1、 Recommendation ITU-R BT.2087-0 (10/2015) Colour conversion from Recommendation ITU-R BT.709 to Recommendation ITU-R BT.2020 BT Series Broadcasting service (television) ii Rec. ITU-R BT.2087-0 Foreword The role of the Radiocommunication Sector is to ensure the rational, equitable, efficient and econ
2、omical use of the radio-frequency spectrum by all radiocommunication services, including satellite services, and carry out studies without limit of frequency range on the basis of which Recommendations are adopted. The regulatory and policy functions of the Radiocommunication Sector are performed by
3、 World and Regional Radiocommunication Conferences and Radiocommunication Assemblies supported by Study Groups. Policy on Intellectual Property Right (IPR) ITU-R policy on IPR is described in the Common Patent Policy for ITU-T/ITU-R/ISO/IEC referenced in Annex 1 of Resolution ITU-R 1. Forms to be us
4、ed for the submission of patent statements and licensing declarations by patent holders are available from http:/www.itu.int/ITU-R/go/patents/en where the Guidelines for Implementation of the Common Patent Policy for ITU-T/ITU-R/ISO/IEC and the ITU-R patent information database can also be found. Se
5、ries of ITU-R Recommendations (Also available online at http:/www.itu.int/publ/R-REC/en) Series Title BO Satellite delivery BR Recording for production, archival and play-out; film for television BS Broadcasting service (sound) BT Broadcasting service (television) F Fixed service M Mobile, radiodete
6、rmination, amateur and related satellite services P Radiowave propagation RA Radio astronomy RS Remote sensing systems S Fixed-satellite service SA Space applications and meteorology SF Frequency sharing and coordination between fixed-satellite and fixed service systems SM Spectrum management SNG Sa
7、tellite news gathering TF Time signals and frequency standards emissions V Vocabulary and related subjects Note: This ITU-R Recommendation was approved in English under the procedure detailed in Resolution ITU-R 1. Electronic Publication Geneva, 2015 ITU 2015 All rights reserved. No part of this pub
8、lication may be reproduced, by any means whatsoever, without written permission of ITU. Rec. ITU-R BT.2087-0 1 RECOMMENDATION ITU-R BT.2087-0 Colour conversion from Recommendation ITU-R BT.709 to Recommendation ITU-R BT.2020 (2015) Scope This Recommendation addresses a method of colour conversion fr
9、om Recommendation ITU-R BT.709 to Recommendation ITU-R BT.2020 for use when HDTV programme content is included within UHDTV programmes. Two sets of conversion equations are specified. One set is based on an opto-electronic transfer function (OETF) and its inverse. The other set is based on an electr
10、o-optical transfer function (EOTF) and its inverse. Keywords UHDTV, colour conversion The ITU Radiocommunication Assembly, considering a) that Recommendation ITU-R BT.2020 Parameter values for ultra-high definition television systems for production and international programme exchange, specifies the
11、 parameter values for the UHDTV image systems, and one of the features of UHDTV is its colour gamut wider than that of HDTV as specified in Recommendation ITU-R BT.709; b) that an increasing number of television broadcasters and programme makers around the world are starting to produce UHDTV program
12、mes; c) that HDTV programmes may well be used for making UHDTV programmes, which necessitates colour conversion from Recommendation ITU-R BT.709 to Recommendation ITU-R BT.2020; d) that it is required that colours of Recommendation ITU-R BT.709 content should be unchanged by the colour conversion to
13、 Recommendation ITU-R BT.2020 and that the conversion method should be mathematically definable, recommends 1 that when colour conversion from Recommendation ITU-R BT.709 to Recommendation ITU-R BT.2020 is required for UHDTV programme production and international exchange, the method described in An
14、nex 1 should be used. 2 Rec. ITU-R BT.2087-0 Annex 1 Method for colour conversion from Recommendation ITU-R BT.709 to Recommendation ITU-R BT.2020 Figure 1 shows a block diagram of the colour conversion from Recommendation ITU-R BT.709 (Rec. 709) to the non-constant luminance signal format in Table
15、4 of Recommendation ITU-R BT.2020 (Rec. 2020). The input and output of this diagram are digitally represented YCBCR signals or RGB signals. FIGURE 1 Block diagram of colour conversion from Rec. 709 YCBCR or RGB to Rec. 2020 YCBCR or RGB for the non-constant luminance signal format in Recommendation
16、ITU-R BT.2020 The functions and equations of each block in Fig. 1 are as follows: Inverse-quantisation of digitally represented luminance and colour-difference signals DYDCBDCR (Rec. 709) in the bit-depth of N709 bits to normalized luminance and colour-difference signals EYECBECR (Rec. 709): = ( 270
17、98 16) 219 = ( 27098 128) 224 = ( 27098 128) 224 Inverse-quantisation of digitally represented colour signals DRDGDB (Rec. 709) in the bit-depth of N709 bits to normalized colour signals EREGEB (Rec. 709): = ( 27098 16) 219 = ( 27098 16) 219 = ( 27098 16) 219 QRGB-1 QRGB 2020 709 QYC-1 M1 M2 M3 QYC
18、709 709 709 709 2020 2020 2020 2020 DYDCBDCR EYECBECR EREGEB EREGEB EREGEB EYECBECR DYDCBDCR EREGEB QYC-1 QRGB-1 DRDGDB DRDGDB Rec. ITU-R BT.2087-0 3 Conversion from normalized luminance and colour-difference signals EYECBECR (Rec. 709) to normalized RGB colour signals EREGEB (Rec. 709): = 1 0 1.574
19、71 0.1873 0.46821 1.8556 0 Non-linear to linear conversion from normalized RGB colour signals EREGEB (Rec. 709) to linearly represented, normalized RGB colour signals EREGEB (Rec. 709) is accomplished by one of two equations which produce slightly different colours from each other: Case #1: In the c
20、ase where the goal is to preserve colours seen on a Rec. 709 display1 when displayed on a Rec. 2020 display2, an approximation of the electro-optical transfer function (EOTF) from Recommendation ITU-R BT.1886 (Rec. 1886) is used: = ()2.40 , 0 1 Case #2: In the case where the source is a direct camer
21、a output and the goal is to match the colours of a direct Rec. 2020 camera output, an approximation of the Rec. 709 inverse opto-electronic transfer function (OETF) is used (see Annex 2): = ()2 , 0 1 NOTE 1: Recommendation ITU-R BT.1886 specifies the reference EOTF which is used to display Rec. 709
22、signals. This transfer function is expressed as L = a(max(V+b),0)2.40; where a =(LW1/2.40LB1/2.40)2.40 and b = LB1/2.40/(LW1/2.40LB1/2.40). The approximated, normalized form of this transfer function is shown in this document, which is found by setting LW = 1 and LB = 0. NOTE 2: The range of E or E
23、is defined within the range of 0 to 1 in Recommendation ITU-R BT.709. However, the definition of the video signal quantization allows values above 1 or below 0. The above equation may also be applied to those values above 1 or below 0. 1 A Rec. 709 display is a display device with RGB primaries that
24、 correspond to those in Recommendation ITU-R BT.709, a D65 white point, and an EOTF which conforms to Recommendation ITU-R BT.1886. 2 A Rec. 2020 display is a display device with RGB primaries that correspond to those in Recommendation ITU-R BT.2020, a D65 white point, and an EOTF which conforms to
25、Recommendation ITU-R BT.1886. M1 4 Rec. ITU-R BT.2087-0 Colour conversion from linearly represented, normalized RGB colour signals EREGEB (Rec. 709) to linearly represented, normalized RGB colour signals EREGEB (Rec. 2020): 2020= 0.6370 0.1446 0.16890.2627 0.6780 0.05930 0.0281 1.061010.4124 0.3576
26、0.18050.2126 0.7152 0.07220.0193 0.1192 0.9505709= 1.7167 0.3557 0.25340.6667 1.6165 0.01580.0176 0.0428 0.94210.4124 0.3576 0.18050.2126 0.7152 0.07220.0193 0.1192 0.9505709= 0.6274 0.3293 0.04330.0691 0.9195 0.01140.0164 0.0880 0.8956709Linear to non-linear conversion from linearly represented, no
27、rmalized RGB colour signals EREGEB (Rec. 2020) to normalized RGB colour signals EREGEB (Rec. 2020) is accomplished by applying the inverse of the non-linear to linear conversion equation. Case #1: In the cases where the goal is to preserve colours seen on a Rec. 709 display, an approximation of the
28、Rec. 1886 inverse EOTF is used: = 1 2 .40 , 0 1 Case #2: In the case where the source is a direct camera output and the goal is to match the colours of a direct Rec. 2020 camera output, an approximation of the Rec. 2020 OETF is used (see Annex 2): = 1 2 , 0 1 NOTE 3: The range of E or E is defined w
29、ithin the range of 0 to 1 in Recommendation ITU-R BT.2020. However, the definition of the video signal quantization allows values above 1 or below 0. The above equation may also be applied to those values above 1 or below 0. Conversion from normalized RGB colour signals EREGEB (Rec. 2020) to normali
30、zed luminance and colour-difference signals EYECBECR (Rec. 2020): = 0.2627 0.6780 0.05930.1396 0.3604 0.50000.5000 0.4598 0.0402 M2 M3 Rec. ITU-R BT.2087-0 5 Quantisation of normalized colour signals EREGEB (Rec. 2020) to digitally represented colour signals DRDGDB (Rec. 2020) in the bit-depth of N2
31、020 bits: = INT(219 + 16) 220208 = INT(219 + 16) 220208 = INT(219 + 16) 220208 Quantisation of normalized luminance and colour-difference signals EYECBECR (Rec. 2020) to digitally represented luminance and colour-difference signals DYDCBDCR (Rec. 2020) in the bit-depth of N2020 bits: = INT(219 + 16)
32、 220208 = INT(224 + 128) 220208 = INT(224 + 128) 220208. Figure 2 shows a block diagram for the colour conversion from Rec. 709 to the constant luminance signal format in Table 4 of Recommendation BT.2020. The input signals of this diagram are digitally represented RGB and YCBCR. And the output sign
33、als are digitally represented RGB and YCCBCCRC where the addition of the c subscript indicates the constant luminance signal format. FIGURE 2 Block diagram of colour conversion from Rec. 709 YCBCR or RGB to Rec. 2020 YCCBCCRC or RGB for the constant luminance signal format in Recommendation ITU-R BT
34、.2020 The functions and equations of each block in Fig. 2 are as follows: For the five blocks inside the black broken line, the same equations and input/output signals are applied as in the descriptions for Fig. 1. These blocks correspond to the conversion from the digitally represented luminance an
35、d colour-difference DYDCBDCR and colour DRDGDB signals (Rec.709) to the linearly represented, normalized RGB colour signals EREGEB (Rec. 2020). QRGB QYC M2 2020 EREGEB M4 EYC C EYCEREB QYcCc 2020 2020 EYCECBCECRC QRGB 2020 2020 QYC-1 M1 709 709 709 709 EYECBECR EREGEB EREGEB QRGB-1 709 DYDCBDCR DYCD
36、CBCDCRC DRDGDB DRDGDB 6 Rec. ITU-R BT.2087-0 For the M4 and C blocks in Fig. 2 (for the constant luminance signal format) are different compared with the blocks in Fig. 1 (for the non-constant luminance signal format). The same non-linear function and quantization equations are applied for , QYcCc a
37、nd QRGB blocks. To differentiate between the non-constant and constant signal format, the c subscript is added for the constant luminance signal format. Conversion from linearly represented, normalized RGB colour signals EREGEB (Rec. 2020) to normalized constant-luminance signal EYc (Rec. 2020): = 0
38、.2627 0.6780 0.0593 Linear to non-linear conversion from linearly represented, normalized RB colour signals EREB and normalized constant-luminance signal EYc (Rec. 2020) to non-linearly represented, normalized RB colour signals EREB and normalized constant-luminance signal EYc (Rec. 2020) is accompl
39、ished by applying the inverse of the non-linear to linear conversion equation. Case #1: In the case where the goal is to preserve colours seen on a Rec. 709 display when displayed on a Rec. 2020 display, an approximation of the Rec. 1886 inverse EOTF is used: = 1 2 .40 , 0 1 Case #2: In the case whe
40、re the source is a direct camera output and the goal is to match the colours of a direct Rec. 2020 camera output, an approximation of the Rec. 2020 OETF is used (see Annex 2): = 1 2 , 0 1 NOTE 4: The range of E or E is defined within the range of 0 to 1 in Recommendation ITU-R BT.2020. However, the
41、definition of the video signal quantization allows values above 1 or below 0. The above equation may also be applied to those values above 1 or below 0. Conversion from non-linearly represented, normalized RB colour signals EREB and normalized constant-luminance signal EYc (Rec. 2020) to normalized
42、colour-difference signals ECBcECRc (Rec. 2020): = 2 0.9702 ,0.9702 0 2 0.7910 ,0 0.7910 M4 C Rec. ITU-R BT.2087-0 7 = 2 0.8591 ,0.8591 0 2 0.4969 ,0 0.4969 Quantisation of normalized colour signals EREGEB (Rec. 2020) to digitally represented colour signals DRDGDB (Rec. 2020) in the bit-depth of N202
43、0 bits: = INT(219 + 16) 220208 = INT(219 + 16) 220208 = INT(219 + 16) 220208 Quantisation of normalized constant-luminance and colour-difference signals EYcECBcECRc (Rec. 2020) to digitally represented constant-luminance and colour-difference signals DYcDCBcDCRc (Rec. 2020) in the bit-depth of N2020
44、 bits: = INT(219 + 16) 220208 = INT(224 + 128) 220208 = INT(224 + 128) 220208 Annex 2 (informative) Non-linear transfer functions for colour conversion A concept of signal flow from scene light to display light in video systems is modelled as shown in Fig. 3, consisting of four functions: camera adj
45、ustments for creative rendering, opto-electronic transfer function (OETF), electro-optical transfer function (EOTF), and display adjustments to compensate for viewing environment. Camera adjustments include linear segment near black, pre-knee, knee point, knee slope, and other adjustments. The Rec.
46、709 and Rec. 2020 OETFs are similar to a square root function. The deviation of these OETFs from a 1/2.0-power function including the linear segment near black can be decomposed into the camera adjustment function. So the OETF itself can be regarded as a square root function. On the basis of this concept, the square function and square root function should be used for the conversion between linear and non-linear sig
