and he independen a iables o assess he
assump ion o homogenei y o he g adien s. A e adjus ing o co a ia es (p e- es sco es), i is checked again ha
he e was a signi ican di e ence be ween con ol g oup and aining AR g oup on he gain sco es, F1,46=7.47,
p=0.009 (MRT) and F1,46=6.55, p=0.014 (DAT-5:SR).
We also conduc ed a ques ionnai e on usabili y and sa is ac ion wi h he applica ion used. Ques ions a e di ided
in o i e blocks: educa ional ma e ial e ec i eness, con en s’ e iciency, echnology e iciency, sa is ac ion and
opinion. Ha ing in mind educa ional ma e ial e ec i eness, di e en alues a e assigned acco ding o Like ’s scale
o he ollowing aspec s: cou se ma e ial p esen a ion quali y, com o able cou se s uc u e, he cla i y and
comp ehensibili y o o ma i e ideos, adequa e cou se de elopmen (le els and ypology), easy loca ion o
exe cises be ween design’s no ebook and iducially ma ks’ book, p ope size o no ebooks o comple ing exe cises
and manipula ing i ual elemen s, so wa e s abili y using augmen ed eali y applica ion. All alues ob ained a e
qui e high and nea 5 which show ha he ma e ial is pe ec ly sui able o achie ing he pu pose i was c ea ed o .
Assessing con en lea nabili y (e alua ion) ea ned a medium sco e o 5.7 o e 6 s a ing he cou se is a success.
When measu ing e iciency in he su ey we asked abou con en e iciency and echnology. Bo h ecei ed high
sco es ob ained as well. Pa icipan s s a e ha hey we e capable o esol ing all p oposed exe cises in he allo ed

38 Jo ge Ma ín-Gu ié ez e al. / P ocedia Compu e Science 77 ( 2015 ) 33 – 39
ime and also conside ha manipula ion o i ual objec s and pe o mance o ges u es has been easy as AR is e y
in ui i e.
Table 3: Usabili y ques ionnai e. Blocks o ques ions.
Block
Ques ions
Measu e kind
Wha ’s being
measu ed?
S uden da a
Se e al
Se e al
Pe sonal da a
A.- Didac ic
ma e ial alue
A1-A7
Like scale
Didac ic ma e ial
e ec i eness
B.- Cou se
ma e ial alue
B1-B3
B4
Like scale
Like scale
E iciency
E ec i eness
B5
Like scale
Lea nabili y (P ope
lea ning pe o mance
measu emen )
C.- Used
echnology alue
(AR)
C1-C4
Like scale
E iciency
D.- Cou se opinion
D1-D13
simple-dicho omy
Sa is ac ion
Mul iple choices
Seman ic
Di e en ial
E.- Cou se
imp o emen
sugges ions
E1-E8
simple-dicho omy
Opinion and
sa is ac ion
Mul iple choices
Seman ic
Di e en ial
open-ended
ques ion
4. Conclusion
We ha e p esen ed an AR applica ion o he de elopmen o spa ial skills. As a as we know his is he i s AR
applica ion wi h hese cha ac e is ics ha has been c ea ed o imp o ing spa ial abili y using didac ics con en o
enginee ing g aphics
A co a iance analysis (ANCOVA) shows ha he expe imen al g oup signi ican ly imp o ed i s spa ial skills
a e pe o ming his aining compa ed o he con ol g oup ha had no unde gone any spa ial skills aining.
In educa ional applica ions, i is qui e impo an o ocus he a en ion o s uden s on he ask a hand and educe
he cogni i e o e head needed o applica ion use. This mo i a ed us o design a use - iendly sys em in an ag eeable
en i onmen . AR-Dehaes has p o en o be an e icien ma e ial o de eloping spa ial abili ies and o lea ning
enginee ing g aphics con en s.
Augmen ed eali y is a cos -e ec i e echnology o p o iding s uden s wi h a ac i e con en s in espec o
pape books, gi ing new li e o classical pen and pape exe cises.
39
Jo ge Ma ín-Gu ié ez e al. / P ocedia Compu e Science 77 ( 2015 ) 33 – 39
Acknowledgmen s
The Spanish Minis y o Inno a ion and Science, h ough he “No o ien ed Fundamen al In es iga ion P ojec s”
(“Imp o emen o spa ial easoning and isualiza ion h ough echnologically de eloped ools”, p ojec e .
TIN2010-21296-C02-02) pa ially suppo ed his wo k.
Re e ences
1. Fe guson, E. (1992). Enginee ing and he Mind's Eye. Camb idge, Massachuse s: MIT P ess.
2. Connolly, P., & Maiche , K., 2005. The De eloping and Tes ing o an In e ac i e Web-based Tu o ial o O og aphic D awing Ins uc ion
and Visualiza ion Enhancemen . In: he 2005 ASEE Annual con e ence & Exposi ion. Po land (O egon), June 12-15,2005.
3. Ba , R. 1999. Planning he EDG Cu iculum o he 21s Cen u y: A P oposed Team E o . In: Enginee ing Design Jou nal, 63 (1999) 2,
pp. 4-12.
4. So by, S., Wysocki, A., & Baa mans, B., 2003. In oduc ion o 3D Spa ial Visualiza ion: an ac i e app oach. Cli on Pa k, NY: Thomson
Delma Lea ning.
5. Ra i, A., Samsudin, K., & Ismail, A., 2006. On imp o ing spa ial abili y h ough compu e -media ed Enginee ing D awing ins uc ion. In:
Educa ional Technology & Socie y, 9 (2006) 3, pp. 149-159.
6. So by, S., 2009. Educa ional esea ch in de eloping 3-D spa ial skills o enginee ing s uden s. In: In e na ional Jou nal o Science
Educa ion, 31 (2009) 3, pp. 459-480.
7. Con e o, M., Naya, F., Company, P., Sao in, J., & Conesa, J., 2005. Imp o ing Visualiza ion Skills in Enginee ing Educa ion. In: IEEE
Compu e G aphics and Applica ion, 25 (2005) 5, pp. 24-31.
8. Ma ín-Do a, N., Sao ín, J., & Con e o, M., 2008. De elopmen o a Fas Remedial Cou se o Imp o e he Spa ial Abili ies o Enginee ing
S uden s. In: Jou nal o Enginee ing Educa ion, 97(2008) 4, pp. 505-513.
9. Ma ín- Gu ié ez, J., Sao ín, J. L., Ma ín-Do a, N., & Con e o, M., 2009. Do Video Games Imp o e Spa ial Abili ies o Enginee ing
S uden s?. In: In e na ional Jou nal o Enginee ing Educa ion, 25(2009) 6, pp. 1194-1204.
10. Ma ín-Gu ié ez, J. e al., 2010. Design and Valida ion o an Augmen ed Reali y o Spa ial Abili ies De elopmen in Enginee ing S uden s.
In: Compu e & G aphics, 34(2010) 1, pp. 77-91.
11. B ains o m eS udio websi e. Web si e : h p://www.b ains o m.es/pages/es udio.h ml
12. Bloom, B. S. (1956). Taxonomy o Educa ional Objec i es, Handbook I: The Cogni i e Domain. New Yo k: Da id McKay Co Inc.
13. Benne , G., Seasho e, H., & Wesman. (2007). A Di e en ial Ap i ude Tes s. The Psychological Co po a ion. En Spanish o icial e sion:
Handbook o he DAT-5. New Yo k (1947): TEA Ediciones, S.A.
14. Vandenbe g, S., & Kuse, A., 1978. Men al Ro a ions: A G oup Tes o Th ee-Dimensional Spa ial Visualisa ion. In: Pe cep ual and Mo o
Skills, 47(1978) 6, pp. 599-604.
Jo ge Ma ín-Gu ié ez is a P o esso o Technical O ice and Enginee ing G aphics and
CAD a he Uni e sidad de La Laguna, Spain (ULL). He ob ained his Ph.D. in 2010 om
Uni e sidad Poli écnica de Valencia. He wo ked om 1996, as a p ojec enginee ci il. His
esea ch in e es s ocus on echnology enhanced lea ning, de elopmen o spa ial abili ies
using mul imedia echnologies, and augmen ed eali y.
Manuel Con e o is a Full P o esso o Enginee ing G aphics and CAD wi h he G aphic
Enginee ing Depa men a he Uni e sidad Poli écnica de Valencia, Spain (UPV). He ea ned
PhD in Indus ial Enginee ing in 1995 om UPV. His esea ch in e es s ocus on ske ch-
based modeling, collabo a i e enginee ing, human compu e in e ac ion, de elopmen o
spa ial abili ies and, echnology enhanced lea ning.
Ma iano Alcañiz is a Full P o esso o Enginee ing G aphics and CAD wi h he G aphic
Enginee ing Depa men a he Uni e sidad Poli écnica de Valencia, Spain (UPV). He
pe o ms his esea ch in he Labo a o y o Human-Cen e ed Technology ela ed wi h i ual
he apy, i ual ehabili a ion, pe suasi e echnologies and i ual eali y.