1、STD SMPTE RP LB?-ENGL 1995 = 8357LOL 000277b 5b5 RP 187-1 995 Center, Aspect Ratio and Blanking of Video Images 1 Scope Most video standards documents specify tolerances for individual picture blanking edges; center and aspect ratio are only implied by nominal blanking values, and vary as the blanki
2、ng edges move within permitted tolerances. Modern techniques of post- production require compositing of images generated by multiple devices (and sometimes originated in different video standards). These techniques demand a degree of precision not afforded by the implied definitions of center and as
3、pect ratio. This practice defines picture center and aspect ratio for a number of existing video standards, and provides an extensible technique that may be used to define center and aspect ratio for future standards. Test charts and test patterns are described which permit image generation and disp
4、lay devices to be calibrated to the recommended geometry. Recommendations are made for blanking widths at various stages of the production/post-production process. Additionally, a definition is provided for aset of recommended screen units permitting simple cross reference between standards with asp
5、ect ratios of 4 x 3 and 16 x 9. The practice is intended to be used for calibration of image generation and display devices. It is also intended as a reference for designers of such equip- ment (particuiatiy graphics devices), and for designers of processing equipment such as image- manipulation dev
6、ices. 2 Definitions The following paragraphs define terms and concepts as used in this practice: 2.1 aspect ratio: Aspect ratio is the ratio of hori- zontal dimension to vertical dimension of an image area, when displayed according to the Pagel of 12 pages specifications of the video standard. Withi
7、n this practice, the reference for an aspect ratio of 16 x 9 is the reference image lattice. The reference for an aspect ratio of 4 x 3 is any subset of the reference image lattice with 1440 pixels horizon- tally and 1080 pixels (lines) vertically. 2.2 cleanaperture: An image lattice that is a subse
8、t of the production aperture of a video standard, fixed in size and position, and nomi- nally cocentric with the production aperture. The clean aperture lies completely inside the picture area remaining after application of the widest blanking permitted for the standard. The dimen- sions of the clea
9、n aperture for a given video standard define the nominal aspect ratio for that standard. Where both composite and component versions of a standard exist, clean aperture is defined with respect to the image lattice speci- fied for the component version. 2.3 edge processing region: The region of the p
10、roduction aperture that lies outside the clean aperture. . Processing equipment should be designed so that all edge-related artifacts are confined to the edge processing region. Blanking transitions should occur completely within the edge processing region. 2.4 image latuce: A two-dimensional array
11、of pixels. 2.5 production aperture: The image lattice that represents the maximum possible image extent in a given standard. The production aperture represents the desirable extent for image acqui- sition, generation, and processing prior to blank- ing. Where both composite and component versions of
12、 a standard exist, production aperture is defined with respect to the image lattice speci- fied for the component version. Copyright Q 1995 by the SOCIETY OF MOTION PICTURE AND TELEVISION ENGINEERS 595 W. Hartsdale Ave., White Plains, NY 10607 (914) 761-1100 Approved December 6,1995 RP 187-1 995 2.6
13、 reference image lattice (RIL): An image lat- tice of square pixels with 1920 pixels horizon- tally and 1080 pixels (lines) vertically. 2.7 square pixel: A picture element that repre- sents an equal distance horizontally and verti- cally when the image is displayed at its defined aspect ratio. 3 Pro
14、duction and clean apertures This practice recognizes that the video signal is oper- ated on by many processing devices. Many of these operations (including most filters) produce edge- related artifacts. Blanking often is applied many times to the signal, and this usually results in distortion of the
15、 blanking edge or in an increase in blanking width. The result of this processing is that the useful image area is gradually reduced. It is, therefore, advanta- geous to acquire the maximum possible image area, and to define a region where artifacts are acceptable. The production aperture, as define
16、d above, is the maximum possible image area that can be supported by the video standard. The clean aperture is a slightly smaller image area, of similar aspect ratio to the production aperture, and contained within it. The part of the production aperture not included in the clean aperture is the edg
17、e processing region. Ideally, a video source will capture or generate the full production aperture. This may not always be practical, but source blanking should be as close as possible to this ideal. When these conditions are met, a test chart of the correct aspect ratio may be imaged and super- imp
18、osed on an electronic graticule representing the clean aperture, thus ensuring correct aspect ratio and centering. Although underscanned monitors are used for techni- cai monitoring, the ideal display for program viewing will display exactly the clean aperture. An electronic graticule representing t
19、he clean aperture may be used as a reference for display adjustment. It should be noted that it is not the intent of this practice that information outside the clean aperture be dis- carded. Some or all of this information may be blanked (horizontally and/or vertically), depending upon the transmiss
20、ion standards and operational practice. Subject to these limitations, this information should be retained and, to the extent permitted by transmission blanking standards, it should be transmitted. 4 Center of image The center of an image is defined as the point at the center of the clean aperture. H
21、orizontally, there will be an equal number of pixels within the clean aperture to the left and to the right of the center point. Vertically, there will be an equal number of pixels (lines) within the clean aperture above and below the center point. The center position for each video standard is spec
22、i- fied in the appropriate table of clause 11. 5 Horizontal blanking 5.1 Picture sources As described above, it is desirable to capture the maximum possible extent of picture information. Some authorities believe the optimum position for horizontal blanking is where the midpoint of the blanking rise
23、 and fall are coincident with the first and last pixels of the production aperture. Others prefer that the blanking edges be contained (just) within the production aperture. 52 In-plant blanking In-plant blanking should remain as narrow as possi- ble, and should always be narrowerthan transmission b
24、lanking. 5.3 Narrow transmission blanking This is the narrowest blanking permitted by a trans- mission standard and is provided for reference in appropriate tables of clause 11. This width of blanking may be very useful within a facii for signals that may be aired directly (without further blanking)
25、, but where it is desired to maintain as much information as pos- sible for further processing. This figure may represent a useful operating practice for blanking of taped signals that may be aired directly or that may be used as inputs for further processing. 5.4 Wide transmission blanking This fig
26、ure, quoted in appropriate tables of clause 11, represents the widest blanking permitted by a trans- mission standard and is placed just outside the clean aperture. It should be noted that once this blanking has been applied, further processing or blanking Will Page 2 o 12 pages STD-SMPTE RP 187-ENG
27、L 1775 8357403 0002778 338 RP 187-1995 likely result in distortion of the blanking edge andlor loss of information from the clean aperture. 6 Vertical blanking Vertical blanking is less frequently impacted by processing equipment than is horizontal blanking, but processes such as vertical filtering
28、will result in distortion at the top and bottom of the image. In general, the full height of the production aperture should be captured or generated at the source. A transmission standard may speciy an exact number of lines that should be transmitted. Where processing requires wider vertical blankin
29、g than determined by the production aperture, such blanking should remain outside the clean aperture. It should be noted that increasing the width of horizontal blanking but not of vertical blanking, as is common practice, results in an apparent change in aspect ratio. For this reason, both source a
30、nd display aspect ratios should be set using an electronic graticule repre- senting the clean aperture. 7 Screen units Many items of equipment such as picture manipula- tion devices make use of a system of screen units as part of the operator interface. In equipment designed for systems with an aspe
31、ct ratio of 4:3, it is common practice to define a grid 8 screen units horizontally by 6 screen units vertically. The nominal edges of the picture are then 4 units to the left and right of, and 3 units above and below, the center of the image. For 16:9 systems, a similar approach is recom- mended. T
32、he grid should be 32 screen units horizon- tally by 18 screen units vertically. The nominal edges of the picture are then 16 units to the left and right of, and 9 units above and below, the center of the image. The nominal edges of the picture referenced above should be the boundaries of the clean a
33、perture. The clean aperture will be exactly 32 x 18 screen units in 16:9 systems. The 4:3 subset of a 16:9 image will measure 24 x 18 screen units, which is exactly three times the current practice (8 x 6) for 4:3 systems. 8 Conversion ratios The tables in clause 11 contain horizontal and vertical c
34、onversion ratios to derive each scanning system from the reference image lattice. It is possible, there- fore, to derive conversion ratios between any two scanning systems. Ratios derived in this way should form the target ratios for standards converters. In practice, standards con- verters may use
35、filters with a relatively small number of taps. An appropriate balance of cost and video quality may mean that some practical converters will not use the exact ratios calculated from the tables. The tables provide a reference for calculating the resulting image size and/or aspect ratio errors. 9 Tes
36、t charts and signals Figure 1 shows a pattern that may be used for both physical test charts and electronic graticules, for both 4:3 and 16:9 applications. The outer rectangles rep- resent the clean apertures. (The markers represent 2.5%, 5%, 7.5%, and 10% overscan.) 10 Safe action and safe title ar
37、eas Safe action and safe title areas are defined in other SMPTE Recommended Practices as subsets of the image area. The definition of the image area of avideo picture used in deriving safe action and safe title areas should be the clean aperture for the appropriate video standard, as defined by this
38、 practice. 11 Tables of parameters Tables 1-6 define parameters for a number of tele- vision standards. The infomation includes the extent of the production and clean apertures, the picture center, and where appropriate, values for horizontal blanking as discussed in clause 5. All horizontal timings
39、 are expressed as fractions of the total line time r). STD-SMPTE RP 187-ENGL 1775 H 8357qOL 0002777 279 W RP 187-1995 Production aperture l- + +- + I O to 191 9 (reference) O to 1079 (reference) 1888 x 1062 Horizontal extent (pixel #to pixel #, inclusive) Verticai extent (line # to line #, inclusive
40、) Size hixels x lined 77- 11 Clean aperture +MM Horizontal extent 16 to 1903 Vertical extent 9 to 1070 Aspect ratio 16:9 (pixel # to pixel #, inclusive) (line #to line #, inclusive) Pixel aspect ratio i (horizontalivertical) i +MM Derivation from RIL - Horizontal Derivation from RIL - Vertical NOTE
41、- Markers represent overscan in increments of 2% picture widthheight. Figure 1 - Test chartlgraticule nia nia Table 1 - Reference image lattice (RIL) 1920 x 1080 Size (pixels x lines) 959 and 960 Horizontal (midway between pixel # and pixel #) I Center of image Vertical 1 (midwav between line # and
42、line #) 1 539 and 540 nia Horizontal position with respect to OH Page 4 of 12 pages STD.SMPTE RP LB7-ENGL 1775 B357LtOL 0002780 T7b RP 187-1995 Pixel aspect ratio (horizontal/vertical) Table 2 - 525/59.94,4:3 aspect ratio, 2:l interlace (ANSVSMPTE 125M) - 160 (approx. 0.904) 177 Production aperture
43、Derivation from RIL - Horizontal Derivation from RIL - Vertical Clean aperture 1 2 160 1 x-x- 177 2 x- Center of image Narrow Nominal Wide NTSC transmission blanking (50% points) pixel #717 to pixel #1 approximately 10.5 ps pixel #716 to pixel #2 approximately 10.7 ps pixel #714 to pixel #5 approxim
44、ately 11 .O ps Size (pixels x lines) 720 x 486 O to 719 Horizontal extent (pixel # to pixel #, inclusive) Vertical extent (line #to line #, inclusive) 283 (field Il) to 263 (field I) Size (pixels x lines) 708 x 480 Horizontal extent (pixel #to pixel #, inclusive) 6 to 713 22 (field I) to 524 (field
45、il) Vertical extent (line #to line #, inclusive) Aspect ratio 4:3 I Horizontal (midway between pixel # and pixel #) 359 and 360 404 (field Il) and 142 (field I) Vertical (midway between line # and line #) Horizontal position with respect to OH 963 321 1716 - 572 TH - aDDrox. 35.667 us Page 5 of 12 p
46、ages STD*SMPTE RP 187-ENGL 1995 8357YOL 0002783 922 Vertical extent (line # to line #, inclusive) RP 187-1995 283 (field il) to 263 (field I) Table 3 - 525/59.94,16:9 aspect ratio, 21 interlace, 13.5-MHz sampling (ANSVSMPTE 267M) I I Ho rizo nt al (midway between pixel #and pixel #) Vertical (midway
47、 between line#and line#) Centerof image Production aperture 359 and 360 404 (field Il) and 142 (field I) Size (pixels x lines) 720 x 486 O to 719 Horizontal extent (pixel # to pixel #, inclusive) Clean aperture Size (pixels x lines) 708 x 480 Horizontal extent (pixel # to pixel #, inclusive) 6 to 71
48、3 Vertical extent (line #to line #, inclusive) 22 (field I) to 524 (field Il) Aspect ratio 16:9 Pixel aspect ratio (horizontaVvertica1) - 640 (approx. 1.205) 531 Derivation from RIL - Horizontal 31 x4x2 Derivation from RIL - Vertical 160 1 177 5 x- Horizont qbosit ion with respectto0 963 321 1716 -
49、572 TH - approx. 35.667 ps Page6 d12 pages STD.SMPTE RP 187-ENGL 1775 8357403 0002782 b7 RP 17-1995 Table 4 - 525/59.94,16:9 aspect ratio, 2:l interlace, l&MHz sampling (ANSVSMPTE 2671111) Production aperture Clean aperture Center of image Size (pixels x lines) Horizontal extent (pixel # to pixel #, inclusive) Vertical extent (line #to line #, inclusive) Size (pixels x lines) Horizontal extent (pixel # to pixel #, inclusive) Vertical extent (line #to line #, inclusive) Aspect ratio Pixel aspect ratio (horizontalhertical) Derivation from RIL - Horizontal Derivation from R
