ITU-T J 81-1993 Transmission of Component-Coded Digital Television Signals for Contribution-Quality Applications at the Third Hierarchical Level of ITU-T Recommendation G 702 - Tel(Stu.pdf

《ITU-T J 81-1993 Transmission of Component-Coded Digital Television Signals for Contribution-Quality Applications at the Third Hierarchical Level of ITU-T Recommendation G 702 - Tel(Stu.pdf》由会员分享，可在线阅读，更多相关《ITU-T J 81-1993 Transmission of Component-Coded Digital Television Signals for Contribution-Quality Applications at the Third Hierarchical Level of ITU-T Recommendation G 702 - Tel(Stu.pdf（79页珍藏版）》请在麦多课文档分享上搜索。

1、. INTERNATIONAL TELECOMMUNICATION UNION ITU-T TELECOMMUNICATION STANDARDIZATION SECTOR OF ITU _.- J.81 (ex CMlT.723) (09/93) TELEVISION AND SOUND TRANSMISSION TRANSMISSION OF COMPONENT-CODED CONTRIBUTION-QUALITY APPLICATIONS OF ITU-T RECOMMENDATION G.702 DIGITAL TELEVISION SIGNALS FOR AT THE THIRD H

2、IERARCHICAL LEVEL ITU-T Recommendation J.81 (Formerly Recommendation ITU-R CMTT.723) 1TU-T RECMN*J98L 93 m 4862593 0609657 938 FOREWORD The IT Telecommunication Standardization Sector (-T) is a permanent organ of the International Telecom- munication Union. The ITU-T is responsible for studying tech

3、nical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. I The World Telecommunication Standardization Conference (WTSC), which meets every four years, established the topics for study by the ITU-T Study Groups wh

4、ich, in their turn, produce kecommendations on these topics. IT-T Recommendation 5.81 (formerly Recommendation ITU-R CMlT.723) was revised by the former ITU-R Study Group CMTT and was approved under the CCIR Resolution 97 procedure on 8 September 1993. See Note 1 below. NOTES 1 As a consequence of a

5、 reform process within the International Telecommunication Union (ITU), the CCIIT ceased to exist as of 28 February 1993. In its place, the ITU Telecommunication Standardization Sector (ITU-T) was created as of 1 March 1993. Similarly, in this reform process, the CCIR and the iFRB have been replaced

6、 by the Radiocommunication Sector (ITU-R). Conforming to a joint decision by the World Telecommunication Standardization Conference, Helsinki, March 1993, and the Radiocommunication Assembly, Geneva, November 1993, the ITU-R Study Group CM?T was transferred to ITU-T as Study Group 9, except for the

7、satellite news gathering (SNG) study area which was transferred to ITU-R Study Group 4. 2 telecommunication administration and a recognized operating agency. In this Recommendation, the expression “Administration” is used for conciseness to indicate both a O ITU 1995 All rights reserved. No part of

8、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 the ITU. ITU-T RECMN*J.BL 93 Im 4862593 O609658 879 CONTENTS Page Annex A . Digital coding of component television signals fo

9、r contribution-quality applications at about 34 or 45 Mbiifs A.l Scope A.2 Recommendations and standards A.3 Definitions, symbols and abbreviations A.4 Summary specification of component TV codecs for about 34 or 45 Mbiifs . AS Video coding and discrete cosine transformation (DCT) . A.6 Quantization

10、 of DCT coefficients . A.7 Variable length coding for DCT coefficients and motion vector differences . A.8 Video framing and forward error correction . A.9 Additional services . . A.10 Service multiplex A.l 1 Network adaptation . A.12 Scrambling for conditional access of transmitted data . Appendix1

11、 . I . 1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 Guidelines for implementation Introduction Mode choice Motion estimation . Truncation or rounding of coefficients . Buffer regulation . Error concealment . Refresh strategy Criticality Transmission of composite signals . Appendix II II.1 II.2 II.3 II . 4 II.5

12、 11.6 II.7 . Results of 34 Mbiifs codec tests . . Codec tested Tests carried out and test methods used Subjective tests . Results of subjective tests . Results of tests of codec recovery time . Additional observations Conclusions References . Recommendation J.81 (09/93) 1 1 2 2 3 3 10 17 22 27 36 40

13、 44 55 55 55 55 55 55 55 55 56 56 56 56 57 61 62 65 65 65 66 1 ITU-T RECMN*J-BA 93 m 4862593 0609657 700 M . Recommendation J.81) TRANSMISSION OF COMPONENT-CODED DIGITAL TELEVISION SIGNALS FOR CONTRIBUTION-QUALITY APPLICATIONS AT THE THIRD HIERARCHICAL LEVEL OF ITU-T RECOMMENDATION G.702 (1 990; rev

14、ised 1993) The ITU-R, considering (a) that for contribution-quality application, transmission should be based on component-coded digital video signals conforming with Recommendation IW-R BT.601; (b) that such transmission would satisfy user requirements for contribution-quality codecs at about 34 or

15、 45 Mbit/s, as specified by Radiocommunication Study Group 11; (cl that according to these user requirements such transmission should preserve the picture quality inherent to the 4:2:2 encoding process based on Recommendation ITU-R BT.601 to the maximum extent possible, taking into account the bit r

16、ate available to the user; (4 that such transmission should similarly preserve the down stream processing capabilities by maintaining the spatial and temporal resolution of the 4:2:2 signals as given by Recommendation IW-R BT.601; (e that additional transmission capacity should be provided for stere

17、o sound-channels, ancillary signals (e.g. Teletext, test signals) and accompanying error-protection data; (0 that through the use of appropriate bit-rate reduction techniques it is likely that these objectives can be reached at an acceptable level of complexity and cost for transmission at bit rates

18、 of about 34 or 45 Mbit/ and p: integer between O and 52. Tr(m) and Th(rn) are parameters depending on criticality (m) and are defined in Table A.4. 12 Recommendation 5-81 (09193) . Po (k, 1) O 1 2 3 4 5 6 7 . 12345678 - O03468811 O 12 3 6 8 913 2 2 3 4 7 91016 3 4 5 5 8101216 5 6 6 7 911 1317 8 7 9

19、 9 11 14 16 21 10 11 11 11 14 16 19 24 12 12 12 12 17 18 20 26 To9019BO-9UdO7 . .- ,v 12 FIGURE A.7lJ.81 Relative visibility matrix for chrominance * Nab 12 linear law 10 c C it I) + sign C(k,/) =2Z(k,/)lS(k,i,m,f) where S(k, i, m, f) = 2Nk*“* )Il6 n levels /8 2bits 8bts 6bits 1 bit Block Tranc- Coe

20、f- Y,C criticality mission ficien factor position FIGURE A.WJ.81 Overview of quantizer TO901 990-94/608 Recommendation J.81 (09/93) 13 . - Min (n, 48) (DC) P Min (2p - 48, f) visibility- threstwids 6 A A t f t t (Not specified) 2 bits 1 bit I I a Buffer occupancy Black FIGURE A.9tJ.81 Overview of ca

21、lculation of the scalar control parameter n(k, i, m, j) Y, c 14 Recommendation J.81 (09193) n Criticality Translation Tr (m) Limit for Y (ml Y or CR C, coefficients Th (ml O +8 No (i.e. 44 + 8) (i.e. 44 + 2) 1 +2 No 2 O 34 3 O 24 T9020i the second bit of each pair, and both the bits of the last pair

22、, if the word length is 18, is the information bit and can assume the value O or i. The length of the available words can vary from 2 to 18, as indicated in Table A.8, and the words having less than 18 bits end with a IO Xol pair. TABLE A.WJ.81 Available word lengths The total number of words is 153

23、4 of which 66 are used to code the run length of zeros, the special words EOBo and EOBI, indicating the end of each block (see A.8.1.4), and the word NULL (used in case of underflow). In order to avoid underflow, the zero values can be coded using the NULL word. In this case, a run length of n zeros

24、 can be coded with n NULL words. It is allowed to mix the NULL words and run length of zeros to obtain the desired length of n. The NULL word is considered as a normal value, so a “+i” must be transmitted between two NiJLL words. Thecodes 11 li 11 il 11 11 11 11 li and 10 i0 10 10 10 10 10 10 loarer

25、eserved. The remaining 1466 words are available to code the quantized levels from -733 to +733. Two assignment tables for the values of X (quantized levels) lower than 17 and greater than -17 and for the run lengths up to 28 zeros are used. In Table A.9 the value of luminance or chrominance coeffici

26、ents assigned to codewords differ in 20 cases, thus two different assignment tables are required to specify the luminance or the chrominance coefficients. The assignments of EOBo and EOB I are fixed for all the tables. 18 Recommendation J.81 (09/93) . TABLE A.9lJ.81 Assignment for luminance (Y) and

27、chrominance (C) coefficients Meword Y C Codeword Y C - 11 II II II II 11 11 11 11 11 11 11 11 11 11 11 11 11 11 11 11 11 11 11 11 11 11 11 11 11 11 - O0 O1 10 10 11 11 10 10 10 10 11 11 11 11 10 10 10 10 10 10 10 10 11 11 11 Il 11 11 11 11 - 30 31 DO DI 10 10 11 11 10 10 11 11 10 10 10 10 11 11 11 1

28、1 10 10 10 10 11 11 11 11 - 3 4 reserved, EOB,) 30 31 30 DI 30 31 DO o1 10 10 11 11 10 10 11 11 10 10 11 11 10 10 11 11 - 30 Dl DO o1 O0 o1 O0 o1 O0 o1 O0 O1 O0 o1 DO o1 - 0*5 0*7 6 7 8 0*13 0*15 0*17 1 2 o* 1 0*3 0*5 0*7 ( 3 4 0*9 0*11 5 0*13 0*15 0*17 6 7 8 0*19 0*21 0*23 0*25 0*27 9 10 11 12 13 1

29、4 15 i6 - 10 IO 10 IO IO IO IO 10 10 10 10 10 10 10 10 IO 10 10 10 IO 10 10 10 10 10 10 10 10 10 10 10 - )I 10 Il (1 10 IO 11 II II 11 10 10 10 10 11 Il 11 II il 11 11 11 10 10 10 10 10 10 10 10 - D O reserved. EOBo) 11 11 10 10 11 11 10 10 I1 11 11 11 10 10 10 10 11 11 11 II 10 10 10 10 _- - 31 30

30、31 DO 31 DO DI O0 11 11 10 10 11 11 10 10 11 11 10 10 11 11 10 10 - O1 O0 o1 O0 o1 O0 o1 O0 o1 O0 O1 O0 o1 O0 O1 O0 - -3 -4 0*6 0*8 -6 -7 -8 0*14 0*16 0*18 -1 -2 0*2 0*4 0*6 0*8 -3 -4 0*10 0*12 -5 O* 14 0*16 0*18 -6 -7 -8 0*20 0*22 0*24 0*26 0*28 -9 -10 -1 1 -12 -13 -14 -15 -16 NOTE - O*n refers to

31、a run-length of n zeros. Recommendation 5.81 (09/93) 19 The information Qts for the positive values and the corresponding negative values are complemented. SimilarIy, the information bits for EOBo and EOBI are also complemented. The assignments for run lengths of zeros from 29 up to 63 and for the c

32、odeword NULL are shown iiTable A.lO. Run of zeros 0*29 0*31 0*33 0*35 0*37 0*39 0*41 0*43 0*45 0*47 0*49 0*51 0*53 0*55 0*57 0*59 0*61 0*63 TABLE A.lO/J.Sl Assignments for run length of zeros 10 11 10 11 10 11 10 11 10 11 10 11 10 11 10 11 10 11 10 11 10 11 10 11 10 11 10 11 10 11 10 11 10 10 10 10

33、10 10 10 O0 10 10 10 o1 10 10 11 O0 10 10 11 o1 10 11 10 O0 10 11 10 o1 10 11 11 O0 10 11 11 o1 11 10 10 O0 11 10 10 o1 11 10 11 O0 11 10 11 o1 11 11 10 O0 11 11 io o1 11 11 11 O0 11 11 11 o1 10 10 10 O0 I Codeword h 11 O0 11 o1 10 O0 10 o1 11 O0 11 o1 10 O0 10 o1 11 O0 11 o1 10 O0 10 o1 11 O0 11 o1

34、 10 O0 I 1y ;: 11 11 11 11 11 11 11 11 10 10 10 10 10 10 10 10 11 11 11 11 I1 11 10 10 10 10 11 11 11 11 10 10 10 10 11 11 11 11 11 11 11 11 11 11 11 11 11 11 11 11 11 11 11 11 Run of zeros 0*30 0*32 0*34 0*36 0*38 0*40 0*42 0*44 0*46 0*48 0*50 0*52 0*54 0*56 0*58 0*60 0*62 Null 10 10 10 10 10 10 10

35、 10 10 10 10 10 10 10 10 10 11 11 I I NOTE - O*n refers to a run-length of n zeros. Examples of the use of run lengths and end of blocks are given below: Example I: Coefficients: -2 O00 +1 +1 O0 +2 O O O O End of block Codewords: 1001 111000 o1 o1 lo00 1100 101101 111101 Transmitted: 1001 111OOo o1

36、1000 1100 111101 +1 is not transmitted because there are two +Is between two runs of zeros, and a run of four zeros is not transmitted because the EOB is anticipated. Example 2: Coefficients: -2 000000000 +1 End of block Codewords: 1001 11 101 100 o1 111101 Transmitted : 1001 11 101 100 111101 +1 is

37、 not transmitted because it is between a run of nine zeros and the EOB. 20 Recommendation 5.81 (09193) The assignments foi values of quantized levels X are made in twos-complement using the following rules: Range of x rules * -733 . 479 Add +1501, then generate nine info bits, set continuation bit o

38、f last pair to one (see example 1). 478 . -17 Add -34, then generate up to nine info bits. not considering all the leading bits equal to one (see example 2). -16 . +16 Use Table A.9 for assignment. +17 . +478 Add +33, then generate up to nine info bits, not considering all the leading bits equal to

39、O (see example 3). +479 . +733 Add +34, then generate nine info bits, set continuation bit of last pair to one (see example 4). Example 1: from -733 to -479 X = -733 -733 + 1501 = 768 = (O01 100000000) = (il 111 Oll Oll Oll Oll Oll o11 Oll 0I)VLC ( x8 x7 x6 x5 and - if the motion vector difference (

40、MV, MV,) is different from (0,O). The corresponding macro-blocks are identified by MI = “10. A.8 Video framing and forward error correction A.8.l Video framing A single stream is produced for variable length coded (VLC) and fixed length coded (FLC) data. Ali data are transmitted MSB first. , A.8.1.1

41、 General structure Even fields are organized as follows: FSW O0 FCP BOF FSW O1 FCP BOF FSW 10 FCP BOF SSW SNo BOO TFYo TFCo (MI,CT,VLCj)o STUFFICRCo SSW SNi BOi TFY; TFCi (MI,CTjVLC,)i STUFFJCRCj SSW SNml BO,-1 TFYm-1 TFCm-1 (MI, CTj VLCj)m-i STUFFlCRCm-1 m = 36 for a 50 Hz system m = 31 for a 60 Hz

42、 system i varies from O to m - 1 for each stripe andj varies from 1 to 45 (position of the macro-block in the stripe). VLCj are VLC data for the jth macro-block. and have the form : MV,MV, VLCy1 EOB vLC,b EOB VLCy2 EOB VLC, EOB Odd fields are organized in a similar manner where i varies from m to 2m

43、-1. 22 Recommendation J.81 (09193) . O1 11 O0 11 o1 11 10 O0 11 10 o1 11 11 O0 (11 11 O1 reserved. 11 10 10 O0 11 10 10 o1 II 10 11 O0 11 10 11 o1 11 11 10 O0 11 11 10 o1 11 11 11 O0 11 11 11 o1 11 10 10 10 O0 11 10 10 10 o1 11 10 10 11 O0 11 10 10 Il o1 11 10 11 10 O0 11 10 11 10 o1 11 10 11 11 O0

44、11 10 11 11 o1 11 11 10 10 O0 11 11 10 10 o1 11 11 10 11 O0 11 II 10 11 o1 11 11 11 10 O0 11 11 11 10 o1 11 11 11 11 O0 11 11 11 11 o1 Il IO 10 10 10 O0 11 10 10 10 10 o1 11 10 10 10 11 O0 11 10 10 10 11 o1 11 10 10 11 10 O0 II 10 IO 11 10 o1 li 10 10 11 11 O0 11 10 10 11 11 o1 11 10 11 10 10 O0 11

45、10 11 10 10 o1 11 10 11 10 II O0 11 10 11 10 11 o1 11 10 11 11 10 O0 I 11 10 11 11 10 o1 II 10 11 11 11 O0 11 11 10 10 10 O0 11 11 10 10 10 o1 11 11 10 10 11 O0 11 11 10 IO 11 o1 II 11 10 11 10 O0 11 11 10 11 10 o1 11 11 10 11 11 O0 11 11 10 11 11 o1 11 11 11 10 10 O0 II 11 II 10 10 o1 11 11 11 10 1

46、1 O0 II 10 II Il 11 o1 Codeword I TABLE A.11IJ.81 Codewords for the motion vector differences I MV, or MVY O +os +1 +IS +2 +2.5 EOBI) +3 +3.5 +4 +4.5 +5 +5.5 +6 +6.5 +7 +7.5 +8 +8.5 +9 +9.5 +10 +los +I 1 +11.5 +I2 +12.5 +13 +13.5 +I4 +14.5 +I5 +15.5 +i6 +16.5 +I7 +17.5 +18 +18.5 +I9 +19.5 +20 +20.5

47、+21 +21.5 +22 +22.5 +23 +23.5 +24 +24.5 +25 +25.5 +26 +26.5 +27 +27.5 +28 Codeword O0 10 o1 10 O0 10 11 o1 10 11 O0 10 10 o1 110 10 Wreserved. 10 11 11 o1 10 11 11 O0 10 11 10 o1 10 11 10 O0 10 10 11 o1 10 10 11 O0 10 10 10 o1 “- IO lo lo O0 10 11 11 11 o1 10 11 11 11 O0 10 11 11 10 o1 10 11 11 10 O

48、0 10 11 10 11 o1 10 11 10 11 O0 10 11 10 10 o1 10 11 10 10 O0 10 10 11 11 o1 10 10 11 11 O0 10 10 11 10 o1 10 10 11 10 O0 10 10 10 11 o1 10 10 10 11 O0 10 10 10 10 o1 10 10 10 10 O0 10 11 11 11 11 o1 10 11 11 11 11 O0 10 11 11 11 10 O1 10 11 11 11 10 O0 10 11 11 10 11 o1 10 I1 II 10 11 O0 10 11 11 1

49、0 10 o1 10 11 11 10 10 O0 10 11 10 11 11 o1 10 11 10 11 11 O0 10 II i0 11 10 O1 10 11 10 11 10 O0 10 11 10 10 11 o1 10 11 10 10 11 O0 10 11 10 10 10 o1 10 11 10 10 10 O0 10 10 11 11 11 o1 10 10 11 11 11 O0 10 10 11 11 10 o1 10 10 11 11 10 O0 10 10 Il 10 11 o1 10 10 11 10 11 O0 10 10 11 10 10 o1 10 10 11 10 10 O0 10 10 10 11 11 o1 10 10 10 II II O0 10 10 10 11 10 o1 MV, or MVY -0.5 -1 -1.5 -2 -2.5 -3 EOBo) -3.5 4 -4.5 -5 -5.5 -6 -6.5 -7 -7.5 -8 -8.5 -9 -9.5 -10 -10.5 -I 1 -1 1.5 -12 -12.5 -1 3 -13.5 -14 -14.5 -1 5 -15.5 -1 6 -16.5 -