International Journal of Cosmetic Science, 2015, 37, 56–62 doi: 10.1111/ics.

12168

Analysing the movement of a hair swatch using video and image

analysis: A promising technique for exploring the dynamic
properties of hair

A. Galliano*, M. Lheur* and R. Santoprete†

*L’Or
eal Research and Innovation, Centre Charles Zviak, Saint-Ouen, France and †L’Or
eal Research and Innovation, Aulnay-sous-bois, France

Received 25 July 2014, Accepted 22 September 2014

Keywords: computer modelling, hair movement, hair treatment

Synopsis
OBJECTIVE: The aim of this paper is to present a new instru-
mental evaluation method of hair movement in order to propose
new criteria in the performances of hair products. The in vitro
evaluation of hair swatches movement allows us to better under-

echantillons de cheveux permet de mieux comprendre les
stand the phenomena that appear when various hair care and
styling technologies are applied upon the mechanisms of hair
movements.
METHODS: The method consists of a video acquisition system
with fast cameras (100 frames/second). The swatch is placed on a
motor drive and moved with a back and forth movement. The fre-
quency and amplitude are set by the operator. Facing each camera,
a backlight provides high contrast to capture all the fibres of the
swatch. A software was specifically developed to analyse the video
recordings and provide the physical parameters that characterize
the movement. Accordingly, it is possible to follow the surface and
shape of the swatch during movement from the ‘skeleton’ of the
swatch and to study the contraction and expansion phases of the
swatch(es) during the movement together with ‘the envelope’ of
the swatch.
RESULTS: Results show that applying various hair treatments
modifies the behaviour of the swatch through the hair shapes con-
traction or expansion induced by the oscillating movement. Expan-
sion of the swatch suggests that the interactions between hair
fibres (adhesions or frictions) play an important role, leading to an
‘individualization’ of hair swatch (more ‘space’ between fibres) or
to a ‘wider’ movement related to the increase of the total surface
developed by the swatch. As an example, the application of a
highly conditioning formula on a hair swatch largely decreases its
expansion by a two-fold factor during movements and increases
the horizontal amplitude by a two-fold factor, as compared to a
commercial bland shampoo.
CONCLUSION: In conclusion, this method is complementary to
the ones used generally to characterize global visual perceptions
(colour, styling, shine). This dynamic component not only opens
doors to new qualitative evaluations, linked with sensorial experts
and consumers perceptions, but also to new quantitative parame-
ters.
Correspondence: Anthony Galliano, L’Or
eal Research and Innovation,
Centre Charles Zviak, 11-13 rue Dora Maar, 93400 Saint-Ouen,
France. Tel.: +33 (0)1 49 18 02 51; e-mail: agalliano@rd.loreal.com
Re
sume

OBJECTIF: Le but de cet article est de pr
esenter une nouvelle
m
ethode d’
evaluation instrumentale de mouvement de cheveux
afin de proposer de nouveaux criteres dans les performances de
produits capillaires. L’
evaluation in vitro du mouvement des
ph
enomenes qui apparaissent lorsque diff
erentes technologies de so-
ins capillaires et coiffage sont appliqu
ees sur les m
ecanismes de
mouvements de cheveux.


METHODES: Le proc
ed
e consiste en un systeme d’acquisition par
vid
eo-cam
eras rapides (100 images/seconde). L’
echantillon est
plac
e sur un moteur d’entra^ınement et d
eplac
e avec un mouve-
ment avant – arriere. La fr
equence et l’amplitude sont d
etermin
ees
par l’op
erateur. Face a  chaque cam
era, un r
etro-
eclairage offre un
contraste
elev
e pour capturer toutes les fibres de l’
echantillon.
L’Asoftware a
et
e sp
ecifiquement d
evelopp
ee pour analyser les en-
registrements vid
eo et fournir les parametres physiques qui ca-
ract
erisent le mouvement. En cons
equence, il est possible de suivre
la surface et la forme de l’
echantillon pendant le mouvement de la
«squelette» de l’
echantillon et d’
etudier les phases de contraction et
d’expansion de l’
echantillon (s) pendant le mouvement avec “l’en-
veloppe” de l’
echantillon.
RESULTATS: Les r
esultats montrent que l’application de divers
traitements capillaires modifie le comportement de l’
echantillon
a travers la contraction ou l’expansion induite par le mouve-
ment d’oscillation des cheveux. L’expansion de l’
echantillon
suggere que les interactions entre les fibres capillaires
(adh
erences ou frictions) jouent un r^ole important, ce qui con-
duit a  une “individualisation” des cheveux (plus «d’espace»
entre les fibres) ou a  un mouvement «plus large» li
ee a  l’aug-
mentation de la surface totale d
evelopp
ee par l’
echantillon. A
titre d’exemple, l’application d’une formule de conditionnement
intense sur un
echantillon de cheveux diminue en grande par-
tie son expansion par un facteur 2 lors des mouvements et
augmente l’amplitude horizontale par un facteur 2 par rapport
a un shampooing doux commercial.
CONCLUSION: En conclusion, cette m
ethode est compl
ementaire
a celles utilis
ees g
en
eralement pour caract
eriser les perceptions
visuelles globales (couleur, style,
eclat). Cette composante dyna-
mique non seulement ouvre les portes a  de nouvelles
evaluations
qualitatives, li
ees avec des experts et des consommateurs sensori-
elles perceptions, mais aussi a  de nouveaux parametres quantita-
tifs.

56 © 2014 Society of Cosmetic Scientists and the Soci
et
e Francßaise de Cosm
etologie

Hair movement evaluation
Introduction
Once called a ‘naked ape’ in a famous book [1], Homo sapiens is
glabrous on most his body at the exception of his scalp from where
100 000–150 000 hairs emerge. The latter constitute a head of
hair, with a wide mosaic of shapes, lengths, colours, sizes, making
two head-hairs rarely looking alike. Such multifaceted features are
even amplified by personal changes in shape and colour brought,
ad libitum, by cosmetic products dedicated to hair.
As single organic fibre, the human hair is the focus of many sci-
entists, including being Dermatologists Biologists, Physicists, Chem-
ists, Cosmeticians. All continuously devote their skills to decipher
its properties, composition, growth, loss, shape that are summarized
in famous textbooks [2, 3].
Imparting style, bounce, style retention to head of hair is a
major daily challenge and expertise of hair professionals. As a
matter of fact, most studies deal with the characteristics of single
hair (shape, colour, chemical composition, structure etc.). In con-
trast, very few works detail the interactive relationships of hairs,
knowing that a full head-hair comprise 100 000–150 000 hairs of
various lengths and their behaviour in real-life conditions [4, 5, 6,
7]. To our knowledge, only two unpublished proprietary testing
methods have been developed [http://www.crodapersonalcare.
com/home.aspx?s=157&r=266&p=3669, http://www.kosmoscience.
net/?q=cabelo ] to describe the dynamical behaviour of hair
swatches during forced movements. It has to be kept in mind
that the human head-hair is daily submitted to movements, that
is gestures, positions, touches, wind, caps or hats, hairdrier con-
ferring it variable appearances along the day. Such aspect is in
addition highly conditioned by hairstyle fashions. Four to five dec-
ades ago, many consumers favoured a fixed hairstyle using very
popular hair lacquers of strong fixative properties. Such a desire
has largely changed and, nowadays, most people favour a ‘free’
head of hair, capable of liberated flowing and living movements.
They probably express a quest for individual hallmarks and/or
freedom vis a vis aesthetic standards, that is a head of hair
reflecting self-esteem combined with the affirmation of intimate
personality [8].
The present work aimed at: (i) measuring the dynamic and geo-
metric properties of hairs assembled into a swatch, when forced to
controlled movements, through a standardized and automated test,
(ii) to establish relevant parameters of such movements and (iii) as
preliminary results, to address the influence of some cosmetic prod-
ucts upon these shape criteria.
Material and methods
Hair swatches
These were supplied by IHIP Company (International Hair Import-
ers & Products, 70 Westmoreland Avenue, White Plains, NY
10606) assembling natural human hairs in standardized condi-
tions, of a same length (27 cm) and weight (2.7 g  0.1 g). In all
cases, these swatches are of a straight shape, type I, as defined by
a previous work [8]. Swatches are realized by aligning all hairs in
a same direction (root to tip) and gathering them, further fixed by
a plastic collar (used to maintain small electric cables), 1 cm below
the root part.
All swatches were initially washed, using a standard bland
shampoo (laurylether sulphate based), further rinsed in water for
10 seconds and dried for 30 min at 60°C.
© 2014 Society of Cosmetic Scientists and the Soci
et
e Francßaise de Cosm
etologie
International Journal of Cosmetic Science, 37, 56–62
A. Galliano et al.
In view of assessing hair care products (conditioners) as a prior-
ity, all swatches were then bleached, immersed in a standard
bleaching solution (with common raw material as sodium persul-
phate, EDTA, ammonia, hydrogen peroxide) during 50 min at
30°C. Bleaching process was stopped by adding sodium thiosulfate
(50% in water) in a buffered solution (sodium sulphate 50% and
tartaric acid 25% in water). Hair swatches were then thoroughly
rinsed with water, washed with a standard bland shampoo (laury-
lether sulphate based) and dried.
In all cases, before being fixed to an oscillating arm, electrostatic
forces appearing on the swatches were neutralized by a brief con-
tact with an anti-static ‘pistol’ (Zerostat anti-static instrument,
Z108812 Aldrich), used in routine in atomic force microscopy
(AFM) technique to neutralize probes charges.
Automated movements and recording equipment
A pentagon-shaped enclosure (Fig. 1) of internal black walls com-
prise a rigid frame at its centre, hosting an electric motor to impart
to an oscillating arm alternative horizontal movements, the ampli-
tude (usually 3 cms) and frequency (1–5 Hz) are fixed by the oper-
ator. Swatches are attached to the oscillating arm through their
plastic collar in such a way that, at rest, they remain in a perfectly
vertical position. At each corner of the pentagon, 5 digital video
cameras (100 images/second, 640 by 480 pixels recording frame),
each facing a LCD screen, allow images to be constantly recorded
under their respective angles. The LCD screen, as a backlight,
ensures a perfect contrast with the swatch during video recordings.
The recording phase of the movements of swatches can be per-
formed through two different protocols: i) during the continuous
oscillations or ii) when movements stop, to record its various
behaviours during the damping phase (time to reach immobiliza-
tion, intermediary shapes). Due to the attachment conditions of the
swatch to the oscillating arm, video recordings only concern 25
centimetres of the swatch, thereby ignoring the first 2 cms of the
attachment region.
Major recorded parameters
Image analysis allows two basic shape criteria to be recorded
(Fig. 2), its ‘backbone’ (the shape skeleton as a summarizing line)
Figure 1 General view of the equipment used for ensuring the movements
of swatches and video recordings.
57
Hair movement evaluation A. Galliano et al.

(a) (b) (c)

Figure 2 Two main reconstructions, determined by image analysis, may describe the overall behaviour of the hair swatch, that is a summarized shape (‘skele-
ton’) (a) and its overall geometric feature (‘envelope’ or developed surface) (b). As example, oscillations at 3 Hz (amplitude 3cms) lead to successive contractions
(minimal area) and expansions (maximal area) of the envelope that can be further completed by a virtual local sub-division of the swatch into 5 segments (5
cms each), Fig. 2c.

and its ‘envelope’, that is the overall surface developed by the

swatch during movements.
Based upon this two shape criteria, it is possible to compute, as
a function of time, each coordinate of the skeleton and the enve-
lope of the swatch during the movement (Fig. 2).
Additional to these shape criteria, some major parameters
emerged from preliminary experiments:
1 The maximal amplitude of the swatch tip during the movement
referred as L, in cm, computed by recording the position of each
points of the skeleton during the movement
2 The highest height attained by the tip of the swatch, referred as
H, in cm, computed by recording the position of each points of
the skeleton during the movement
3 An overall parameter, that is the respective area, in cm², of
two shape features taken by the swatch during oscillations,
that is contraction (Amin) and expansion (Amax) according to Figure 3 Regular phases of contractions and expansions in the envelope
the frequency used. The local contraction and dilation could (overall area) of the hair swatch during oscillations.
be assessed by following the width of the five segments along
the swatch. These two parameters are computed by recording
for recording the maximal amplitude of the dynamics, that is to
the position of each point of the envelope and 5 selected seg-
detect the most evident effect brought by products because some
ments along the swatch (each of a 5 cm length) during the
may expand or contract the hair swatch during movements. Inter-
movement.
actions between hair fibres then play a major role in this phenome-
As shown by Fig. 3, oscillations lead hair swatches to undergo non because possibly influenced by some product-derived properties
alternative phases of contraction and expansion, illustrated by a or mechanisms (entanglement, adhesive bonds).
maximal envelope area (Amax) and a minimal envelope area
(Amin).
Damping period
To observe local contractions and expansions of the swatch dur-
ing movement, such measurement can be applied to various parts A complementary aspect of such movements deals with the damp-
of the swatch, that is to the 5 longitudinal segments. Experimental ing period when oscillations stop. By following the horizontal
curve looks like that observed on Fig. 3, the width of each segment amplitude as a function of time, the progressive decrease in ampli-
oscillating between a minimum and a maximum value. With tude (L) and time required to complete immobility can be recorded
increased frequency (0.5–5 Hz), some changes occur in the values and computed. Horizontal amplitude and time to reach immobility
of the minimum/maximum sections of each segment. The dynam- are recorded after stopping 5 oscillations of 10 cm amplitude at a
ics of the hair swatch behaviour can only be determined at 1.2 Hz frequency. The damping curve can be modelled by a
frequencies above 1 Hz. In addition, it strongly suggests, post-appli- decreasing exponential function from which the damping time can
cation of some products, to ‘scan’ various frequencies (1.5–5 Hz) be extracted.

58 © 2014 Society of Cosmetic Scientists and the Soci
et
e Francßaise de Cosm
etologie

International Journal of Cosmetic Science, 37, 56–62
Hair movement evaluation A. Galliano et al.

(a)
0.5Hz 1 1.5Hz 2Hz 2.5Hz 3Hz 4Hz 5Hz

Horizontal amplitude
(b) Root
Hair length
Tip

Figure 4 Pictures of the various shapes and modified amplitudes of a hair swatch according to frequency from 0.5 to 5 Hz (a). Computed values along various
hair segments, from root to tip. 1.5 Hz frequency clearly leads to the largest horizontal amplitude of the tip (b).

Results

Effect of frequency
Under a 3 cm constant amplitude of oscillations, swatches were
submitted to various frequencies, from 0.5 Hz to 5 Hz, as shown in
Fig. 4. It clearly appears (Fig. 4a,b) that from 0.5 to 1.5 Hz,
swatches respond to pendulum movements by an increased L (hori-
zontal amplitude of their tips), although keeping a rather straight
‘I’ shape.
Inversely, at higher frequencies (2–5 Hz), L clearly decreases,
associated with a change in shape with the appearance of vibration
bounce, towards a ‘S’ aspect. As mentioned above, with regard to
further testing’s, preference was given to both 1.5 and 3 Hz fre-
quencies for assessing a most differentiating behaviour between
swatches, remaining in a rather low range, that is frequencies close
to that of human gestures. Figure 5 Variations in L along the swatch according to the applied treat-
ment comparing to the reference swatch (Control swatches).

Effects of applied products

Effect of Sebum
50 grams of an artificial Sebum (Squalene 15%, Wax-Esters 25%
(15% Palmityl Oleate, 10% Oleyl Oleate), Triglycerides 60% (20%
Triolein, 20%Tri Palmitin, 20%Tri Palmitolein), all of analytical
grades, was specially prepared and 1 g of this fluid lipid mixture
was manually spread along hair swatches, using gloves. Three dif-
ferent treatment procedures (a, b, c) were applied on the control
swatches (d):
a)Sebum only
b)Sebum followed by rinsing with distilled water, then drying with
a standard hairdrier.
c)Sebum + shampoo (2 g of a bland commercial brand, conditioner
free), then rinsing and drying
d)Control swatches: no sebum applied, just shampooed (same as
above), then rinsed and dried
Figure 5 shows the changes in L (at 3 Hz frequency) along seg-
ments, induced by the four treatments. It clearly illustrates the
effect of sebum alone (a), making the hair swatch much stiffer
whereas the other treatments lead to rather comparable behav-
iours. It also shows that rinsing with water (b), under manual
frictions, strikingly leads to a behaviour comparable to shampooing
(c, d) by a likely physical elimination of most sebum.

© 2014 Society of Cosmetic Scientists and the Soci
et
e Francßaise de Cosm
etologie 59

International Journal of Cosmetic Science, 37, 56–62
Hair movement evaluation A. Galliano et al.

Effect of bland and conditioning shampoos less expanded (or more contracted) features, an effect shown strongly
Four series of swatches (8 per test) were shampooed under manual dependent upon the intensity of the conditioning treatment.
and standard procedures (1 g each) by a commercial bland sham- Figure 10 illustrates the various appearances and amplitudes of hair
poo (no conditioning ingredient, used as reference) or by a condi- swatches during movements according to conditioning levels.
tioning shampoo followed by hair care products of increased
conditioning potentials. The hair care products, all being available Reproducibility of measurements:
on the market, were selected according to both their content in Although standardized as much as possible, there remain inherent
conditioning ingredients and successive steps of applications, that is sources of possible variations (nature of the swatches, effect of pro-
carried out under two to three operating steps: ducts, operator-dependance). All recorded values (L, H, envelope,
width. . .) show acceptable reproducibility, that is, S.D (standard
1 ‘Binomial’: a conditioning shampoo followed, after rinsing, by the
deviation) values reach 10–15% of the mean. In most cases, a min-
application of a hair care product with a high content in cationic
imum of 8 swatches per test appears suitable for complying with
polymers. Then swatch is rinsed and gently dried.
statistics rules when attempting to differentiate swatches or effects
2 ‘Standard Trinomial’: Same shampoo as above, followed, after
of products.
rinsing, by the application of an amino-silicone-based hair care
product. Then swatch is rinsed and a leave-on silicon-based
cream is further applied. Discussion
3 ‘High Trinomial’: Same shampoo as above, followed, after rinsing,
The method developed here is a complementary approach, or a
by the application of a silicone-based hair mask. Then swatch is
refinement, to those used by two unpublished techniques, [7]. Due
rinsed and the same leave-on cream as above is applied.
to the lack of informative details –since unpublished – differences
All 32 swatches were then submitted to alternative oscillations
(1.5 Hz, amplitude 3 cms), and various parameters were recorded,
as follows: 6000 Minimum area (cm²)
Maximum area (cm²)
1 Influence on L: As shown in Fig. 6, the amplitude of L greatly var- 5000
Swatch area (mm²)
ies with conditioning potentials.
2 Influence upon vertical amplitude (H, in mm): Comparable to L, the 4000
tip of swatches show increased lifting up with increased condi-
tioning potentials (Fig. 6). 3000
3 Influence upon some shape parameters of the swatches: Fig. 7 and
Fig. 8 show that both the developed surface of swatches (envelope, 2000
in mm2), and width of the mid-part/tip of the swatch (mm),
1000
decrease with increased conditioning potentials during movements.
4 Influence upon the damping period: when movements stop, hair 0
conditioning procedures clearly lead to a shortened time to reach Commercial Binomial Standard High trinomial
a complete immobilisation of the swatches, decreasing from 8 to bland shampoo trinomial
4 s with increased conditioning potentials (Fig. 9).
Figure 7 Decreased shape of hair swatches during movements, induced by
Taken altogether, Figs 6–9 indicate that increased conditioning increased conditioning potential. All results of minimal and maximal areas
treatment leads to a more coherent behaviour of swatches, that is are statically different (P < 0.05).

35 180
Tip vercal amplitude 40
160 Minimum width - mid part
Horizontal amplitude (mm)

30 Tip horizontal amplitude Maximum width - mid part

Vercal amplitude (mm)

140 35
Minimum width - p posion
25 30
Swatch width (mm)

120 Maximum width - p posion

100 25
20
80 20
15 60
15
40
10 10
20
5 0 5
Commercial Binomial Standard High 0
bland trinomial trinomial Commercial Binomial Standard High trinomial
shampoo bland shampoo trinomial

Figure 6 Increased horizontal amplitude (L, in mm) and vertical amplitude
(H, in mm) of the tip of the swatches with increasing conditioning potential.
All results of the tip vertical amplitude are statically different (P < 0.05), at
the exception of Commercial bland shampoo vs. Binomial. All results of the
tip horizontal amplitude are statically different (P < 0.05), except Commer-
cial bland shampoo vs. Binomial.
Figure 8 Changes in the width (max and min values, in mm) of swatches
at their mid-part (segment 3)/tip position (segment 5) with increased condi-
tioning potential. In all cases, minimal and maximal widths of the mid-part
statistically differ (P < 0.05). In all cases, the minimal and maximal widths
of the tip position statically differ (P < 0.05), at the exception of maximal
width of tip, between standard trinomial and binomial regimen.

60 © 2014 Society of Cosmetic Scientists and the Soci
et
e Francßaise de Cosm
etologie

International Journal of Cosmetic Science, 37, 56–62
Hair movement evaluation A. Galliano et al.

in the tested products or experimental conditions, it is difficult to head-hairs. Preliminary tests of wavy or highly curled swatches
compare the respective advantages of these proposed methods with (being natural or permed), that is above type III, led to complex
the one presented here. The latter seems however offering addi- and irreproducible behaviours, of an ‘obscure’ logic, preventing an
tional parameters to being recorded such as changes in the overall adequate selection and the recording of a reliable parameter.
shape (envelope, backbone, contraction. . .) and a possible 3D The selection of straight hair-based swatches (type I or II) allows
reconstruction of the hair swatches. nonetheless a variety of parameters dealing with shape or displace-
Whatsoever, these methods depict a new approach in the studies ments to being recorded. Combined, they offer a versatile feature
of head-hair properties, that is their dynamic response when sub- for best defining/adapting a protocol suitable to quantify subtle dif-
mitted to standardized movements. In the technical conditions used ferences between swatches or products of different effects (condi-
by our presented method, the initial shape of hair swatches (i.e. tioners, straighteners, fixating gels or sprays. . .). The preliminary
straight, type I chosen here) clearly appeared a limiting factor, pre- results from the present work indicate that, among other factors,
venting to study the dynamic behaviour of all possible shapes of interactions between fibres play a strong role in the maintenance
of the structure of the swatch when forced to oscillations. In other
words, the surface status of the hair shaft, neutral/hydrophobic
10 when native or polar/hydrophilic when oxidized or damaged is
9 likely paramount. Coating hair surface with sebum or increasingly
8 potent conditioners decrease interactions between hair fibres by
7 masking/hiding polar/negative groups such as sulfonic moieties.
Damping me (s)

Some preliminary studies with various oils seem confirming a close

6
sebum-like effect. Inversely, washing swatches with a bland sham-
5 poo likely reveals all these superficial polar groups, obviously lead-
4 ing to repulsive forces which, in turn, contribute to an easier de-
3 structuration of the swatch. Simple calculations, assuming hair as
2 a perfect cylinder, indicate that a 2,7 g and 27 cm long swatch
1
formed by hairs of 70l in diameter, comprise about 2000 hairs,
thus forming a vast interactive network between fibres.
0
The latter is, in addition, highly qualitative. With regard to hair
Commercial Binomial Standard High trinomial
bland shampoo trinomial
growth, of about 1 cm/month [10], the tip of a 27-cm long hair
reflects about 2 years of physical assaults (brushing, combing, heat
Figure 9 Decreased time to reach a complete immobilization of the drying. . .) weathering, oxidation, etc. [11]. From root to tip, the
swatches with increased conditioning potential. All treatments are statically qualitative changes are of a gradient aspect, from native to ‘aged’
different (P < 0.05) from commercial bland shampoo. states with length. Such a gradient has been fully described in a

Increased
dilatation
movement
Narrow
movement

Wide

Commercial bland shampoo

L L

Binomial

Standard trinomial

High
trinomial

Increased
contraction

Figure 10 Various shapes and movements of the hair swatches according to conditioning level. The stronger conditioning effect leads to both a much more
coherent appearance and a higher amplitude of movements. In brief, highly conditioned swatches respond to oscillations by an increased ‘suppleness’ and a
more homogeneous behaviour of their fibres.

© 2014 Society of Cosmetic Scientists and the Soci
et
e Francßaise de Cosm
etologie 61

International Journal of Cosmetic Science, 37, 56–62
Hair movement evaluation A. Galliano et al.

previous work [12] where exceptionally long hairs (>2.4 m) show,
with length, progressive alterations such as losses in 18-MEA, in
shine, together with a progressively disorganized cuticle, up to its
disappearance at the tip region, all elements being paralleled by the
increased the presence of surface (anionic) polar groups.
movements of hair swatches seem offering a promising approach
for integrating all factors, being intrinsic or resulting from product
applications. As such, they probably not only better reflect real-
life conditions but confer to cosmetic products an additional
dynamic vision.

Conclusion
The dynamic method exposed here offers a wide facet of parame-
ters to which additional studies are needed for selecting those
(and their experimental conditions) which best quantify a given
property brought to a hair swatch, or comparing swatches
made from different hair types or diameters. These standardized
Acknowledgements
The authors are grateful to Didier Saint L
eger for useful discus-
sions. This work benefited from the valuable support of Delphine
Trillat, Bernard Querleux and Genevieve Loussouarn. The authors
deeply thank Guillaume Marty, Camila Flor and Kiran Nowbuth for
their scientific contribution.

References
1. (Morris, D., ed.) The Naked Ape. Jonathan
Cape, London (1967).
2. (Bouillon, C. and Wilkinson, J., eds.) The
Science of Hair Care, 2nd edn. Taylor and
Francis, Boca raton, Florida 33487
(2005).
3. (Robbins, C., ed.) The Chemical and Physical
Behavior of Human Hair, 3rd edn. Springer-
Verlag, New York, 10010 (1994).
4. Kurihara, T., Anjyo, K. and Thalmann, D.
Hair animation with collision detection. Pro-
ceedings of Models and Techniques in Com-
puter animation, Geneva (1993).
5. Tae-Yong, K. and Neumann, U. Modeling
Rendering, Animating Human Hair, Integrated,
Media Systems. Center University of South-
ern, California (2002).
6. Ward, K., Galoppe, N. and Lin, MC. A simu-
lation-based VR system for interactive hairstyl-
ing. IEEE Virtual Reality 1624074, page 37.
(2006).
7. Ward, K., Galoppe, N. and Lin, M.C. Interac-
tive virtual hair salon. Presence (Camb) 16,
237–251 (2007).
8. Picot-Lemasson, A., Decoq, G, Aghassian, F
and Leveque, JL. Influence of hairdressing
on the psychological mood of women. Int. J.
Cosmet. Sci. 23, 161–164 (2001).
9. De la Mettrie, R., Saint-Leger, D., Lousso-
uarn, G., Garcel, AL, Porter, C and Langa-
ney, A. Shape variability and classification
of human hair: a worldwide approach.
Hum. Biol. 79, 265–281 (2007).
10. Loussouarn, G., El Rawadi, C. and Genain,
G. Diversity in hair growth profiles. Int. J.
Dermatol. 44, 6–9 (2005).
11. Galliano, A., Saint-Olive Bacque, C., Marty,
G., et al. Resistance of human hair cuticle
after a shaking in wet conditions: compari-
son between Chinese and Caucasian hair.
Int. J. Cosmet. Sci.. 32, 356–368 (2010).
12. Thibaut, S., De Becker, E, Bernard, BA.,
et al. Chronological ageing of human hair
keratin fibres. Int. J. Cosmet. Sci.. 32, 422–
434 (2010).

62 © 2014 Society of Cosmetic Scientists and the Soci
et
e Francßaise de Cosm
etologie

International Journal of Cosmetic Science, 37, 56–62
Copyright of International Journal of Cosmetic Science is the property of Wiley-Blackwell
and its content may not be copied or emailed to multiple sites or posted to a listserv without
the copyright holder's express written permission. However, users may print, download, or
email articles for individual use.