1、 g49g50g3g38g50g51g60g44g49g42g3g58g44g55g43g50g56g55g3g37g54g44g3g51g40g53g48g44g54g54g44g50g49g3g40g59g38g40g51g55g3g36g54g3g51g40g53g48g44g55g55g40g39g3g37g60g3g38g50g51g60g53g44g42g43g55g3g47g36g58for estimating image quality Part 1: Overview of psychophysical elementsICS 37.040.01Photography Ps
2、ychophysical experimental methods BRITISH STANDARDBS ISO 20462-1:2005BS ISO 20462-1:2005This British Standard was published under the authority of the Standards Policy and Strategy Committee on 30 January 2006 BSI 30 January 2006ISBN 0 580 47649 9The British Standards which implement international p
3、ublications referred to in this document may be found in the BSI Catalogue under the section entitled “International Standards Correspondence Index”, or by using the “Search” facility of the BSI Electronic Catalogue or of British Standards Online.This publication does not purport to include all the
4、necessary provisions of a contract. Users are responsible for its correct application. Compliance with a British Standard does not of itself confer immunity from legal obligations.Summary of pagesThis document comprises a front cover, an inside front cover, the ISO title page, pages ii to v, a blank
5、 page, pages 1 to 17 and a back cover.The BSI copyright notice displayed in this document indicates when the document was last issued.Amendments issued since publicationAmd. No. Date CommentsA list of organizations represented on this committee can be obtained on request to its secretary.Cross-refer
6、encesenquiries on the interpretation, or proposals for change, and keep UK interests informed; monitor related international and European developments and promulgate them in the UK.National forewordThis British Standard reproduces verbatim ISO 20462-1:2005 and implements it as the UK national standa
7、rd.The UK participation in its preparation was entrusted to Technical Committee CPW/42, Photography, which has the responsibility to: aid enquirers to understand the text; present to the responsible international/European committee any Reference numberISO 20462-1:2005(E)INTERNATIONAL STANDARD ISO204
8、62-1First edition2005-11-01Photography Psychophysical experimental methods for estimating image quality Part 1: Overview of psychophysical elements Photographie Mthodes psychophysiques exprimentales pour estimer la qualit dimage Partie 1: Aperu gnral des lments psychophysiques BS ISO 20462-1:2005ii
9、iiiContents Page Foreword iv Introduction v 1 Scope . 1 2 Normative references . 1 3 Terms and definitions. 1 4 Specification of the experimental conditions and results 5 4.1 Observer characteristics 5 4.2 Stimulus properties 6 4.3 Instructions to the observer 6 4.4 Viewing conditions . 7 4.5 Experi
10、mental duration 7 4.6 Results . 8 4.7 Summary of reported quantities . 8 Annex A (informative) Selection of an appropriate psychophysical method 9 Annex B (informative) Stimulus differences, paired comparison proportions, and JNDs . 11 Annex C (informative) Example of a report of a psychophysical ex
11、periment 13 Annex D (informative) Comparison of selected psychometric methods 15 Bibliography . 17 BS ISO 20462-1:2005iv Foreword ISO (the International Organization for Standardization) is a worldwide federation of national standards bodies (ISO member bodies). The work of preparing International S
12、tandards is normally carried out through ISO technical committees. Each member body interested in a subject for which a technical committee has been established has the right to be represented on that committee. International organizations, governmental and non-governmental, in liaison with ISO, als
13、o take part in the work. ISO collaborates closely with the International Electrotechnical Commission (IEC) on all matters of electrotechnical standardization. International Standards are drafted in accordance with the rules given in the ISO/IEC Directives, Part 2. The main task of technical committe
14、es is to prepare International Standards. Draft International Standards adopted by the technical committees are circulated to the member bodies for voting. Publication as an International Standard requires approval by at least 75 % of the member bodies casting a vote. Attention is drawn to the possi
15、bility that some of the elements of this document may be the subject of patent rights. ISO shall not be held responsible for identifying any or all such patent rights. ISO 20462-1 was prepared by Technical Committee ISO/TC 42, Photography. ISO 20462 consists of the following parts, under the general
16、 title Photography Psychophysical experimental methods for estimating image quality: Part 1: Overview of psychophysical elements Part 2: Triplet comparison method Part 3: Quality ruler method BS ISO 20462-1:2005vIntroduction There are many circumstances under which it is desirable to quantify image
17、quality in a standardized fashion that facilitates interpretation of results within a given experiment and/or comparison of results between different experiments. Such information can be of value in assessing the performance of different capture or display devices, image processing algorithms, etc.
18、under various conditions. There are a number of psychometric methods described in the literature, such as paired comparison, rank ordering, categorical sort, and magnitude estimation, which might be considered as candidates for experimentally measuring image quality. Several textbooks1 3 4 5 9 12hav
19、e reviewed these and other methods and have discussed associated data reduction techniques, which usually are based upon the approach of Thurstone11or analogous reasoning. However, the choice of the best method for a particular application may be difficult to make, and interpretation of the rating s
20、cales produced by the numerical analyses is frequently ambiguous. Furthermore, none of the commonly used techniques provides an efficient mechanism for calibration of the results against a standardised numerical scale or associated physical references, which is desirable when results of different ex
21、periments are to be compared or integrated. The value of new calibrated psychometric methods in developing comprehensive models of imaging system quality has been demonstrated in a recent work6that contains more detailed discussions of many of the informative topics superficially considered herein.
22、The three parts of ISO 20462 address the need for documented means of determining image quality in a calibrated fashion. Part 1 provides an overview of practical psychophysics; specific experimental methods and associated data reduction techniques are described in Part 2 (triplet comparison method8
23、10) and Part 3 (quality ruler method6). Informative Annex A aids in identifying the better choice between the two alternative methods of Parts 2 to 3, which are complementary and together are sufficient to span a wide range of applications. It is the intent of these methods to produce results that a
24、re not merely directional in nature, but are expressed in terms of relative or fixed scales that are calibrated in just noticeable differences (JNDs), so that the significance of experimentally measured stimulus differences is readily ascertained. BS ISO 20462-1:2005blank1Photography Psychophysical
25、experimental methods for estimating image quality Part 1: Overview of psychophysical elements 1 Scope This part of ISO 20462 is part of a multiple-part standard pertaining to the subjective evaluation of pictorial still image quality. This part of ISO 20462 a) defines the units by which image qualit
26、y is quantified (just noticeable differences, or JNDs); b) describes the influence of stimulus properties, observer characteristics, and task instructions on results obtained from rating experiments; c) provides a flow chart for choosing the preferred psychophysical method for determining image qual
27、ity from among those defined in subsequent parts of ISO 20462. 2 Normative references The following referenced documents are indispensable for the application of this document. For dated references, only the edition cited applies. For undated references, the latest edition of the referenced document
28、 (including any amendments) applies. ISO 3664, Viewing conditions Graphic technology and photography 3 Terms and definitions For the purpose of this document, the following terms and definitions apply: 3.1 artefactual attribute attribute of image quality that, when evident in an image, nearly always
29、 leads to a loss of overall image quality EXAMPLE Examples of artefactual attributes include noise and aliasing. NOTE The commonly used terms defect and impairment are similar in meaning. 3.2 attribute aspect, dimension, or component of overall image quality cf. artefactual attribute (3.1) and prefe
30、rential attribute (3.12) BS ISO 20462-1:20052 EXAMPLE Examples of image quality attributes include image structure properties such as sharpness and noise; colour and tone reproduction properties such as contrast, colour balance, and relative colourfulness; and digital artefacts such as aliasing, con
31、touring, and compression defects. 3.3 attribute just noticeable difference attribute JND measure of the detectability of appearance variations, corresponding to a stimulus difference that leads to a 75:25 proportion of responses in a paired comparison task in which univariate stimuli pairs are asses
32、sed in terms of a single attribute identified in the instructions cf. quality JND (3.14) NOTE 1 As an example, a paired comparison identifying the sharper of two stimuli that differ only in their generating system modulation transfer function (MTF), would yield results in terms of sharpness attribut
33、e JNDs. If the MTF curves differed monotonically and did not cross, the outcome of the paired comparison would depend primarily upon the observers ability to detect changes in the appearance of the stimuli as a function of MTF variations, with little or no value judgement required of the observers.
34、The relationship between paired comparison proportions and stimulus differences is discussed in greater detail in Annex B. NOTE 2 If observers are instead asked to choose which of a pair of stimuli is higher in overall image quality, and if the stimuli in aggregate are multivariate, such that the ob
35、server should make value judgements of the importance of a number of attributes, rather than focussing on one aspect of image appearance, it is observed experimentally that larger objective stimulus differences (for example, MTF changes) are required to obtain a 75:25 proportion of responses, which
36、in this case corresponds to a quality JND. NOTE 3 A JND is a statistical quantity, derived from a number of observations. An observer assessing a single pair of images differing by one attribute JND is unlikely to be confident that he or she has detected the sample difference. A stimulus difference
37、of approximately three JNDs is usually needed for an observer of average sensitivity to feel reasonably certain of his or her assessment. 3.4 categorical sort method psychophysical method involving the classification of a stimulus into one of several ordered categories, at least some of which are id
38、entified by adjectives or phrases that describe different levels of image quality or attributes thereof NOTE The application of adjectival descriptors is strongly affected by the range of stimuli presented, so that it is difficult to compare the results of one categorical sort experiment to another.
39、 Range effects and the coarse quantization of categorical sort experiments also hinder conversion of the responses to JND units. Given these limitations, it is not possible to unambiguously map adjectival descriptors to JND units, but it is worth noting that in some experiments where a broad range o
40、f stimuli have been presented, the categories excellent, very good, good, fair, poor, and not worth keeping have been found to provide very roughly comparable intervals that average about six quality JNDs in width. 3.5 image quality impression of the overall merit or excellence of an image, as perce
41、ived by an observer neither associated with the act of photography, nor closely involved with the subject matter depicted NOTE The purpose of defining image quality in terms of third-party (uninvolved) observers is to eliminate sources of variability that arise from more idiosyncratic aspects of ima
42、ge perception and pertain to attributes outside the control of imaging system designers. 3.6 instructions set of directions given to the observer for performing the psychophysical evaluation task 3.7 just noticeable difference JND stimulus difference that leads to a 75:25 proportion of responses in
43、a paired comparison task cf. attribute JND (3.3) and quality JND (3.14) BS ISO 20462-1:200533.8 magnitude estimation method psychophysical method involving the assignment of a numerical value to each test stimulus that is proportional to image quality; typically, a reference stimulus with an assigne
44、d numerical value is present to anchor the rating scale NOTE The numerical scale resulting from a magnitude estimation experiment is usually assumed to constitute a ratio scale, which, ideally, is a scale in which a constant percentage change in value corresponds with one JND. In practice, modest de
45、viations from this behaviour occur, complicating the transformation of the rating scale into units of JNDs without inclusion of unidentified reference stimuli (having known quality) among the test stimuli. 3.9 multivariate describing a series of test or reference stimuli that vary in multiple attrib
46、utes of image quality 3.10 observer individual performing the subjective evaluation task in a psychophysical method 3.11 paired comparison method psychophysical method involving the choice of which of two simultaneously presented stimuli exhibits greater or lesser image quality or an attribute there
47、of, in accordance with a set of instructions given to the observer NOTE Two limitations of the paired comparison method are as follows. a) If all possible stimulus comparisons are done, as is usually the case, a large number of assessments are required for even modest numbers of experimental stimulu
48、s levels if N levels are to be studied, N(N 1)/2 paired comparisons are needed. b) If a stimulus difference exceeds approximately 1,5 JNDs, the magnitude of the stimulus difference cannot be directly estimated reliably because the response saturates as the proportions approach unanimity. However, if
49、 a series of stimuli having no large gaps are assessed, the differences between more widely separated stimuli may be deduced indirectly by summing smaller, reliably determined (unsaturated) stimulus differences. The standard methods for transformation of paired comparison data to an interval scale (a scale linearly related to JNDs) perform statistically optimized procedures for inferring the stimulus differences, but they may yield unreliable results when saturated responses are included i
