1、 Contents 1 Scope 2 General 3 Scanning 4 System colorimetry and transfer characteristics 5 Digital representation 6 Raster structure 7 Digital timing reference signals (SAV, EAV) 8 Ancillary data Annex A System colorimetry, transfer characteristics and luminance coding Annex B Analog representation
2、of the sync waveforms and video levels for the systems included in this standard Annex C Recommended filter characteristics Annex D Bibliography 1 Scope 1.1 This standard defines the digital representation of stationary or moving two-dimensional images for television production. The representation i
3、s sampled linearly in the spatial domain and sampled temporally at a constant frame rate. The scanned image has an aspect ratio of 16:9. The scanning format details are given in table 1. This standard includes both R,G,B and Y,CB,CRexpressions for the signal representation. 1.2 The principal applica
4、tion of this standard is for the production of content for EDTV-II, which employs an NTSC letterbox encoding scheme, compatible with SMPTE 170M. NOTES 1 RGB and YCBCRrefer to nonlinear gamma corrected signals derived from the linear components arising from the analysis of the image which are denoted
5、 as R,G,B, etc. 2 The signals CB,CRrefer to digital expressions of the color-difference signals. The signals PBPRrefer to the analog color-difference signals. Although this standard defines the digital representation, references to analog signal representations are given where considered appropriate
6、. 3 The primary purpose of this standard is to specify the parameters of the active area of the image. An additional set of parameters is included for real-time applications that consist of the frame rate, vertical and horizontal blanking, and total number of lines. The values of these parameters ar
7、e included in table 1. 4 The term line in this standard refers to the digital line in the 483-line frame raster as described in figure B.1. In the case of an analog video signal, the word analog line is used. The line numbering is based on the 525 digital line numbers as described in figures 1 and B
8、.1. Page 1 of 13 pages SMPTE 293M-2003 Revision of ANSI/SMPTE 293M-1996 Copyright 2003 by THE SOCIETY OF MOTION PICTURE AND TELEVISION ENGINEERS 595 W. Hartsdale Ave., White Plains, NY 10607 (914) 761-1100 Approved February 7, 2003 SMPTE STANDARD for Television 720 x 483 Active Line at 59.94-Hz Prog
9、ressive Scan Production Digital Representation SMPTE 293M-2003 Page 2 of 13 pages Table 1 Scanning system System nomenclature Samples per digital active line (S/AL) Lines per active image Frame rate Sampling frequency fs (MHz) Samples per total line Total lines per frame Colorimetry 720 x 483/59.94
10、720 483 60/1.001 27.0 858 525 SMPTE 170M 2 General Digital code word values in this standard are expressed as decimal values in the ten-bit representation. An eight-bit system shall round or truncate to the most significant eight bits according to provisions described in 5.10. 3 Scanning 3.1 Scannin
11、g parameters are specified in terms of the reference clock frequency indicated in table 1. The clock frequency shall be maintained to a tolerance of 3 ppm. 3.2 A frame shall comprise the total lines per frame indicated in table 1. Each line is of equal duration, determined by the sampling frequency
12、and the samples per total line (S/TL). Each line shall be scanned uniformly from left to right; lines in a frame shall be scanned uniformly from top to bottom. Lines are numbered in time sequence according to the raster structure described in annex B. 3.3 Each line shall be represented by the number
13、 of samples specified in table 1. These samples shall be equally spaced. Sample instants shall be defined with respect to a horizontal datum denoted by 0H, which is also the sync reference instant for the analog signal. The timing of sample instants with respect to 0His shown in figure B.1. 4 System
14、 colorimetry and transfer characteristics Signals in conformance with this standard shall be computed by the color analysis, optoelectronic transfer, scaling, and coding characteristics defined in annex A, unless indicated otherwise in an associated video index message (see SMPTE RP 186). NOTE The v
15、alues indicated in annex A are identical to those of SMPTE 170M for NTSC production. 5 Digital representation 5.1 Signals may be expressed as either R,G,B or Y,CBCRdigital component sets. 5.2 The digital signals described here are assumed to have been filtered to reduce or prevent aliasing upon samp
16、ling. Annex C provides additional information regarding such filtering. 5.3 R,G,B signals and the Y signals have a nominal bandwidth of 12 MHz. CBand CRsignals shall have a bandwidth nominally one-half that of the associated Y signal. 5.4 R,G,B signals and the Y signal of the Y,CBCRset shall be samp
17、led orthogonally, line and picture repetitive, at the reference clock frequency fs. The period of the sampling clock is denoted t = 1/fs. CBand CRsignals shall be horizontally subsampled by a factor of two with respect to the Y component. The CBand CRsamples are cosited with even-numbered Y samples
18、(sample 0, sample 2, sample 4, etc.) of each line. SMPTE 293M-2003 Page 3 of 13 pages 5.5 A sampling instant in a line is denoted in this standard by a sample number from zero through one less than the total number of samples in a line. Sample number 736 corresponds to the 0Hdatum. The sample number
19、ing is shown in figure 2 (for additional details, refer to figure B.1). 5.6 Sampled data at the interface shall be such that appropriate sin(x)/x correction occurs during conversion of the signal to the analog domain. 5.7 Digital R, G, B, and Y components shall be computer as follows: Ld= INT (219DL
20、 + 16D + 0.5); D = 2n-8where L is the component value (0 L 1), n takes the value 8 or 10 corresponding to the number of bits in the digital code word, and Ldis the resulting digital code. NOTE This scaling places the peak values of the R,G,B and Y components at code words 64 and 940 in a ten-bit rep
21、resentation or code words 16 and 235 in an eight-bit representation. 5.8 Digital CBand CRcomponents of the Y,CBCRset shall be computed as follows: Cd= INT (224DC + 128D + 0.5); D = 2n-8where C is the component value ( 0.5 C + 0.5), n takes the value 8 or 10 corresponding to the number of bits in the
22、 digital code word, and Cdis the resulting digital code. NOTE This scaling places the peak values of the CBand CRcomponents at code words 64 and 960 in a ten-bit representation or code words 16 and 240 in an eight-bit representation. 5.9 Code values having the eight most-significant bits all zeros o
23、r all ones that is, ten-bit codes 0, 1, 2, 3, 1020, 1021, 1022, and 1023 are employed for synchronizing purposes and shall be prohibited from video or ancillary data. 5.10 A system having an eight-bit code word precision may round video signals of 10-bit precision to eight bits and then discard the
24、two least significant bits. The two least significant bits of all other data across the interface shall be truncated without rounding. 5.11 For Y, R, G, and B signals, undershoot and overshoot in video processing may be accommodated by the use of code words 4 through 63 and code words 941 through 10
25、19 in a ten-bit system, or code words 1 through 15 and code words 236 through 254 in an eight-bit system. For CBand CRsignals, undershoot and overshoot in video processing may be accommodated by the use of code words 4 through 63 and code words 961 through 1019 in a ten-bit system, or code words 1 t
26、hrough 15 and code words 241 through 254 in an eight-bit system. 5.12 For Y, R, G, and B signals, the data words occurring during the digital blanking intervals, that are not used for timing reference codes EAV, SAV, or for ancillary data/signal, shall be filled with the sequence of code word 64 (fo
27、r 10-bit data) or 16 (for 8-bit data). For CBand CRcolor-difference signals, the code word shall be 512 (for 10-bit data) or 128 (for 8-bit data). 6 Raster structure 6.1 The details of the digital horizontal line are shown in figure 2. The figure also includes reference to the location of 0Hof the a
28、nalog horizontal sync. 6.2 The details of the vertical scanning structure are shown in figure 1. The figure also includes the line-numbering definition and details of digital vertical blanking. 6.3 The production aperture is defined by the number of active lines and active pixels as shown in table 1
29、. SMPTE 293M-2003 Page 4 of 13 pages 6.4 The clean aperture shall extend vertically from line 44 to line 523, inclusive. Horizontally, the clean aperture shall extend from sample 6 to sample 713, inclusive. Analog blanking edges and other edge processing artifacts should be confined to the area outs
30、ide the clear aperture. 6.5 The center of the picture is defined as the center of the clean aperture. Vertically, the center of the picture lies between line 283 and line 284. Horizontally, the center of the picture lies between sample 359 and sample 360. 6.6 Each active line has a 4 sample/byte pre
31、amble, start of active video (SAV), and a postamble end of active video EAV), which are described in clause 7. bottomoftheframetopoftheframebottomoftheframeDigital vertical blankingDigital verticalblankingActive imageTotal lines Bottom line # Top line # Line #1 Lines available for data 525 525 43 Ne
32、xt to bottom line 42 Figure 1 Vertical timing details specified in line numbers and the coding of EAV/SAV data SMPTE 293M-2003 Page 5 of 13 pages 10-bit 8bit960 240CB/CR512 12864 16940 235R,G,B and Y 64 16edcab f(sample #0)sample clocksadSAVEAVSAVEAVSAVEAVSAVEAVSample number for R, B and Y signals a
33、 b c d e f Total samples per line Start of EAV End of EAV Analog sync (0H) Start of SAV End of SAV Start of active line 858 720 723 736 854 857 0 Figure 2 Horizontal timing details specified in sample numbers SMPTE 293M-2003 Page 6 of 13 pages 7 Digital timing reference signals (SAV, EAV) 7.1 SAV (s
34、tart of active video) and EAV (end of active video) timing reference signals delineate the active video data and may be used for synchronization across the digital interface. Figures 1 and 2 show the relationship of the SAV and EAV sequences to digital video. The corresponding relationship to analog
35、 video is shown in annex B. 7.2 An SAV or EAV sequence shall comprise four consecutive code words: a code word of all ones, a code word of all zeros, another code word of all zeros, and a code word including F, V, H (field, vertical, horizontal) activity indicators, and P3, P2, P1, and P0 (parity) b
36、its. An SAV sequence shall be identified by having H = 0; EAV shall have H = 1. Tables 2 and 3 show details of the coding. 7.3 Each active line of video data (and, optionally, any blanked lines in the vertical interval) shall be preceded by a four-word SAV sequence and terminated by a four-word EAV
37、sequence. The EAV sequence immediately preceding the 0H datum of line 1 shall be considered to be the start of the digital frame. Table 2 Video timing reference codes Bit number 9 8 7 6 5 4 3 2 1 0 Word Value (MSB) (LSB) 0 1023 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 0 0 0 0 3 Se
38、e table 3 1 F V H P3 P2 P1 P0 0 0 Table 3 Word 3, data and protection bits for SAV and EAV Bit number 9 8 7 6 5 4 3 2 1 0 1 F V H P3 P2 P1 P0 0 Function Value Fixed Fixed Fixed SAV line numbers from 43 to 525 512 1 0 0 0 0 0 0 0 0 0 EAV line numbers from 43 to 525 628 1 0 0 1 1 1 0 1 0 0 SAV line nu
39、mbers from 1 to 42 684 1 0 1 0 1 0 1 1 0 0 EAV line numbers from 1 to 42 728 1 0 1 1 0 1 1 0 0 0 NOTE The value of F and V may be different in various interface standards. It will be assigned in corresponding interface documents. 8 Ancillary data 8.1 Ancillary data may be included optionally in the
40、blanking intervals of a digital interface according to this standard. 8.2 Any horizontal blanking interval between EAV and SAV may be employed to convey ancillary data packets. 8.3 The interval between SAV and EAV of any line that is outside the vertical extent of the picture, and that is not employ
41、ed to convey digitized ancillary signals, may be employed to convey ancillary data packets. SMPTE 293M-2003 Page 7 of 13 pages Annex A (normative) System colorimetry, transfer characteristics and luminance coding A.1 Picture information shall be represented by tristimulus (linear) red, green, and bl
42、ue (R,G,B) primary components, lying in the range 0 (reference black) to +1 (reference white). The primary and reference white chromaticities (in the CIE S002 system) are as follows: Chromaticities for primaries: x y Green 0.310 0.595 Blue 0.155 0.070 Red 0.630 0.340 Chromaticities for reference whi
43、te: x y White 0.3127 0.3290 A.2 From the linear red, green, and blue tristimulus values, three nonlinear primary components, R, G, and B, shall be computed according to the optoelectronic transfer function of SMPTE 170M, where L denotes a tristimulus value and V denotes a nonlinear primary signal: V
44、 = 4.5L, for 0 L 0.018 V = 1.099L0.45 0.099 for 0.018 L 1 A.3 The Y component shall be computed as a weighted sum of nonlinear R, G, and B primary components, according to: Y = 0.299R + 0.587G + 0.114B NOTES 1 The Y component is calculated in accordance with SMPTE 170M. 2 The Y component, being a we
45、ighted sum of nonlinear R,G,B components, has no simple relationship with the CIE luminance tristimulus component denoted Y used in color science. The choice of coefficients for Y has no influence on color reproduction per se, but does have a small effect on the magnitude of the small errors in repr
46、oduced luminance and colors that are caused by the bandwidth limitation of the color-difference components. A.4 Analog color-difference component signals PBand PRshall have the same 0.7 Vp-p excursion as the Y component (figure B.3) and shall be computed as follows: )Y (B 0.11410.5P B = )Y (R 0.2991
47、0.5P R = The digital components are coded as CBand CRcomponents for digital transmission, as defined in clause 5. A.5 To ensure the proper interchange of picture information between analog and digital representations, signal levels shall be completely contained in the range specified between referen
48、ce black and reference white specified in clause 5, except for overshoots and undershoots due to processing, as noted. Annex B (informative) Analog representation of the sync waveforms and video levels for the systems included in this standard This annex describes the relationship of the analog sync
49、hronizing signals to the digital data stream defined in this standard. The details are shown in figures B.1, B.2, and B.3. SMPTE 293M-2003 Page 8 of 13 pages Ancillary data (optional) Digital horizontal blanking A E analog line (n+1) analog line (n) analog line (n-1) 0 H datum Digital active line Analog waveform k a b c d EAV SAV EAV C B D k a b c d e e f g h i j digital line (n+1)
