1、Information technology MPEG video technologies Part 1: Accuracy requirements for implementation of integer-output 8x8 inverse discrete cosine transform AMENDMENT 1: Software for integer IDCT accuracy testingAmendment 1:2009 (IDT) toNational Standard of CanadaCAN/CSA-ISO/IEC 23002-1-08(ISO/IEC 23002-
2、1:2006, IDT)NOT FOR RESALE.PUBLICATION NON DESTINE LA REVENTE. ISO/IEC 2013. CSA Group 2015. All rights reserved. Unauthorized reproduction is strictly prohibited.Technical Corrigendum1:2015(IDT) toAmendment1:2009(IDT) to CAN/CSA-ISO/IEC23002-1:08(ISO/IEC 23002-1:2006, IDT)National Standard of Canad
3、aTechnical Corrigendum1:2015(IDT) to Amendment1:2009(IDT) toCAN/CSA-ISO/IEC 23002-1:08Informationtechnology MPEG videotechnologiesPart 1: Accuracyrequirements for implementationof integer-output8x8 inversediscrete cosinetransform(ISO/IEC 23002-1:2006, IDT)Standards UpdateServiceTechnical Corrigendum
4、 1:2015 (IDT) toAmendment 1:2009 (IDT) to 23002-1:08January 2015Title: Information technology MPEG video technologies Part 1: Accuracy requirements forimplementationof integer-output 8x8 inverse discrete cosine transformTo register for e-mail notification about any updates to this publication go to
5、shop.csa.ca click on CSA Update ServiceThe List ID that you will need to register for updates to this publication is 2420149.If you require assistance, please e-mail techsupportcsagroup.org or call 416-747-2233.Visit CSA Groups policy on privacy at csagroup.org/legalto find out how we protect your p
6、ersonalinformation.INTERNATIONAL STANDARD ISO/IEC 23002-1:2006/Amd.1:2008TECHNICAL CORRIGENDUM 1 Published 2013-05-15 INTERNATIONAL ORGANIZATION FOR STANDARDIZATION ORGANISATION INTERNATIONALE DE NORMALISATIONINTERNATIONAL ELECTROTECHNICAL COMMISSION COMMISSION LECTROTECHNIQUE INTERNATIONALEInformat
7、ion technology MPEG video technologies Part 1: Accuracy requirements for implementation of integer-output 8x8 inverse discrete cosine transform AMENDMENT 1: Software for integer IDCT accuracy testing TECHNICAL CORRIGENDUM 1 Technologies de linformation Technologies vido MPEG Partie 1: Exigences dexa
8、ctitude pour limplmentation de la transformation cosinus inverse discrte de sortie du nombre entier 8x8 AMENDEMENT 1: Logiciel pour essai de prcision IDCT entier RECTIFICATIF TECHNIQUE 1 Technical Corrigendum 1 to ISO/IEC 23002-1:2006/Amd.1:2008 was prepared by Joint Technical Committee ISO/IEC JTC
9、1, Information technology, Subcommittee SC 29, Coding of audio, picture, multimedia and hypermedia information. Replace the original electronic attachment by the new version attached to this technical corrigendum. ICS 35.040 Ref. No. ISO/IEC 23002-1:2006/Amd.1:2008/Cor.1:2013(E) ISO/IEC 2013 All rig
10、hts reserved Technical Corrigendum 1:2015 (IDT) to Amendment 1:2009 (IDT) to CAN/CSA-ISO/IEC 23002-1:08Copyright NoticeThe Canadian adopg415 on of this Amendment to the Internag415 onal Standard contains informag415 on copyright protected by CSA Group. All rights reserved. No part of this adopted Am
11、endment may be reproduced in any form whatsoever without the prior permission of CSA Group. ISO/IEC material is reprinted with permission.Requests for permission to reproduce this adopted Amendment or parts thereof should be addressed to:Manager, SalesCSA Group5060 Spectrum Way, Suite 100Mississauga
12、, Ontario L4W 5N6Copyright violators will be prosecuted to the full extent of the law.CSA Standards Update ServiceAmendment 1:2009 toCAN/CSA-ISO/IEC 23002-1-08October 2009Title: Information technology MPEG video technologies Part 1: Accuracy requirements for implementation of integer-output 8x8 inve
13、rse discrete cosine transform AMENDMENT 1: Software for integer IDCT accuracy testingPagination: 7 pages (iii preliminary and 4 text)To register for e-mail notification about any updates to this publicationgo to www.ShopCSA.caclick on E-mail Services under MY ACCOUNTclick on CSA Standards Update Ser
14、viceThe List ID that you will need to register for updates to this publication is 2419164.If you require assistance, please e-mail techsupportcsa.ca or call 416-747-2233.Visit CSAs policy on privacy at www.csagroup.org/legal to find out how we protect your personal information.Reference numberISO/IE
15、C 23002-1:2006/Amd.1:2008(E)ISO/IEC 2008INTERNATIONAL STANDARD ISO/IEC23002-1First edition2006-12-01AMENDMENT 12008-06-01Corrected version2008-07-01Information technology MPEG video technologies Part 1: Accuracy requirements for implementation of integer-output 8x8 inverse discrete cosine transform
16、AMENDMENT 1: Software for integer IDCT accuracy testing Technologies de linformation Technologies vido MPEG Partie 1: Exigences dexactitude pour limplmentation de la transformation cosinus inverse discrte de sortie du nombre entier 8x8 AMENDEMENT 1: Logiciel pour essai de prcision IDCT entier ISO/IE
17、C 23002-1:2006/Amd.1:2008(E) PDF disclaimer This PDF file may contain embedded typefaces. In accordance with Adobes licensing policy, this file may be printed or viewed but shall not be edited unless the typefaces which are embedded are licensed to and installed on the computer performing the editin
18、g. In downloading this file, parties accept therein the responsibility of not infringing Adobes licensing policy. The ISO Central Secretariat accepts no liability in this area. Adobe is a trademark of Adobe Systems Incorporated. Details of the software products used to create this PDF file can be fo
19、und in the General Info relative to the file; the PDF-creation parameters were optimized for printing. Every care has been taken to ensure that the file is suitable for use by ISO member bodies. In the unlikely event that a problem relating to it is found, please inform the Central Secretariat at th
20、e address given below. COPYRIGHT PROTECTED DOCUMENT ISO/IEC 2008 All rights reserved. Unless otherwise specified, no part of this publication may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying and microfilm, without permission in writing from
21、either ISO at the address below or ISOs member body in the country of the requester. ISO copyright office Case postale 56 CH-1211 Geneva 20 Tel. + 41 22 749 01 11 Fax + 41 22 749 09 47 E-mail copyrightiso.org Web www.iso.org ii ISO/IEC 2008 All rights reservedAmendment 1:2009 toCAN/CSA-ISO/IEC 23002
22、-1-08ISO/IEC 23002-1:2006/Amd.1:2008(E) ISO/IEC 2008 All rights reserved iiiForeword ISO (the International Organization for Standardization) and IEC (the International Electrotechnical Commission) form the specialized system for worldwide standardization. National bodies that are members of ISO or
23、IEC participate in the development of International Standards through technical committees established by the respective organization to deal with particular fields of technical activity. ISO and IEC technical committees collaborate in fields of mutual interest. Other international organizations, go
24、vernmental and non-governmental, in liaison with ISO and IEC, also take part in the work. In the field of information technology, ISO and IEC have established a joint technical committee, ISO/IEC JTC 1. International Standards are drafted in accordance with the rules given in the ISO/IEC Directives,
25、 Part 2. The main task of the joint technical committee is to prepare International Standards. Draft International Standards adopted by the joint technical committee are circulated to national bodies for voting. Publication as an International Standard requires approval by at least 75 % of the natio
26、nal bodies casting a vote. Attention is drawn to the possibility that some of the elements of this document may be the subject of patent rights. ISO and IEC shall not be held responsible for identifying any or all such patent rights. Amendment 1 to ISO/IEC 23002-1:2006 was prepared by Joint Technica
27、l Committee ISO/IEC JTC 1, Information technology, Subcommittee SC 29, Coding of audio, picture, multimedia and hypermedia information. This corrected version incorporates the following corrections: reference to ISO 23002 now reads ISO/IEC 23002; reference to ISO/IEC 13818-2 now reads ITU-T Rec. H.2
28、62 | ISO/IEC 13818-2; the last paragraph of C.2 clarifies that the reference software package testbeds are for testing “drift effects”, as opposed to “drift”; changes to the presentation of C.3.3 to C.3.5, to improve legibility; changes to C.4 that provide additonal information about default test va
29、lues, specify to which clauses/subclauses of ISO/IEC 23002-1 the test modes relate, and remove the invertibilty test mode. Amendment 1:2009 toCAN/CSA-ISO/IEC 23002-1-08ISO/IEC 23002-1:2006/Amd.1:2008(E) ISO/IEC 2008 All rights reserved 1Information technology MPEG video technologies Part 1: Accuracy
30、 requirements for implementation of integer-output 8x8 inverse discrete cosine transform AMENDMENT 1: Software for integer IDCT accuracy testing Add the following annex: Annex C (normative) Reference software C.1 Introduction This annex specifies reference software that implements 8x8 IDCT precision
31、 tests as defined in this part of ISO/IEC 23002. Several additional tests (precision, dynamic range, linearity, and drift), implemented by the reference software are also specified. C.2 Reference software overview The reference software is implemented using the C programming language. It contains th
32、e following modules: 1. testbed.h header file to contain prototypes and lists of fixed-point IDCTs available for testing 2. testbed.c module implementing all tests specified in this part of ISO/IEC 23002 and containing main program 3. example_idct.c example fixed-point IDCT used for testing. The map
33、ping between the precision tests and functions in module testbed.c is provided in Table C.1. Amendment 1:2009 toCAN/CSA-ISO/IEC 23002-1-08ISO/IEC 23002-1:2006/Amd.1:2008(E) 2 ISO/IEC 2008 All rights reservedTable C.1 ISO/IEC 23002-1 testing procedures and corresponding functions in reference softwar
34、e Subclauses of ISO/IEC 23002-1 Procedure Corresponding function(s)in reference software 5.1 Ideal real-valued 8x8 forward DCT ref_double_fdct() 5.2 Ideal integer-valued 8x8 forward DCT ref_fdct() 5.3 Ideal real-valued 8x8 inverse DCT ref_double_idct() 5.4 Ideal integer-valued 8x8 inverse DCT ref_id
35、ct() 5.5.1.2 Process for pseudo-random number generation rand_gen() rand_gen_reset() 5.5.1.1 5.5.1.3 IDCT test procedure using pseudo-random input data idct_test() 5.5.2 Test of all-zero input behaviour zeroinput_test() A.2 Test of near-DC inversion behaviour neardc_test() 5.5.1 B.2 C.3.2 Pseudo-ran
36、dom data IDCT tests Additional pseudo-random input data tests Extended dynamic range pseudo-random input data tests run_tests(), idct_test() C.3.3 IDCT linearity test linearity_test() C.3.4 DCT pseudo-random input data test dct_test() C.3.5 DCT- IDCT pair pseudo-random input data test pr_test() In a
37、ddition, the reference software package includes ITU-T Rec.H.262ISO/IEC 13818-2 (MPEG-2 Video) and ISO/IEC 14496-2 (MPEG-4 Visual) based testbeds for testing the drift effects of fixed-point IDCT implementations. These testbeds are located in subdirectories MPEG2test and MPEG4 test, respectively. C.
38、3 Additional IDCT and DCT tests C.3.1 Introduction Subclauses C.3.2 to C.3.5 specify additional (informative) accuracy tests that are useful for testing of IDCT and DCT designs. C.3.2 Extended dynamic range pseudo-random input data tests For the sets of values of the variables L, H, and Q specified
39、below in this subclause, the tests using pseudo-random input data specified in 5.5.1.1 are performed using the pseudo-random number generator specified in 5.5.1.2. The desired accuracy for these additional tests using pseudo-random input data should be as specified in 5.5.1.3. The sets of values of
40、the variables L, H, and Q are specified as follows. L = 1 and H = 1 and Q = 1 000 000 L = 512*2B-8and H = 512*2B-8and Q = 1 000 000 L = 1805*2B-8and H = (1805*2B-8)-1 and Q = 1 000 000 Amendment 1:2009 toCAN/CSA-ISO/IEC 23002-1-08ISO/IEC 23002-1:2006/Amd.1:2008(E) ISO/IEC 2008 All rights reserved 3C
41、.3.3 IDCT linearity test For s = 07, t = 07 and z = 1, 3, 5 , 528*2B-8-1, the following test is performed. 1. For u = 07 and v = 07, set F zvu = 0 and Gzvu = 0. 2. Set F zst = z, and Gzst = -z. 3. Apply the IDCT under test to the 8x8 transform coefficient matrix F zto produce a corresponding integer
42、-valued 8x8 sample matrix fzfor testing. 4. For x = 07 and y = 07, set f zyx = clip3( -2B, (2B)-1, fzyx ). 5. Apply the IDCT under test to the 8x8 transform coefficient matrix G zto produce a corresponding integer-valued 8x8 sample matrix gzfor testing. 6. For x = 07 and y = 07, set gzyx = clip3( -2
43、B, (2B)-1, gzyx ). 7. For x = 07 and y = 07, set wyx = max( wyx, abs( f zyx - g zyx ) ). For x = 07 and y = 07, the resulting value of wyx, which may be referred to as the peak absolute error (PAE) for location yx for the linearity test, should be 0. C.3.4 DCT pseudo-random input data test For the p
44、urpose of testing precision of FDCT implementations, the process of 5.5.1.1 is employed with the following alteration. Instead of steps 3.c to 3.h specified in 5.5.1.1, the following steps are performed: c. For x = 07 and y = 07, set gzyx = Fzyx. d. Apply the FDCT under test to the integer-valued 8x
45、8 sample matrix fzto produce a corresponding integer-valued 8x8 transform coefficient matrix Gzfor testing. e. For u = 07 and v = 07, set Gzyx = clip3( -2B+3, (2B+3) -1, Gzyx ). f. For x = 07 and y = 07, set hzyx = Gzyx. g. For x = 07 and y = 07, set pyx = max( pyx, abs( hzyx - gzyx ) ). This proces
46、s is executed using the pseudo-random number generator specified in 5.5.1.2 for the following set of variables: L = 2Band H = (2B) -1 and Q = 1 000 000. C.3.5 Joint DCT-IDCT pair pseudo-random input data test For the purpose of testing precision of FDCT implementations, the process of 5.5.1.1 is emp
47、loyed with the following alteration. Instead of steps 3.b to 3.h specified in 5.5.1.1, the following steps are performed: b. For x = 07 and y = 07, set gzyx = fzyx. c. Apply the FDCT under test to the integer-valued 8x8 sample matrix fzto produce a corresponding integer-valued 8x8 transform coeffici
48、ent matrix Gzfor testing. d. Apply the IDCT under test to the integer-valued 8x8 transform coefficient matrix Gzto produce a corresponding integer-valued 8x8 sample matrix hzfor testing. e. For x = 07 and y = 07, set hzyx = clip3( -2B, (2B)-1, hzyx ). f. For x = 07 and y = 07, set pyx = max( pyx, abs( hzyx - gzyx ) ). This process is executed using the pseudo-random number generator specified in 5.5.1.2 for the sets of variables L, H, and Q specified in 5.5.1. A
