MinDig TV



MinDig TV

Video Decoding

Implementation Guideline

for

Digital Television Receivers

for use in the

Hungarian Digital Terrestrial Television Broadcasting

Part of the MinDig TV Receiver Specification

Version 0.9.3

2009-11-17

Created by: Viktor Till

Company: Antenna Hungária

Contact: tillv@ahrt.hu

Table of contents

1. Introduction 3

1.1. Scope 3

1.2. Document History 3

1.3. References 4

1.4. Terminology 4

1.5. Definitions and Abbreviations 5

1.5.1. General Definitions 5

1.5.2. Definitions of the Present Document 5

1.5.3. Abbreviations 5

2. Overview of Video Decoding 6

3. Reference Model 7

3.1. SD Set-top-box Reference Model 7

3.2. HD Set-top-box Reference Model 7

3.3. Integrated Digital Television Reference Model 7

4. MPEG Video Decoding 8

4.1. General 8

4.2. Profiles and Levels 8

4.3. Video Bandwidth 8

4.4. Resolutions and Frame Rates 8

4.5. Video Format Signalling 9

4.5.1. ISO/IEC 13818-2 Sequence Header 9

4.5.2. ISO/IEC 14496-10 Sequence Parameter Set 10

4.6. Format Switching 11

4.7. Service Switching 11

4.8. Overscan 11

5. Active Format Pre-processing 12

5.1. General 12

5.2. Transmitted Signals 12

5.3. Related Viewer Preferences 12

5.4. Detailed Reference Model 13

5.5. Active Format Pre-processing Rules 14

5.6. Applying AFD 14

6. Decoder Format Conversion 16

6.1. General 16

6.2. Related Viewer Preferences 17

6.2.1. Aspect Ratio of the Targeted Display 17

6.2.2. Presentation Modes 17

6.2.3. Required Combinations of the Related Viewer Preferences 18

6.3. Applying the Rules 18

6.4. Conversions between Standard and High Definition Formats 19

6.4.1. Feeding the HD Output Interface 19

6.4.2. Feeding the SD Output Interface 19

7. Video Connections to the Display 20

7.1. Analogue Interfaces 20

7.1.1. SCART 20

7.1.2. CVBS 20

7.1.3. Y-Pb-Pr 20

7.2. HDMI 20

7.3. IDTV with Embedded Display 20

Annex A - Comprehensive Table 21

1. Introduction

1 Scope

The present document is part of the “MinDig TV Receiver Specification”.

It expands the descriptions of the “MinDig TV General Receiver Requirements” on video decoding and presentation issues.

To develop a MinDig TV compliant receiver, all the mandatory requirements of the present document shall be satisfied, while it is strongly recommended to follow its recommendations. Furthermore, a MinDig TV receiver shall satisfy all the parts of the “MinDig TV Receiver Specification”.

The MinDig TV Receiver Specifications does not require the implementation of the IEC 62216 “E-Book” [9], but on fields where there is no requirement described here, the IEC 62216 should be handled as a guideline.

2 Document History

|Version |Date |Comments |Editor |

|0.1.x |14.09.2009 |First public version |V. Till |

|0.9.x |16.11.2009 |Pre-release |V. Till |

3 References

[1] ISO/IEC 13818-1

[2] ISO/IEC 13818-2, "Information Technology - Generic Coding of Moving Pictures and Associated Audio; Part 2: Video"

[3] ISO/IEC 14496-10, "Information technology - Coding of audio-visual objects - Part 10: Advanced Video Coding"

[4] ETSI TS 101 154, "Digital Video Broadcasting (DVB); Implementation guidelines for the use of Video and Audio Coding in Broadcasting Applications based on the MPEG-2 Transport Stream"

[5] ETSI TR 101 211

[6] ETSI EN 300 294, “Television systems; 625-line television Wide Screen Signalling (WSS)”

[7] ETSI EN 301 775, “Digital Video Broadcasting (DVB); Specification for the carriage of Vertical Blanking Information (VBI) data in DVB bitstreams”

[8] HDMI v1.3a

[9] IEC 62216:2009 “E-Book”

4 Terminology

Shall (must, mandatory, required) These words mean that the item is mandatory.

Shall not (must not) These words mean that the item is forbidden.

Should (recommended, preferred) These words mean that the item is not mandatory, but is highly recommended.

MinDig TV receiver (receiver) The device is suitable for handling DVB-T services according to the requirements of the MinDig TV General Receiver Requirements and the related guidelines. The MinDig TV receiver may be an STB, IDTV or PC-based device or a part of the above.

5 Definitions and Abbreviations

1 General Definitions

See the “General Definitions” section of the “MinDig TV General Receiver Requirements” specification.

2 Definitions of the Present Document

3 Abbreviations

2. Overview of Video Decoding

Short summary of the chapters below.

3. Reference Model

Short description about the three types of reference model.

6 SD Set-top-box Reference Model

[pic]

Figure 1

7 HD Set-top-box Reference Model

[pic]

Figure 2

8 Integrated Digital Television Reference Model

[pic]

Figure 3

4. MPEG Video Decoding

9 General

A MinDig TV compliant receiver shall be able to decode MPEG-2 (ISO/IEC 13818-2 [2]) and H.264/AVC (ISO/IEC 14496-10 [3]) video streams constrained according to ETSI TS 101 154 [4]. The video decoder of the receiver shall comply at least with the following sections of the ETSI TS 101 154:

- ‘25 Hz MPEG-2 SDTV IRDs and Bitstreams’,

- ‘Specifications Common to all H.264/AVC IRDs and Bitstreams’,

- ‘H.264/AVC SDTV IRDs and Bitstreams’,

- ‘H.264/AVC HDTV IRDs and Bitstreams’

Furthermore, the video decoder of the receiver should comply with all the relevant sections of the IEC 62216 “E-Book” [9], according to the constraints above, unless the present document orders differently.

10 Profiles and Levels

The receiver shall support the decoding of MPEG-2 standard definition video services encoded according to MP@ML, H.264/AVC standard definition video services encoded according to MP@L3 or HP@L3, and H.264/AVC high definition video services encoded according to MP@L4 or HP@L4.

11 Video Bandwidth

The receiver shall properly handle an MPEG-2 SD video stream if its video rate is between 0.5 and 15 Mbit/s.

The receiver shall properly handle an H.264/AVC video stream if its video rate is between 250 kbit/s and 5 Mbit/s.

The receiver shall support the decoding and displaying of still pictures for all AVC profiles. To signal a video stream containing still pictures, the broadcast will set the AVC_still_present flag in the AVC descriptor, as specified in ISO/IEC 13818-1. The receiver shall decode the still picture frame and repeat displaying it until the subsequent (still picture) frame is available to display. The receiver shall properly handle AVC still pictures down to a bitrate of 100 kbit/s.

The receiver shall be able to continue outputting decoded video pictures undisturbed by dynamic changes of the video rate.

12 Resolutions and Frame Rates

In case of the MPEG-2 and AVC/H.264 standard definition video services, the receiver shall support the “full screen” luminance pixel resolutions of

- 720 x 576 - 50i,

- 704 x 576 - 50i,

- 544 x 576 - 50i,

- 480 x 576 - 50i,

- 352 x 576 - 50i,

- 352 x 288 - 50i

or “less than full screen” resolutions as described in TS 101 154.

In case of the H.264/AVC high definition video services, the receiver shall support the “full screen” luminance pixel resolutions of

- 1 920 × 1 080 - 50i / 25p,

- 1 440 × 1 080 - 50i / 25p,

- 1 280 × 1 080 - 50i / 25p,

- 960 × 1 080 - 50i / 25p,

- 1 280 × 720 - 50p,

- 960 × 720 - 50p,

- 640 × 720 - 50p

or “less than full screen” resolutions as described in TS 101 154.

The 1080 line video frame is broadcast in 1088 line encoded video. The receiver shall handle frame cropping signalling of the sequence parameter set that specifies which 8 lines shall be cut off. If the frame cropping information is not present in a video stream with 1088 encoded lines, the receiver shall discard the bottom 8 lines.

13 Video Format Signalling

1 ISO/IEC 13818-2 Sequence Header

The receiver shall be able to interpret the video format signalling of the video sequence header. According to the “25 Hz MPEG-2 SDTV IRDs and Bitstreams” clause of TS 101 154, the receiver shall be able to decode video services with values of aspect_ratio_information of "0010", "0011" and "0001", corresponding to 4:3 and 16:9 aspect ratio respectively. Handling of 2.21:1 aspect ratio and interpreting of sequence_display_extension and picture_display_extension are optional.

Table 1 shows the horizontal scaling factors required to restore the horizontal number of pixels to 720.

Note that, the table below is just a reference that shows which logical scaling factors shall be applied between the incoming “full screen” resolution frames (restored by the video decoder unit), and the picture presented on a display with a certain horizontal resolution. Depending on the applied active format pre-processing, selected presentation mode, and the resolution of the connected display; the receiver shall be able to release additional scaling methods by following the logic of this example.

|Decoded Picture |Displayed Picture |

| |Horizontal up-sampling |

|Luminance resolution |Source Aspect Ratio |4:3 display |16:9 display |

|(horizontal x vertical) | | | |

|720 × 576 |4:3 |× 1 |× 3/4 |

| |16:9 |× 4/3 (see note 1) |× 1 |

|544 × 576 |4:3 |× 4/3 |× 1 |

| |16:9 |× 16/9 (see note 1) |× 4/3 |

|480 × 576 |4:3 |× 3/2 |× 9/8 |

| |16:9 |× 2 (see note 1) |× 3/2 |

|352 × 576 |4:3 |× 2 |× 3/2 |

| |16:9 |× 8/3 (see note 1) |× 2 |

|352 × 288 |4:3 |× 2 |× 3/2 |

| |16:9 |× 8/3 (see note 1) |× 2 |

|NOTE 1: The up-sampling with this value is applied to the pixels in case of centre-cut-out presentation. |

Table 1

There must not be any up-scaling between 704 × 576 and 720 × 576. In this case, the 704 × 576 video shall be positioned into the centre of the 720 × 576 raster, while black blanking pixels shall fill the remaining left and right side 9 columns.

The interpretation of ETSI EN 300 468 component_descriptor is not required. It may not be broadcast in the bitstream. However, in the case when it is present but the contained data is not corresponding to the encoded video service, then it must not cause any impact on the video decoding.

2 ISO/IEC 14496-10 Sequence Parameter Set

The receiver shall be able to interpret the ISO/IEC 14496-10 sequence parameter set, and it shall be able to extract the relevant scaling factors from the VUI. According to the “H.264/AVC SDTV IRDs and Bitstreams” and “H.264/AVC HDTV IRDs and Bitstreams” clauses of TS 101 154, the receiver shall be able to decode 4:3 and 16:9 standard definition and 16:9 high definition video services.

Table 2 shows the horizontal scaling factors required to restore the horizontal number of pixels to 720, while Table 3 shows the horizontal scaling factors required to restore the horizontal number of pixels to 1920/1280.

Note that, the tables below are just references that shows which logical scaling factors shall be applied between the incoming “full screen” resolution frames (restored by the video decoder unit), and the picture presented on a display with certain horizontal resolutions. Depending on the applied active format pre-processing, selected presentation mode, and the resolution of the connected display; the receiver shall be able to release additional scaling methods by following the logic of this example.

|Decoded Picture |Displayed Picture |

| |Horizontal up sampling |

|Luminance resolution |Source Aspect Ratio |Aspect_ratio_idc |4:3 display |16:9 display |

|(horizontal x vertical) | | | | |

|544 × 576 |4:3 |4 |× 4/3 |× 1 |

| |16:9 |12 |× 16/9 (see note 1) |× 4/3 |

|480 × 576 |4:3 |10 |× 3/2 |× 9/8 |

| |16:9 |6 |× 2 (see note 1) |× 3/2 |

|352 × 576 |4:3 |6 |× 2 |× 3/2 |

| |16:9 |8 |× 8/3 (see note 1) |× 2 |

|352 × 288 |4:3 |2 |× 2 |× 3/2 |

| |16:9 |4 |× 8/3 (see note 1) |× 2 |

|NOTE 1: The up-sampling with this value is applied to the pixels in case of centre-cut-out presentation. |

Table 2

|Decoded Picture |Displayed Picture |

| |Horizontal up-sampling |

|Luminance resolution |Source Aspect Ratio |Aspect_ratio_idc |16:9 display |

|(horizontal x vertical) | | | |

|1 920 × 1 080 |16:9 |1 |× 1 |

|1 440 × 1 080 |16:9 |14 |× 4/3 |

|1 280 × 1 080 |16:9 |15 |× 3/2 |

|960 × 1 080 |16:9 |16 |× 2 |

|1 280 × 720 |16:9 |1 |× 1 |

|960 × 720 |16:9 |14 |× 4/3 |

|640 × 720 |16:9 |16 |× 2 |

Table 3

The interpretation of ETSI EN 300 468 component_descriptor is not required. It may not be broadcast in the bitstream. However, in the case when it is present but the contained data is not corresponding to the encoded video service, then it must not cause any impact on the video decoding.

14 Format Switching

The receivers shall be able to continue outputting decoded video pictures undisturbed by changes in the video format parameters provided that these changes are constrained as follows:

- changes are implemented at a sequence boundary (ISO/IEC 13818-2) or random access point (ISO/IEC 14496-10),

- vertical_size and display_vertical_size remain unchanged,

- the buffer models defined by MPEG are met,

- the field parity of the first displayed field of the new sequence complements that of the last displayed field of the preceding sequence (interlaced video only).

Specifically, this allows the following parameters to be changed:

- coded picture width (horizontal_size), and

- coded pixel aspect ratio (aspect_ratio_information, display_horizontal_size).

The receiver should minimize the impact of a format change to a few seconds (depending on random access points in the video).

15 Service Switching

After service switching, the video elementary stream which feeds the MPEG Video Decoder unit is broken, and when it is restored again, it may contain differently encoded video that was provided before. From the time, when correct video elementary stream is available again at its input, the decoder unit shall be able to recover the decoding process at latest in 1 second (typically in 0.5 second) by assuming that GOP length is 12. The picture on the display during the zapping time shall be either frozen or black.

16 Overscan

For services carrying ISO/IEC 14496-10 video, the broadcaster may use the overscan_info_present and overscan_appropriate flags in the VUI area to indicate whether the receiver should apply this typical overscan or should display the complete broadcast video image. The flags will be encoded according to Table 1.

|overscan_info_present_flag |overscan_appropriate_flag |Usage |

|0x0 or not broadcast |n/a |No preferred display method |

|0x1 |0x0 |Important information in entire |

| | |video region |

|0x1 |0x1 |Decoded picture suitable for |

| | |overscan |

Table 4

Unless the user requests otherwise, the receivers shall interpret and follow the overscan flags according to Table 2.

|overscan_info_present_flag |overscan_appropriate_flag |Usage |

|0x0 or not broadcast |n/a |Implementation dependent |

|0x1 |0x0 |Overscan not applied |

|0x1 |0x1 |Overscan applied |

Table 5

For receivers with HDMI output, the receiver shall pass the video without overscan related reformatting to its output, by setting the bits in the AVI Infoframe (see CEA 861) in accordance with Table 3.

|overscan_info_present_flag |overscan_appropriate_flag | |

| | |(in HDMI AVI Infoframe) |

|0x0 or not broadcast |n/a | |

|0x1 |0x0 | |

|0x1 |0x1 | |

Table 6

5. Active Format Pre-processing

17 General

At the output of the MPEG Video Decoder unit, the set of pixels of the decoded video frames and its full format aspect ratio are available. Based on the rules described by the tables of 4.5.1. and 4.5.2 clauses, the Decoder Format Converter unit shall render the output video signal according to the viewer preferences (Chapter 6).

However, certain parts of the decoded video frames contain blanking areas included at the content provider or operator sides which do not carry content “of interest”. In this case, additional signaling may be included in the broadcast to inform the receiver, or indirectly the display, about the portion of the 16:9 or 4:3 coded frame that is “of interest”. If proper signaling is present, then the receiver shall perform additional pre-processing on the decoded frames before passing them to the Decoder Format Converter unit.

The signalling mentioned above can be either the WSS (Wide Screen Signalling) included by the DVB bitstream [7] [1]; or AFD (Active Format Description) as defined in TS 101 154 [8], Annex B. If AFD is present, then its interpretation and usage is mandatory for the receiver, while the handling of WSS is optional, as secondary solution, in the case when AFD is not available, see Figure 4 and 5.

Note that, active format description is only additional information for the basic aspect ratio signalling system to announce the part “of interest” part of decoded frame. However, AFD or WSS itself does not define the aspect ratio of the encoded frame, as this is described elsewhere in the MPEG-2 or H264/AVC. Furthermore, active format description does not tell directly anything about the presentation mode of the content, as it is the role of the Decoder Format Converter based on the viewer preferences.

Instead, active format description provides detailed information about the interest part of the decoded frame which information, combined with the viewer preferences on the presentation mode, determines the necessary pre-processing method to be executed before the rendering process. As its result, the decoded frame may be cropped and the aspect ratio of the pre-processed frame may become different from its source.

18 Transmitted Signals

The active format pre-processing is done on the decoded video frames provided by the MPEG Vide Decoder. The aspect ratio of these frames is defined in accordance with the clauses 3.1. and 3.2. of TS 101 154, Annex B. These frames shall be handled as a composition of areas showed by Figure 4.

In case of MPEG-2 video services, AFD is carried in the video elementary steam at the picture level, as defined in TS 101 154, Annex B.6.

In case of H.264/AVC video services, AFD is carried in the video elementary stream as "User data registered by ITU-T Recommendation T.35 SEI message" syntactic element, as defined in TS 101 154, Annex B.7.

If AFD is not present then the embedded WSS (if present) can be used as secondary solution for describing the active format. [7]

19 Related Viewer Preferences

Active format information shall be interpreted and used in default mode. Although this feature can be disabled by the viewer, it is not recommended to offer this possibility.

The receiver should offer the possibility of changing the “shoot & protect” mode (see Table 7). However, this feature shall be disabled by default.

[pic]

Figure 4

20 Detailed Reference Model

If active format handling is enabled but AFD data is not present then the embedded WSS (if present) can be used as secondary solution. If AFD data is present, but the value of active_format_flag is ‘0’ or the value of AFD is ‘0’ or reserved then any action shall not be done by the Active Format Pre-processing unit.

[pic]

Figure 5

21 Active Format Pre-processing Rules

Basic rules of the required active format handling performed by the reference model receiver:

1. The blanking areas shall be cut off in each case.

2. If “shoot & protect” mode is enabled, then unprotected areas shall be cut off.

3. The unprotected areas shall be handled in adaptive mode that can override the second rule. It means that unprotected areas shall be cut off only in those cases when the presentation mode selected by the viewer (described in Chapter 6) determine that these areas will not be set back with blanking areas by the rendering process. Otherwise, unprotected areas shall not be cut off.

As a result of the active format pre-processing, inside the receiver, additional resolutions may be created to the ones defined by clause 4.4. The receiver shall be able to handle all these possible resolutions, and the appropriate scaling factors shall be applied.

For example, if a 16:9 frame encoded with 1440 pixel horizontal resolution contains 14:9 pillar-box image, then the pre-processed frame, after removing the blanking areas, will have 1440 x (14/16) = 1260 pixel horizontal resolution.

22 Applying AFD

Table 7 shows how the reference model receiver shall perform pre-processing on the incoming signals (decoded frame and its full format aspect ratio) based on the viewer preferences.

|Transmitted signals |Viewer preferences |Pre-processing |

|AFD |Decoded frame |Aspect ratio |Presentation |Shoot & |Aspect ratio of the |Pre-processed |

| |and its full format aspect|of the connected |mode |protect mode |pre-processed |frame |

| |ratio |display | | |frame | |

|‘1000’ | |x |x |x |4:3 | |

| |As the coded | | | |frame | |

| |4:3 frame | | | | |[pic] |

| | | | | | | |

| |[pic] | | | | | |

Table 7 – part 1

|‘1001’ |4:3 (centre) |x |x |x |4:3 | |

| | | | | |frame | |

| |in a 16:9 frame | | | | | |

| |[pic] | | | | | |

| | | | | | | |

| |in a 4:3 Frame | | | | |[pic] |

| |[pic] | | | | | |

|‘1010’ |16:9 (centre) |x |x |x |16:9 | |

| | | | | |frame | |

| |in a 16:9 frame | | | | | |

| |[pic] | | | | | |

| | | | | | | |

| |in a 4:3 frame | | | | |[pic] |

| |[pic] | | | | | |

|‘1011’ |14:9 (centre) |x |x |x |14:9 | |

| | | | | |frame | |

| |in a 16:9 frame | | | | | |

| |[pic] | | | | | |

| | | | | | | |

| |in a 4:3 frame | | | | |[pic] |

| |[pic] | | | | | |

|‘1101’ | |16:9 |x |enabled |14:9 |[pic] |

| |4:3 | | | |frame | |

| |(with shoot and protect | | | | | |

| |14:9 centre) | | | | | |

| | | | | | | |

| | | | | | | |

| |in a 16:9 frame | | | | | |

| |[pic] | | | | | |

| | | | | | | |

| | | | | | | |

| | | | | | | |

| |in a 4:3 frame | | | | | |

| |[pic] | | | | | |

| | |4:3 |letterbox |enabled |4:3 | |

| | | | | |frame |[pic] |

| | | | |disabled |16:9 | |

| | | | | |frame |[pic] |

| | | |centre-cut-out |x |16:9 | |

| | | | | |frame |[pic] |

Table 7 – part 3

6. Decoder Format Conversion

23 General

The Decoder Format Converter unit receives the set of pixels of the decoded frame and information about its full format aspect ratio that input may be modified by the active format pre-processing, as it is described in Chapter 4 and Chapter 5. The current clause describes how the Decoder Format Converter unit shall compose the output video frames from its input in accordance with the viewer preferences.

The receiver shall be able to drive at least 4:3 (optional in case of IDTV) and 16:9 type displays. The current clause describes the applicable presentation modes; while Chapter 7 defines which modes shall be applied in case of the different output interfaces.

In case of 4:3 displays, the receiver shall compose the final format of the output video signal, while in case of 16:9 displays, the receiver may rely on the format conversion capability of the targeted display unit.

Although the decoded frame can be either 4:3 or 16:9 type, the active format pre-processing may release 14:9 type frame, therefore, the Decoder Format Converter unit shall be able to handle the 14:9 format as well.

24 Related Viewer Preferences

1 Aspect Ratio of the Targeted Display

The receiver shall support at least the 4:3 and the 16:9 type displays. In case of IDTV, 4:3 may be missed.

2 Presentation Modes

The receiver shall support at least the following presentation modes (only pillar-box is required in case of IDTV):

- letterbox,

- centre-cut-out (or pan&scan, if implemented),

- pillar-box or widescreen.

Letterbox This mode is proposed to be used in case of 4:3 displays by those viewers who want to see every part of the decoded frame that carries content “of interest”. These viewers should tolerate the appearance of black blanking areas which may be part of the original content, or if it is necessary, which are inserted by the receiver. Before any letterbox conversion, line 23 and line 623 shall be masked to avoid the irritating half lines. This mode does not assume any display side format conversion.

Centre-cut-out This mode is proposed to be used in case of 4:3 displays as well by those viewers who do not want to see any blanking areas on the surface of their display unit, even if this mode loses some parts of content “of interest”. Therefore, in this case the receiver shall cut out the largest possible 4:3 rectangle from the centre of the real video content. This mode does not assume any display side format conversion.

Pillar-box This mode is proposed to be used in case of 16:9 displays. In this mode, the receiver shall compose the whole output video signal, with the necessary black blanking areas, that has the same (logical) aspect ratio than the targeted display panel. By this way, this video output shall be appropriate to continually fill out the whole surface of the targeted display unit, while any display side format conversion is not assumed.

Widescreen This mode is proposed to be used in case of 16:9 displays as well. In this mode, the receiver shall release the highest possible horizontal (and vertical) resolution on the output interface. Therefore, display side format conversion can be used when it is necessary.

In addition to the mandatory presentation modes, the receiver should support the full-frame and 14:9 modes as well.

Full-frame The output video signal is the full width and height of the decoded frame.

14:9 This mode is equal to the centre-cut-out mode, with the only difference, as in this case the largest possible 14:9 rectangle shall be cut out from the centre of the real video content.

Finally, the receiver is enabled to provide any other (proprietary) presentation mode in addition to the mandatory ones required by this document.

3 Required Combinations of the Related Viewer Preferences

At least the following combinations of the above mentioned viewer preferences shall be available for the viewer:

- 4:3 – letterbox ,

- 4:3 – centre-cut-out,

- 16:9 – pillar-box and/or 16:9 widescreen.

In case of IDTVs with 16:9 type embedded display unit, only the 16:9 – pillar-box mode is mandatory.

The main goal of the minimally required modes mentioned above is to always provide video presentation with correct aspect ratio on the surface of the targetted display unit. However, besides the mandatory presentation modes, any proprietary one is enabled, even if it provides anamorphic result.

25 Applying the Rules

|Receiver |Required output signals |Display device |

|Pre-processed frame |Aspect ratio of|Presentation |SCART pin8, line|Output video frame |Display format |Displayed video |

| |the targeted |mode |23 WSS |(may be transmitted in |conversion |signal |

| |display | | |anamorphic format) | |(always with correct |

| | | | | | |geometry) |

| |16:9 |pillar-box |6 V |[pic] |- | |

| | | | | |(note 1) | |

| | | |’1110’ | | | |

| | | | | | |[pic] |

| | | | | | | |

| | | | | | | |

|14:9 frame | | | | | | |

| | | | | | | |

|[pic] | | | | | | |

| | |center-cut-out|12 V |[pic] |- |[pic] |

| | | | | | | |

| | | |’0001’ | | | |

Table 8 – part 1

| |16:9 |pillar-box |6 V |[pic] |- |[pic] |

| | | | | |(note 1) | |

| | | |’1110’ | | | |

| | | | | | | |

| | | | | | | |

| | | | | | | |

|16:9 frame | | | | | | |

| | | | | | | |

|[pic] | | | | | | |

| | |center-cut-out|12 V |[pic] |- |[pic] |

| | | | | | | |

| | | |’0001’ | | | |

|Note 1: In case of standard definition output interface, where the resolution shall be 720 x 576 pixel, the video content is transmitted in |

|anamorphic format. Therefore, if the targeted display panel has higher resolution, then it will perform the necessary upscalling after it |

|recognized the SCART and/or VBI signaling. Otherwise, in case of high definition interface, the receiver should provide the video content in|

|that resolution which meets the horizontal and vertical resolutions of the targeted display as much as possible (see Chapter 7). |

| |

|Note 2: As an alternative, the receiver can use 14:9 letterbox format output as well. In this case, 12 V and ’1000’ signaling shall be |

|applied. |

Table 8 – part 2

26 Conversions between Standard and High Definition Formats

Outputting the selected video service in standard definition format is mandatory for every receiver, while outputting in high definition format (Y-Pb-Pr and/or HDMI) is optional.

The receiver shall support the conversions between any possible standard and high definition resolutions and vice versa.

1 Feeding the HD Output Interface

If HD output interface is available (embedded HD display, HDMI and/or Y-Pb-Pr interface – Figure 2 and 3), then the receiver shall up-scale the 704 pixel horizontal size window of the SD video (available at the output of the active format pre-processing unit) to fit the HD video output in 16:9 pillar-box mode.

In case of HD incoming services, the receiver shall be able to perform the necessary horizontal and vertical scaling between every possible HD format.

2 Feeding the SD Output Interface

In case of SD incoming services, the receiver shall be able to perform all the pre-processing and rendering methods described in Chapter 5 and Chapter 6, and it shall be able to output the result to the SD output interfaces.

If HD interface is not available as it is shown by Figure 1, then the output of the Decoder Video Converter can directly feed the output interfaces. Otherwise, Figure 2 shows the minimum required logical implementation.

The minimum required behavior when both SD and HD output interfaces are available is that the receiver shall be able to display the video signal prepared for HD presentation in letterbox mode on the SD interface.

Down-conversion of pictures shall be implemented from any of the HD full screen luminance resolution values (1920x1080, 1440x1080, 1280x1080, 960x1080, 1280x720, 960x720 and 640x720) to SD resolution (720x576).

When down-converting any 1:1 pixel aspect ratio format (i.e. 1280x720 or 1920x1080) to 720x576 resolution, the target shall be 702x576 pixels to be centered in the 720x576 grid, while black blanking pixels shall fill the remaining left and right side 9 columns.

7. Video Connections to the Display

27 Analogue Interfaces

A MinDig TV compliant receiver (except IDTV) shall contain at least either SCART (with RGB and CVBS) or CVBS (1x RCA) SD interface. In addition, it may contain HD or SD resolution Y-Pb-Pr interface as well.

It is important to note that the line 23 WSS signalling value is not the same that may be transmitted in the DVB bitstream [7]. This signalling shall be always generated by the receiver according to Table 8.

1 SCART

The 4:3 – letterbox, 4:3 – centre-cut-out and either 16:9 – pillar-box or 16:9 widescreen modes shall be supported on SCART. The SCART pin8 signalling shall be used according to Table 8. Furthermore, line 23 WSS signalling should be used according to Table 8 as well.

2 CVBS

The 4:3 – letterbox, 4:3 – centre-cut-out and 16:9 – pillar-box modes shall be supported on CVBS. 16:9 – pillar-box can be replaced by 16:9 widescreen mode only in that case if line 23 WSS is correctly implemented in accordance with Table 8.

3 Y-Pb-Pr

16:9 – pillar-box modes shall be supported on Y-Pb-Pr. It can be replaced by 16:9 widescreen mode only in that case if line 23 WSS is correctly implemented in accordance with Table 8.

28 HDMI

HDMI is an optional interface for a MinDig TV compliant receiver. If it is present, then it shall apply 16:9 – pillar-box presentation mode, and it should provide HD quality output.

The receiver shall rely on the (E-) EDID information returned by the display device - indicating the video formats the display can receive and render - to stay within its capabilities.

Aspect ratio signalling in the HDMI AVI Infoframe bits R0..R3, M0, M1 (see CEA 861) shall be set in accordance with the properties of the video on the output.

After the first HDMI connection between the receiver and a display, the receiver shall find the highest video quality supported by both devices. Then, it is strongly recommended for the receiver to continually render video output on HDMI scaled according to this video format, independently from any other condition, e.g. the resolution of the actually incoming video service. Thus, the automatic re-calibration process of the display, e.g. occurring at service switching, can be avoided. By this way, the zapping time can be reduced significantly.

29 IDTV with Embedded Display

A MinDig TV compliant IDTV shall apply at least a 16:9 – pillar-box compliant presentation mode.

Annex A - Comprehensive Table

The following table combines the Table 7 and 8 in order to give a comprehensive view on the required connections between the different kinds of incoming video signals and the way how these shall be presented on the surface of the targeted display.

|Transmitted signals |Viewer preferences |Result of AFD |Displayed video |

| | |pre-processing |signal |

| | | |(always with correct |

| | | |geometry) |

AFDDecoded frameAspect ratio of the connected displayPresentation modeShoot & protect mode‘1000’

As the coded

16:9 frame

[pic]16:9xx

16:9 frame

[pic][pic]4:3letterbox[pic]centre-cut-out[pic]‘1000’

As the coded

4:3 frame

[pic]16:9xx

4:3 frame

[pic][pic]4:3letterbox[pic]centre-cut-out[pic]‘1001’4:3 (centre)

in a 16:9 frame

[pic]

in a 4:3 Frame

[pic]16:9xx

4:3 frame

[pic][pic]4:3letterbox[pic]centre-cut-out[pic]Table 9 – part 1

‘1010’16:9 (centre)

in a 16:9 frame

[pic]

in a 4:3 frame

[pic]16:9xx

16:9 frame

[pic][pic]4:3letterbox[pic]centre-cut-out[pic]‘1011’14:9 (centre)

in a 16:9 frame

[pic]

in a 4:3 frame

[pic]16:9xx

14:9 frame

[pic][pic]4:3letterbox[pic]centre-cut-out[pic]‘1101’

4:3

(with shoot and protect 14:9 centre)

in a 16:9 frame

[pic]

in a 4:3 frame

[pic]16:9xenabled14:9 frame

[pic]

[pic]disabled4:3 frame

[pic]

[pic]4:3letterboxx

4:3 frame

[pic]

[pic]centre-cut-outx

4:3 frame

[pic]

[pic]Table 9 – part 2

‘1110’

16:9

(with shoot and protect 14:9 centre)

in a 16:9 frame

[pic]

in a 4:3 frame

[pic]16:9xx

16:9 frame

[pic]

[pic]4:3letterboxenabled14:9 frame

[pic]

[pic]disabled16:9 frame

[pic]

[pic]centre-cut-outx

16:9 frame

[pic]

[pic]‘1111’

16:9

(with shoot and protect 4:3 centre)

in a 16:9 frame

[pic]

in a 4:3 frame

[pic]16:9xx

16:9 frame

[pic]

[pic]4:3letterboxenabled4:3 frame

[pic]

[pic]disabled16:9 frame

[pic]

[pic]centre-cut-outx

16:9 frame

[pic]

[pic]Table 9 – part 3

-----------------------

In the context of the present document these areas are called „blanking areas”

In the context of the present document these areas are called „unprotected areas”

Black regions indicate areas of the picture that do not contain useful information

Grey regions that lie outside the smallest rectangle enclosing the white regions indicate areas of the picture that may be cropped by the receiver without significant loss to the viewer

Bounding box represents the whole decoded frame

This smallest rectangle enclosing the white regions indicates the area of essential picture information

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download