1、INTERNATIONAL TELECOMMUNICATION UNION ITU-T TELECOMMUNICATION STANDARDIZATION SECT0 R OF ITU P.930 (08196) SERIES P: TELEPHONE TRANSMISSION QUALITY Audiovisual quality in multimedia services Principles of a reference impairment system for video ITU-T Recommendation P.930 (Previously CCITT Recommenda
2、tion) ITU-T P-SERIES RECOMMENDATIONS TELEPHONE TRANSMISSION QUALITY Vocabulary and effects of transmission parameters on customer opinion of Series P.10 transmission quality Subscribers lines and sets Series P.30 Transmission standards Series P.40 Objective measuring apparatus Series P.50 Objective
3、electro-acoustical measurements Series P.60 Measurements related to speech loudness Series P.70 Methods for objective and subjective assessment of quality Series P.80 Audiovisual quality in multimedia services P.300 P.500 P.800 For further details, please refer to ITU-T List of Recommendations. ITU-
4、T RECOMMENDATION P.930 PRINCIPLES OF A REFERENCE IMPAIRMENT SYSTEM FOR VIDEO Summary This Recommendation describes the principles of an adjustable video reference system that can be used to generate the reference conditions necessary to characterize the subjective picture quality of video produced b
5、y compressed digital video systems. A Reference Impairment System for Video (RISV) can be utilized to simulate the impairments resulting from the compression of video sequences, independent of compression scheme. The subjective evaluation methods are described in Recommendation P.9 10. Appendix I de
6、scribes VIRIS (a Video Reference Impairment System developed by Bellcore), which is a specific implementation of an adjustable reference impairment system for video. Although the studies done at Bellcore were with MPEG-1, VIRIS can also be used with other compression schemes, such as H.261. It is re
7、commended that a RISV be capable of producing the following categories of distortions, either singly or in combinations, with independent adjustment of each impairment level: a b) c) In this Recommendation, five types of impairments (block distortion, blurring, edge busyness, noise, and jerkiness) a
8、re defined and general methods for implementing these impairments are provided. Appendix I describes a specific implementation of these impairments in VIRIS. Other impairments are the subject for future study. From the viewers point of view, the impairments produced by the RISV should be a good appr
9、oximation of impairments generated by digital video coding and transmission systems. Three possible applications for the RISV are: 1) 2) 3) Although this Recommendation describes the principles of an RISV, before an implementation can be recommended, validation tests are required. Artifacts due to c
10、onversions between analog and digital formats (e. g. noise and blurring). Artifacts due to coding and compression (e. g. jerkiness, edge busyness, and block dis tortion). Artifacts due to transmission channel errors (e. g. errored blocks). I creating reference conditions in subjective tests of digit
11、al video systems to ensure that the quality of the scenes presented to viewers covers the entire range of picture quality; defining standard video impairment levels that can be used to compare subjective test results; and quantifying the user-perceived quality of a video system with respect to a kno
12、wn reference. Source ITU-T Recommendation P.930 was prepared by U-T Study Group 12 (1993-1996) and was approved under the WTSC Resolution No. 1 procedure on the 30th of August 1996. STD.ITU-T RECHN P.930-ENGL 199b m VBb2591 Ob20748 477 FOREWORD ITU (International Telecommunication Union) is the Unit
13、ed Nations Specialized Agency in the field of telecommunications. The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of the U. The lTU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing t
14、elecommunications on a worldwide basis. The World Telecommunication Standardization Conference (WTSC), which meets every four years, establishes the topics for study by the ITU-T Study Groups which, in their turn, produce Recommendations on these topics. The approval of Recommendations by the Member
15、s of the ITU-T is covered by the procedure laid down in WTSC Resolution No. 1 (Helsinki, March 1-12, 1993). In some areas of information technology which fall within FU-Ts purview, the necessary standards are prepared on a collaborative basis with IS0 and IEC. NOTE I In this Recommendation, the expr
16、ession “Administration” is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. O ITU 1996 All rights reserved. No part of this publication may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopyi
17、ng and microfilm, without permission in writing from the ITU. 11 Recommendation P.930 (08/96) STD-ITU-T RECMN P-930-ENGL L97b W iBb2591 0b207ii9 323 W 1 2 3 4 4.1 4.2 5 5.1 5.2 5.3 5.4 5.5 5.6 6 CONTENTS Scope References Abbreviations and definitions Attributes of a Reference Impairment System for V
18、ideo (RISV) An adjustable range of video impairment levels Potential applications for a RISV Impairments produced by the RISV . Block distortion Blurring Edge busyness Noise Jerkiness . Combination of impairments . Conclusion . Appendix I . VIRIS. a specific implementation of a RISV Introduction t I
19、 . 1 1.2 Impairment simulation . 1.2.1 Block distortion 1.2.2 Blurring . 1.2.3 Edge busyness . 1.2.4 Noise . 1.2.5 Jerkiness Calculation of the Peak Signal-to-Noise Ratio (PSNR) 1.4 VIRIS programme VIRIS subjective test pian 1.5.1 VIRIS picture sequences . 1.5.2 VIRIS test conditions 1.5.3 VIRIS tes
20、t method 1.5.4 VIRIS viewing conditions 1.5.5 VIRIS test procedures . VIRIS subjective test results . 1.5.6.1 Data analysis . 1.5.6.2 VIRIS results 1.3 1.5 1.5.6 1 1 1 2 2 3 4 4 4 6 7 7 7 8 8 9 10 12 13 13 14 16 16 17 18 19 20 20 20 21 . 111 Recommendation P.930 (08/96) STD-ITU-T RECMN PaS3O-ENGL 19
21、9b m Li8b257L Ob20750 045 1.6 Edge busyness subjective test plan Edge busyness picture sequences . VIRIS test conditions and test material preparation . Edge busyness subjective test procedures Edge busyness subjective test results 1.6.1 1.6.2 1.6.3 1.6.4 1.6.5 Edge busyness subjective test method 1
22、.7 Conclusions Bibliography Page 24 25 25 26 26 27 28 29 iv Recommendation P.930 (08/96) . Introduction Reference video conditions are often used in subjective video quality evaluations because category rating judgements can vary from one test to the next or from one laboratory to another depending
23、on the experimental setting (i.e. range and frequency effects, subject population, etc.). The reference conditions allow for relative comparisons between the reference video conditions and the video conditions of interest and allow for subjective quality to be stated in terms of an objective measure
24、 of the reference. Including reference video conditions in a test allows for a wide range of test conditions to be presented and provides for a measure of video quality that can be more reliably replicated in another test or laboratory. In considering a system capable of producing such a set of adju
25、stable video reference conditions, several issues need to be addressed: 1) the type of video impairments or digital compression artifacts that should be simulated must be determined; 2) the appearance of the simulated video impairments should be similar to the actual impairments; 3) the means of gen
26、erating the impairments should be precisely defined so that they can be reproduced in any laboratory; and 4) the impairment level range should provide a wide range of picture quality performance. Video compression algorithms may introduce numerous impairment artifacts into the picture. Five of the m
27、ore common types are block distortion, blurring, edge busyness, jerkiness, and noise. These impairment artifacts are created in the compression stage and are generally due to loss of significant information by quantization, coding and other compression techniques. They are usually correlated with mo
28、tion in the picture. An adjustable video reference system should be rather simple to implement and should simulate the above five impairments similar to those produced by actual digital video encoders and independent of the compression scheme. The simulated impairment range should cover a wide range
29、 of picture quality for evaluating the picture quality of digital video coders. This requires that subjective tests be performed to determine the picture quality range provided by the simulated impairments. This Recommendation describes some of the impairments that should be produced by an adjustabl
30、e video reference impairment system and the general methods used to produce them. Appendix I contains a description of VIRIS, which is a specific implementation of reference impairment system for video for the evaluation of MPEG-1 digital video coders. Note that before VIRIS or any implementation of
31、 a RISV can be accepted by all Administrations, validation tests are required. This Recommendation reflects the current early status of research on reference impairment system for video. As progress on this work continues, understanding of these methods no doubt improve. As new knowledge is attained
32、, this Recommendation will be revised. 1. Recommendation P.930 (08/96) V - - STD-ITU-T RECMN Py3U-ENGL 1b m Li8b2591 0b207S3 854 Recommendation P.930 PRINCIPLES OF A REFERENCE IMPAIRMENT SYSTEM FOR VIDEO (Geneva, 1996) 1 Scope This Recommendation describes an adjustable video reference system that c
33、an be used to generate the reference conditions necessary to characterize the subjective picture quality of video produced by compressed digital video systems. A Reference Impairment System for Video (RISV) can be utilized to simulate the impairments resulting from the compression of video sequences
34、, independent of compression scheme. Subjective evaluation methods are not covered in this Recommendation but are instead described in Recommendation P.9 1 O. 2 References The following Recommendations and other references contain provisions which, through reference in this text, constitute provisio
35、ns of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; all users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendati
36、ons and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. - ITU-R Recommendation BT.500-6 (1994), Methods for the subjective assessment of the quality of television pictures. ITU-R Recommendation BT.601-4 (1 994), Encoding phameters of digital
37、 television for studios. KU-R Recommendation BT.802 (1992), Test pictures and sequences for subjective assessments of digital codecs conveying signals produced according to Recommendation 601. - - - U-T Recommendation P.920 ( 1996), Interactive test methods for audiovisual communications. FTU-T Reco
38、mmendation P.9 10 (1996), Subjective video quality assessment methods for multimedia applications. ITU-T Recommendation H.261 (1993), Video codec for audiovisual services atp x 64 kbit/s. - - 3 Abbreviations and definitions For the purposes of this Recommendation, the following abbreviations and def
39、initions are used. 3.1 pel (or pixel): A picture element that describes the brightness or colour of a discrete point in an image. 3.2 block: Group of pels. For example, a block of 8 x 8 pels is the smallest coding block used in MPEG-1 algorithms. There are 1320 such blocks in a SIF image, 4.4 in the
40、 horizontal direction (352 pels/8) and 30 in the vertical direction (240 lines/8). 3.3 block distortion (or tiling): Distortion of the image characterized by the appearance of an underlying block encoding structure. Recommendation P.930 (08196) 1 - STD-ITU-T RECMN P-730-ENGL 177b 9 48b2572 Ob20754 7
41、90 3.4 blurring: Global distortion of the entire image, characterized by reduced sharpness of edges and spatial detail. 3.5 colour errors: Distortion of all or a portion of the final image characterized by the appearance of unnatural or unexpected hues or saturation levels. These hues or saturation
42、levels were not present in the original image. 3.6 edge busyness: Distortion concentrated at the edges of objects, and further characterized by its temporal and spatial characteristics. 3.7 error blocks: A form of block distortion where one or more blocks in the image bear no resemblance to the curr
43、ent or previous scene and often contrast greatly with adjacent blocks. 3.8 jerkiness (or jerky motion): Motion that was originally smooth and continuous perceived as a series of distinct “snapshots“. 3.9 mosquito noise: Form of edge busyness distortion sometimes associated with movement, characteriz
44、ed by moving artifacts andor blotchy noise patterns superimposed over the objects (resembling a mosquito flying around a persons head and shoulders). 3.10 quantization noise: A “snow“ or “salt and pepper“ effect similar to a random noise process but not uniform over the image. 3.11 CIF: Common Inter
45、mediate Format used by H.261 coders, 352 luminance pels x 288 lines. 3.12 QCIF: Quarter CF, 176 luminance pels x 144 lines. 3.13 SIF: Source Input Format used by MPEG coders, a progressive, non-interlaced format of 352 luminance pels x 240 lines X 29.97 Hz or 352 luminance pels x 288 lines x 25 Hz.
46、3.14 CCIR format: U-R (formerly CCIR) digital video standard using interlaced formats of 720 luminance pels x 480 lines x 30 Hz and 720 luminance pels f From the viewers point of view, the impairments produced by the RISV should be a good approximation to the impairments generated by digital video c
47、oding and transmission systems. Next, five of these impairments are described in more detail, and Appendix I describes algorithms for producing these types of impairments. Attributes of a Reference Impairment System for Video (FUSV) conversions between analogue and digital formats (e.g. noise and bl
48、urring); coding and compression (e.g. jerkiness, edge busyness, and block distortion); transmission channel errors (eg. errored blocks). 4.1 The amount of distortion for each impairment should be independently adjustable. The ranges of adjustment should be sufficient to cover the full range of quali
49、ty for digital video coding and transmission systems. In addition, there should also be a means to combine two or more impairments and adjust the combined level of these impairments over a range that is useful for testing these systems. Both these topics are subject of continuing study, and subclause 5.6 provides additional guidance on these topics. An adjustable range of video impairment levels 2 Recommendation P.930 (08/96) STD-ITU-T RECMN P-730-ENGL 177b W 9Bb2571 Ob20755 b27 4.2 Three potential applications for the RISV are: 1) 2) 3) Standardization of t
