Tiêu chuẩn iso tr 19263 1 2017

© ISO 2017 Photography — Archiving systems — Part 1 Best practices for digital image capture of cultural heritage material Photographie — Systèmes d’archivage — Partie 1 Meilleures pratiques pour la c[.]

Photog raphy — A rchi ving systems —

Part 1:

Best practic es for dig ital imag e

Ph to ra hie — S stème s d’ar hiva e —

Partie 1: Meilleure s pratiques p ur la ca tur d’ima es n mériques

du matérie l de p trimoine culture l

COPYRIGHT PROTECTED DOCUMENT

© ISO 2017, P blshed in Sw itz rlan

A ll rig hts r eserved Unles otherw ise spe ified, nopar of this p blc tion ma y be r epr od c d or utilz d otherw ise in an form

or b an me ns, ele tr onic or me hanic l, inclu in p oto opying , or postin on the internet or an intranet , w ithout prior

written permis ion Permis ion c n be req esed from either ISO at the ad r es below or ISO’s member bod y in the c u try of

F reword v

Introduction vi

1 Sc ope 1

2 A nal ysis of imag e qual ty 1

2.1 General 1

2.2 Imag e q alty char acte is ics 1

2.3 ISO 1 2 4 Tes chart te h ical featur es 2

2.4 Grid an g ra y/white featur es 2

2.4.1 General 2

2.4.2 Ru ning scale features (cm and inches) 3

2.4.3 Gra yscale an ru ning g r a y/white/ lack b r featur es 3

2.4.4 C lour p tch featur es 3

2.4.5 MTF measur ement featur es 4

2.4.6 A dditional ISO 1 264 tar g et featur es/r efer enc data 4

2.5 A dditional tar g et 4

2.6 Linear g ra yscale

5 2.6.1 DCSG colour chart 5

2.6.2 Limitations of Chart Based Imag ing Sys em A nalysis 5

3 Imag e q alty levels 6

4 Basic pr inciples of imag e captur e and pr oc es ing 6

4.1 Ove view 6

4.2 S ene r efe red an output r efe red imag e s ates 7

4.3 Use contr ols an r eadout 7

4.3.1 General 7

4.3.2 C lour Pr oc s ing C ntr ols 7

4.3.3 Ex posur e readout 8

4.3.4 Raw pr oc s or r eadout an contr ols 8

4.3.5 Othe use contr ols 8

4.3.6 Unwanted data modification 8

4.4 Mas e imag es an de iv tives 8

4.4.1 General 8

4.4.2 Raw imag e fi es 8

4.4.3 A rtwor k r epr od ction cycle 9

5 Imag ing system setup an cal bratio 10

5.1 General 1

5.2 Position camera sys em 1

5.3 Establsh u iformity-even i umination 1

5.3.1 General 1

5.3.2 Optional flat-fielding 1

5.4 Establsh ex posur e 1

5.5 Establsh tone r epr od ction curve (OEC ) 1

5.6 Cr eate an IC colour pr ofie 1

5.7 A nalyse colour an tone 1

6 A pplcatio of imag e qualty anal ysis 12

6.1 S le tion of imag ing sys ems: pr eflg hting eq ipment or ven ors 1

6.2 Using ISO 1 2 4 tar g et: Initial sys em config uration 1

6.3 Using ISO 1 2 4 tar g et: Sys em pe formanc ev luation (benchmarking ) 1

6.4 Using ISO 1 2 4 tar g et: Ong oing pe formanc monitoring 1

7 Technical metadata for imag e qualty anal ysis 14

A nne x A (informative)Line r Grays ale L* to RGB c onver sio table 15

Biblog raph y 19

ISO (he Int ernational Org nization for Stan ardization) is a worldwide fede ation of national s an ards

b dies (ISO membe b dies) The work of pr p ring Int ernational Standards is normaly car ied out

through ISO t ech ical committ ees Each membe b dy int er st ed in a subje t for w hich a t ech ical

committ ee has be n es a l shed has the right t o be r pr sent ed on that committ ee Int ernational

org nizations, g ove nmental an non-g overnmental, in laison with ISO, also take part in the work

ISO cola orat es closely with the Int ernational Ele trot ech ical C mmis ion (IEC) on al matt ers of

ele trot ech ical s an ardization

The proc d r s used t o develo this document an those int en ed for it furthe maint enanc ar

desc ibed in the ISO/IEC Dir ctives, Part 1 In p rticular the dife ent a pro al c it eria ne ded for the

dife ent ty es of ISO document should be not ed This document was draft ed in ac ordanc with the

edit orial rules of the ISO/IEC Dir ctives, Part 2 ( e www iso org dir ctives)

A tt ention is drawn t o the p s ibi ity that ome of the element of this document ma be the subje t of

p t ent right ISO shal not be held r sponsible for identifying any or al such pat ent right Detais of

any p t ent right identif ied d ring the develo ment of the document wi be in the Introd ction an / r

on the ISO ls of p t ent de larations r c ived ( e www iso org p t ent )

Any trade name used in this document is information given for the convenienc of use s an does not

cons itut e an en orsement

For an ex lanation on the meaning of ISO spe ific terms an ex r s ions r lated to conformity as es ment,

as wel as information a out ISO’s adhe enc to the Wor ld Trade Org nization (WTO) principles in the

Te h ical Bar ie s to Trade (TBT) se the folowing URL: www iso org / iso/ for word html

The committ ee r sp nsible for this document is ISO/TC42, Ph to ra h

Toda digitization pro rams ne d t o satis y the deman s of an int er onne t ed dynamic use

commu ity A digitiz d imag e can be r purposed ac os any n mbe of sy st ems an the efor ne ds t o

be wel def ined, t ech icaly ro us an media a nos ic The digital imag e of an original is int en ed t o

satis y multiple uses inclu ing ac es , ar hiving, r sear h, conse v tion, ed cation, marketing, social

media, r prod ction and dis ribution b th in print an onlne

Int en ed for org nizations, such as cultural he itag e ins itutions, ISO 1 2 4-1 spe if ies a method for

analysing ima ing sy st ems w he e it is imp rtant t o control the degr e of ac uracy an t o ensur that

ima ing q alty is maintained o e time The e ar thr e common a plcations of ISO 1 2 4-1:

a) ima ing sy st em pe formanc ev luation (benchmarking) – used for sy st em develo ment an

sy st em sele tion

b) ima ing sys em pe formanc o timization – used for tai oring the sys em to a p rticular jo

(use case)

c) ima ing sy st em pe formanc monit oring – used for control ng that the q alty of the sy st em

r mains consist ent an within spe if ications o e time

The purp se of this document is t o pro ide practical guidanc on how t o a ply ISO 1 2 4-1 for

cultural he itag e ima ing of two-dimensional originals This inclu es how the imag e q al ty analy sis

is pe formed, the fu ction of t echnical targ et featur s, an how t o adjus / ptimiz the pe formanc

of ima ing sy st ems A dditionaly this document i us rat es how ISO 1 2 4-1 can be used for sele tion

of a pro riat e ima ing sy st ems an how t o es a l sh an maintain imag e q alty in digitization

workflow s

An ex B pro ides information r lat ed t o develo ing a digitization s rat eg inclu ing as es ment of

cole tions, develo ing a hardwar s rat eg an sy st em sele tion

ISO 1 2 2 pro ides definitions for ima ing t erminolo y used in this document an ISO 1 2 4-1

Photog raphy — A rchi ving systems —

Part 1:

This document spe ifies how t o pe form q al ty analy sis of ima ing sy st ems (e.g flat bed scanne s,

planetary scan e s, or digital s i came as) used for digitization of r fle tive two-dimensional

originals

Original mat erials inclu e but ar not l mit ed t o b oks, t extual document , drawings, print ,

phot ogra hs, an p intings Ce tain ty es of two-dimensional mat erials with complex surfac

g eometry an or highly r fle tive surfac element r q ir spe ial i lumination t ech iq es that can fal

out ide the sco e of this document

NOTE ISO/TS 1 26 - 2wil ad res transmis ive materials

2 A nal ysis ofimag e qual ty

2.1 General

In orde t o analyse ima ing sy st em q alty ISO 1 2 4-1 spe if ies a t ech ical targ et (ISO 1 2 4-1 targ et)

designed t o incorp rat e multiplet echnical featur s for the measur ment of key ima ing charact eris ics

from a single imag e C lculations ar pe formed via sof war dedicat ed t o ISO 1 2 4-1targ et analy sis

2.2 Imag e qual ty characteristic

Imag e t echnical analy sis inv lves a n mbe of int er elat ed measur ment st eps, ty icaly the analy sis

proc s begins with v ldating w hit e balanc an t one r prod ction folowed b ad itional calculation

st eps as l st ed below When al measur ment ar within a set of def ined t ole anc s, an ima ing sy st em

me t a def ined q alty level R esolution an g eometry ar analy sed aft er firs analy sing cor imag e

q alty element

— White Balance: adjus ment of ele tronic s i l pictur colour chan el g ins or imag e proc s ing

so that radiation with r lative spe tral p we dis ribution eq al t o that of the sc ne i umination

sour e is r n e ed as a visual neutral

— Tone Repro uction C rve (TRC): curve gra hicaly desc ibing the r lationship betwe n the input

t ones an the output t ones in an ima ing proc s

— Gain Mod latio (highl ght / the pat ches):v riation of the g in o e the signal level

— Noise: u want ed v riations in the r sponse of an ima ing sy st em

— Dynamic Rang e: the dife enc , o e a given pe iod of time, betwe n ma imum an minimum

signal levels, ex r s ed in de ibels, contras ratios or f-st ops

— Banding: u want ed s ripes or b n s that oc ur in a digital imag e

— Defect Pix els: pix el or sub ixel that o e at es in a wa othe than the one in w hich it is driven

— Colour A ccuracy: a i ty of an ima ing sy st em t o r prod c the colours of some int en ed o je t, as

spe if ied using somecolour dife enc metric

— S mplng Rate (dife enc betwe n claimed an o tained): n mbe of samples pe unit of time,

angle, r v lutions or othe me hanical, in ependent v ria le for u iformly sampled data

— Resolution (lmiting): measur of the a i ty of a came a sy st em, or a comp nent of a came a

sy st em, t o depict pictur detai

— Sharpening: amplf ication of the SFR b means of imag e proc s ing t o achieve sharpe a pearing

imag es Also, a clas of imag e proc s ing o e ations that enhanc s the contras of sele tive sp tial

fr q encies, usualy visualy imp rtant ones

— MTF 50: the mod lation trans e function is, a measur of the trans e of mod lation (or contras )

from the subje t t o the imag e an is used t o measur sp tial fr q ency r sp nse ( SFR) In othe

words, it measur s how faithfuly the ima ing sy st em r prod c s (or trans e s) detai from the

targ et t o the digital imag e MTF 0 r fe s t o that sp tial fr q ency (ex r s ed in l nes pe mm) at

w hich the imag e r tains 5 % of the t es targ et’ s contras , se ISO 1 2 3

— I luminat io n n-uniformity (arget siz r lated): a pplcation of visible radiation (lght)to an o je t

— Colour mis-re istratio : colour-t o-colour sp tial dislocation of othe wise sp tial y coincident

colour featur s of an imag ed o je t

— Distortio : displac ment from the ideal sha e of a subje t (lying on a plane paralel t o the imag e

plane) in the r corded imag e

— Reprod ct ion s ale: ratio of the siz of an o je t in a digital imag e an the siz of the original o je t

2.3 ISO 19264 Test chart technical features

The ISO 1 2 4-1 targ et is def ined in ISO 1 2 4-1:—, Annex A In ivid al chart featur s ar r prod c d

he e t o i lus rat e fu ctionalty An ISO complant targ et should contain al of the t echnical featur s

A dditional targ et ar uti z d for charact erizing ima ing sy st em colour an t one

2.4 Grid and g ray/white fe tures

2.4.1 General

Figure 1 — E ample of grid and gray/w hite fe tures

Gra /w hit e grids ar used for analy sing i umination non-u iformity an dist ortion I lumination

non-uniformity is simiar t o w hit e b lanc , but a ples t o i lumination at al t onal levels ac os the entir

ima ing f ield an can be adve sely afe t ed b the introd ction of non-imag e forming lght an or lens

falof Dist ortion is oft en cor e t ed digitaly, but doing so r calculat es each pix el location in an imag e,

this ma neg tively influenc imag e r solution but ma also contribut e t o an o e al impro ement in

imag e r prod ction ac uracy I lumination-non u iformity r sult ar ex r s ed as ΔL* dife enc s

betwe n the ma imum an minimum L* v lues

2.4.2 Running s ale fe tures (cm and inches)

Figure 2 — Example of running s ale

S ales ar used t o det ermine X an Y r solution, and t o t es for cons ant mo ement scan e s, s it ching

sy st ems)

NOTE T is me sured fu ction identif ies the actual ima ed values in b th x an y directions, as uring s ale

inte rity of the ima es

2.4.3 Grays ale and running g ray/white black ba fe tures

Figure 3 — Example of gray scale and running gray/w hite/black ba fe tures

The gray scale an run ing gra /w hit e b rs ar used t o det ermine OEC (t one r cording), g in

mod lation, noise, an signal t o noiseratio

Ima ing sy st ems should conve t the t one v lues in the original sc ne t o digital v lues; this t ech ical

t erm is OECF (Opt o-Ele tronic C nve sion Fu ction) Valdation of the cor e t sele tion of these

p ramet ers an a pro riat e r pr sentation of the digital information for the sele t ed p ramet ers is a

c itical fu ction of imag e q alty analy sis

Gain mod lation r fe s t o the v riation of the g in (dis ribution of t onal v lues) o e the signal level

an is a c itical fact or in r prod ction ima ing an colour ac uracy R ep rt ed as ΔL*v lues The smale

the deviation between the L* of the p t ches in the r fe enc targ et an the L*v lues r pr sent ed b the

digital codev lues the mor ac urat e the t one r prod ction

Noise is g ene aly the digital eq iv lent of f ilm grain, an pr sent it elf as pix el-t o-pixel fluctuations

oft en se n in de pshadow ar as Noise has the efe t of r d cing the o e al pe c ived smo th t onalty

of an imag e Noise can also take a one-dimensional form caled b n ing or s r aking

Signal t o noise ratio is the ratio of the inc emental output signal t o the ro t mean sq ar (rms) noise

level, at a particular signal level

2.4.4 C lour patch fe tures

Figure 4 — Example of colour patch fe tures

The colour p t ch element is used for det ermination of colour ac uracy, t es of the colour sp c , v ldation

of IC prof iles, an survey of colour v riation ac os the scan ing ar a

R esult ar r port ed in ΔE 2 00* v lues in the form of a ta le for each in ivid al p t ch t og ethe with

the r sult for the mean an the ma v lue for al p t ches It is sufficient t o r port the mean an the

ma v lue only Obse v tion ofthe bes 90 % can be helpful t o help identify outlying data but is not

mandat ory

ΔE 2 0 * v luesar calculat ed using a l near ( SL= 1) formula ( e ISO 1 2 4-1)

2.4.5 MTF me surement fe tur es

Figure 5 — Example of MTF me surement fe tures

The MTF element ena les measur ment of sampl ng r solution ac ording t o ISO 1 0 7-1 (up t o

1 0 P I ma )

R esolution (Limiting) is the highes fr q ency ( p cing) that imag e detai can be dis inguished

S an e s an came as ma claim ve y high r solutions that ar unachiev ble d e t o design lmitations

of the t otal ima ing sy st em This measur identifies the actual achieved r solution an should not be

confused or conside ed eq iv lent t o sampl ng rat e

This chart element also helps calculat e sampl ng efficiency, an pro ides for visual r solution che k

up t o 1 lp /mm Samplng eff iciency is also calculat ed using the MTF Ex mple-if the o je t ca tur d

is 1 in long an the sensor has 40 0 pix el featur s ca turing the 1 inches, the samplng rat e is

40 pix els /in Mos ima ing sy st ems cannot achieve 1 0 % sampl ng eff iciency An ac urat e samplng

rat e is es ential t o k owing the siz of the original o je t

2.4.6 Additional ISO 1 264 tar g et fe tures/r eferenc e data

A dditional chart ar as ma be designat ed for la el ng, ad itional t es p tt erns or chart featur s an

man facturing information Chart R efe enc Data ar ty icaly cust om measur d an delve ed from

t es chart ven ors in t ext a le form t o be used as a r fe enc for calculations Chart r fe enc data set

an measur ment methods should be document ed

2.5 Additional tar g ets

In ad ition t o the ISO 1 2 4-1 targ et othe targ et ma be used t o charact eriz the ima ing sy st em

The folowing targ et aid in the charact erization of ima ing sy st ems

2.6 Linear g rays ale

Key

1 semi g los values (5 * to 9 *L)

2 glos black (4L*)

3 me surement s ale, in mm

4 per ceptual mid le value (5 L*)

Figure 6 — Example of l ne r gray scale

A l near gray scale is useful for conf iguration an ve ification of t one r prod ction (OECF) an g in

mod lation The targ et incorporat es semi-glos spe tral y neutral pigment eq aly sp c d in 5 * st eps

from L*5 t o L*9 with ad itional glos black p t ches The glos pat ches ext en the dynamic rang e an

ar used t o visualy as es lghting r fle tions an glar from impro e lghting g eometry

2.6.1 DCSG c olo r char t

Figure 7 — Example of DCSG colo r chart

The X Rit e C lour Che ke

®

Digital SG (DCSG) colour chart is useful for colour calbration (devic

charact erization) C lour chart ma v ry in t erms of subs rat e, glos fact or, colour g mut an n mbe

of p t ches A colour chart that closely mat ches the surfac q alty an colour g mut of the original

artwork ma be uti z d

2.6.2 Limitatio s of C ar t Based Imag ing System A nal ysis

Being that ISO 1 2 4 is b sed up n analy sis of t es chart with t ech ical p tt erns an r fe enc v lues

the e ar inhe ent l mitations that ne d t o be conside ed Fabrication of t ech ical targ et v ries o e

time, an targ et ha e a finit e l fe sp n Baselne data used t o define t echnical targ et (chart r fe enc

data) can also v ry between use s an ven ors Ven ors ma impro e ly implement the analy sis

methods outlned in ISO 1 2 4 Bey n these p s ible v ria les, the e ar v ria les in the surfac

q alties of original artworks, ca tur i uminant an sensors that lmit the a i ty t o ensur an ex ct

colourimetric or pe c ptual mat ch

3 Imag e qual ty levels

Imag e analy sis of a t echnical targ et r sult in an ar a of v lues A cor element of 1 2 4-1 is the use

of aims an t ole anc s t o pro ide v lua le insight int o imag e q al ty These aims an t ole anc s have

be n de ived via ext ensive t es ing an fe db ck from cultural he itag e ima ing use s an pro ram

manag ers

ISO 1 2 4-1 def ines thr e imag e q alty levels pr sent ed as a matrix It is imp rtant t o not e that hese

q alty levels ar not pro ided for any spe ific use case or cat eg ory of artwork the efor r aching the

highes ima ing q alty thr shold for al cat eg ories is not a u ive sal r q ir ment The q alty levels ar

meant t o pro ide use s with a r fe enc t o g ug e r lative imag e q alty an t o help es a l sh workflow

b selnes En use s, use commu ities, or ins itutions ma r fe t o the 1 2 4-1 q alty level matrix

as ne ded t o ad r s dife ent o je t y es, t o document and shar r sult or t o spe ify imag e q alty

r q ir ment as p rt of contractual a r ement with out ide digitization ven ors Pro ram manag ers

ma cho se t o conf igur an maintain sy st ems that ex ce d the t ole anc definition matrix defined in

ISO 1 2 4-1 It is imp rtant t o document any sit e or proje t pe if ic q alty aims

Please r fe t o the imag e q alty ta le in ISO 1 2 4-1

4 Basic principles ofimag e captur e and proc es ing

In orde t o r cord an original digital ima ing sy st ems g ene aly folow the st eps outlned in the flow

dia ram shown in Figur 9 w hich i lus rat es a ty ical ar a sensor devic

Figure 8 — Typical ar ay sensor device

sp tial patt ern of r peat ed, e.g r d, gr en an blue, signals A t this p int hese raw data cons itut e the

firs form of ‘ raw’ r corded imag e, the raw cor e t ed C A data

The next st ep in a ty ical proc s ing path is the g ene ation of a fuly po ulat ed thr e-colour imag e

ar a Pro riety alg orithms, aimed at minimizing colour art efact , can be a pled he e This de-

mosaicing o e ation is the int erp lation of the single-r cord ar a t o a ‘ raw’ int erpolat ed r d, gr en

an blue data set Whie de-mosaicing alg orithms ha e impro ed o e time, r prod ction of c rtain

originals with halft ones, et chings an othe mat erials with high fr q ency visual p tt erns can sufe

from colour Moir art efact Moir is def ined as a spatial beat phenomenon g ene at ed b the mod lation

of n me ous sp tial fr q encies Moir art efact can impact b th luminanc an chrominanc Line

scanne s, and multi-shot sensor sy st ems minimiz the oc ur enc of colour Moir art efact as

de-mosaicing is not ne es ary in these ima ing sy st ems

Whit e-b lanc , an matrix colour-cor e tion o e ations ar usualy a pled next The r sult isan imag e

data set that isin a sc ne-r fe r d colour encoding

The final st ep in the imag e proc s ing chain is the r n e ing, usual y for displa The r sult is a f inished

imag e data ar a in an output-r fe r d colour encoding This st ep ma be a simple colour-sp c

trans ormation, but can also inclu e choic s for g mut ma ping an colour pr fe enc

Whi e the a o e st eps ar common in colour imag e acq isition sy st ems, spe ific implementation detai s

wi v ry Un e s an ing the signal (colour) encoding of a raw imag e is as imp rtant as a r ement on a

p rticular file format

4.2 Sc ene referred and output referred imag e states

The t erms sc ne r fe r d an output r fe r d ar es ential t o un e s an bes practic for artwork

digitization ISO 1 2 4-1 employ s o je tive methods t o help c eat e imag es that r fe t o the original

sc ne or o je t, in othe words: a sc ne r fe r d imag e Whi e scanne s ar ty ical y engine r d

t o pro ide a sc ne-r fe r d r sp nse, the majority of comme cial y a aia le ima ing sy st ems

ar engine r d t o del ve finished output r fe r d imag es o timiz d for “ pleasing” r n itions

Unfortu at ely each man factur r an o se ver ma have dife ent subje tive o inions a out w hat is

pleasing as o posed t o w hat isac urat e A sc ne-r fe r d imag ecan be r purp sed an r formatt ed t o

any media as it contains information trac a le b ck t o the original o je t When a sc ne-r fe r d imag e

is conve t ed (via IC or othe colour conve sion) or visualy edit ed an o timiz d for r prod ction t o a

spe if ic medium or devic , it be omes output-r fe r d

4.3 User c ontrols and r eadouts

4.3.1 General

Digital ima ing syst ems (came as or scan e s) an r lat ed control sof war should pro ide use s

ne es ary ac es t o controls r lev nt t o ISO 1 2 4-1 sy st em o timization If an ima ing sy st em l mit

ac es t o c itical controls an only ofe s output r fe r d or “ act ory” imag e proc s ing fu ctionalty,

imag e q alty ma sufe an use s ma be u a le t o con gur sy st ems t o me t defined q alty

c it eria Ifan ima ing sy st em does not ofe a pro riat e use controls and r adout the a plcation

of ISO 1 2 4-1 ma be lmit ed t o ima ing sy st em pe formanc ev luation an ima ing sy st em

pe formanc monit oring only, se ISO/TR 1 3 1-3

4.3.2 C lour Pr oc es in C ntr ols

The aim of the colour proc s ing for ISO 1 2 4-1 is t o prod c ac urat e sc ne colourimetry, with the

sc ne ado t ed w hit e chromatical y ada t ed t o the chromaticity of the imag e encoding ado t ed w hit e

ISO 1 3 1-1 spe if ies came a charact erization metrolo y ISO/TR 1 3 1- 2 pro ides conside ations

for det ermining sc ne analy sis trans orms Came as and scanne s should fuly supp rt cust om use

charact erization methods such as IC colour profiles (ISO 1 076-1), or DNG digital neg tive prof iles

(DCP) Use s should be a le t o sele t any v ld working colour sp c , des ination colour sp c , cust om

g ene at ed input colour prof iles an should be a let o disa le any fact ory or pro rietary colour r n e ing