JP2000197047A - Method and device for processing digital signal - Google Patents

Abstract

PROBLEM TO BE SOLVED: To convert a transfer stream which includes plural individual video channels to a high definition television transfer stream through formation of a single encoded image by connecting respective selected elements and spatially and temporally arranged. SOLUTION: High definition television HDTV signals 200 are divided into respective elements in a signal splitter 201, and a digital data stream 202 is generated for the respective elements. It is encoded in a separate standard television encoder 203, a main profile is formed of a main level digital bit stream and sent to a standard multiplexer 204, and a standard multiplexed bit stream 205 is formed. Then, the bit stream 205 is changed into a high level HDTV suited stream in a postprocessor 206 to be suited to HDTV and transmitted via a transmission network 208 to a high definition decoder 209, and the data stream is decoded and high definition television images are displayed.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a digital television, and more particularly, to encoding of a high definition television (HDTV).

[0002]

2. Description of the Related Art A standard digital television picture has a resolution of 480.times.704 pixels and displays 30 frames per second. High definition television (HDTV) differs from standard television in that the resolution is much higher. Many different H
DTV standards have emerged, each of which has a different image resolution and / or frame rate. The table below shows some of the most common HDTV resolutions, including standard digital television for comparison.

[0003]

[Table 1]

[0004] Generating a digitally compressed HDTV signal requires a dedicated encoder for the high sampling rate used for HDTV signals. Therefore,
Due to its complexity, HDTV encoders are substantially more complex and more expensive than standard television encoders.

[0005]

Because HDTV images have high resolution and the data rates associated with large images are much larger and exceed the capabilities of standard television encoders,
Standard television encoders cannot encode HDTV images.

The applicant's co-pending UK patent application 9
8072066.9 describes a standard encoding device that can be used to encode HDTV images.
Therefore, HD encoding is mainly performed using a standard encoding device.
It is desirable to generate a TV data stream. It is an object of the invention to convert a transport stream containing a plurality of individual video channels into an HDTV transport stream.

[0007]

According to a first aspect of the invention, a plurality of individually encoded images are created to create a single encoded image that can be reconstructed to represent the image. An apparatus for processing elements, comprising a coordinator for selectively arranging each element spatially and temporally, and combining the selectively arranged elements to form a single coded image And a coupling for connection. According to a second aspect of the present invention, there is provided a method for processing a plurality of individually encoded image elements to create a single encoded image that can be reconstructed to represent the image. Selectively adjusting each element to spatially and temporally arrange each element, and combining the selectively arranged elements to form a single encoded image. A method is provided comprising:

[0008]

BRIEF DESCRIPTION OF THE DRAWINGS The invention will be described, by way of example, with reference to the accompanying drawings, in which: FIG. The table below shows some of the common HDTV resolutions and the number of standard television encoders required to encode a full HDTV image by that resolution.

[0009]

[Table 2]

FIG. 1 shows a high definition (high definition) television (HDTV) image 100 divided into a number of sub-elements 101, AF.
FIG.

FIG. 2 is a diagram showing an outline of the present invention. HD
The TV signal 200 is input to the signal splitter 201.
HDTV signal 200 is provided by an HDTV source such as a storage device or other means. The signal splitter 201 divides the HDTV image 100 into elements 101 and generates a separate digital data stream 202 for each element. In this example, the HDTV image is divided into six standard television images, as shown in FIG. In addition to splitting the signal, splitter 201 also inserts a series of time codes into the data stream. The time code is later used by the post-processor 206 to ensure time synchronization between the elements. The function of the signal splitter is
Details are set forth in Applicant's co-pending patent application 9807205.1. Each data stream 202 is encoded by a separate standard television encoder 203 to form a main profile with a main level (MP @ ML) digital bit stream. Element A of FIG. 1 is encoded by encoder A of FIG. 2, element B of FIG. 1 is encoded by encoder B of FIG. 2, and so on.

Next, the six individual standard television digital bit streams are sent to a standard multiplexer 204, which may include a statistical multiplexer (not shown). Multiplexer 204 multiplexes the six individual bit streams into one multiplexed bit stream 2
05 is formed. The multiplexed bit stream 205 is a standard multiplexed bit stream, and each bit stream is independent of one another at this stage. Inserting a standard receiver at this point allows any of the individual bit streams that are elements of the image 100 to be decoded.

[0013] To comply with HDTV, the multiplexed bit stream is processed to convert from a standard bit stream containing six MP @ ML streams to a high level (MP @ H
L) Must be converted to HDTV compatible stream. This conversion process according to the present invention is performed in post-processor 206. The converted data stream is then passed via transmission network 208 to high-definition decoder 2
09, this network is an artificial satellite,
Terrestrial, microwave and other transmission networks. Here, the data stream is decoded and the high definition television image is displayed on the high definition television.

FIG. 1 shows one line 105 of the HDTV image 100. This line is made up of several slices 102, 103 and 104. In the present invention, each individually encoded element represents a line of each element using a single slice, such as slice 102 in element A, but in other embodiments, multiple slices may be used. May be used.

[0015] The slices are preceded by a slice header, but the slices are combined to form a frame used in the encoding process. The frames are then usually organized into groups of twelve to form groups of pictures, ie, GOPs.
To form Each GOP starts with a GOP header,
The header has control data and includes the time code inserted by the signal splitter 201.

The post-processor 206 converts the six standard multiplexed television bit streams into a single H
It performs many functions to convert to a DTV bitstream.

FIG. 3 is an enlarged view of the post processor 206 of FIG. A multiplexed transport stream 205 including a plurality of standard television bit streams is input to resynchronization and repositioning unit 300, where the transport stream is processed to create input data for conversion to an HDTV data stream. Packets from individual video streams are distributed throughout the multiplexed transport stream as shown in FIG. 5a. Because data is transmitted serially, video data from one element arrives slightly before data from another element. Because the multiplexing is performed in this manner, the data is not necessarily continuous from any one image, so that the video packet for one element is
It may be several frames ahead or behind the video packet of another element. The resynchronizer alleviates this problem by resynchronizing the video data as it arrives and maintaining synchronization throughout the image. This process is described in detail below.

Another problem is that of spatial positioning. The input transport stream includes a plurality of independent video streams, each of which represents one element of a complete HDTV image as shown in FIG. In order to be displayed correctly by the HDTV decoder, the individual slices must be repositioned so that each slice is displayed in the correct position for image reconstruction.
In the example shown in FIG. 1, the HDTV line 105 must be correctly represented by slices 102, 103 and 104. Each slice must be properly positioned and synchronized in time so that each slice is from the same point in time. Otherwise, the different elements of the image will be offset in time from each other, and the boundaries of the individually encoded elements will be visible in the reconstructed image.

At this point, no individual slice has a horizontal or vertical offset. This is because each slice is usually represented by a standard television decoder and thus has no offset. The repositioner changes or adds an offset to each slice in the respective data stream so that a continuous and uniform image is displayed by the decoder. Returning to FIG. 1, slice 102 of element A does not require a horizontal offset because element A is in the normal display position. Slices 103 and 104 of elements B and C need to be added with a horizontal offset, respectively, so that slice 103 is displayed adjacent to slice 102 and slice 104 is displayed adjacent to slice 103. Each of the slices 102, 103 and 104 must be repositioned to form one continuous line of the original image 100.

When the video data has been repositioned and resynchronized, it is stored in memory 301. Next, the stream recombining unit 302 stores the processed data in the memory 30.
1 and recombines it to form an HDTV compatible transport stream. The recombination process is described in more detail below.

FIG. 4 is an enlarged view of the resynchronization / repositioning unit of FIG. A transport stream 205 including a plurality of video streams is provided to the stream stripper 400.
Is input to The stream stripper 400 separates the incoming transport stream into its constituent parts: video data, control data, and other data. Control data is stored in the control store 404 and includes data regarding image size, image format, and temporal order. The “other” data includes audio data and timing information and is stored in the packet store 403 in a corrected form, and the video data is further processed by the slice resynchronization unit 401 and the slice repositioning unit 402 to obtain the video store 405. Is stored. Packet store 403,
The control store 404 and the video store 405 are all part of the memory 301 shown in FIG.

FIG. 5a shows an example of a transport stream, which includes a number of video packets from each element AF and a number of audio packets, clock reference packets and stuffing packets. The data is stored in the packet store 403 as shown in FIG. When data is removed from the transport stream,
An integer indicating the number of packets removed from the stream is stored instead of the removed data. The transport stream is stripped in this way to separate all time-critical packets, such as audio and clock references, from video data that requires extensive processing to convert to an HDTV compliant data stream. This has the added advantage of preserving the time-critical order of the original data stream.

FIG. 7 shows details of the slice resynchronization unit 401. The input video data including only the video packet is transmitted to the video stream identification unit 700, the image group (G
OP) Header identification unit 701, time code extraction unit 703
And re-synchronization control section 702. The video stream identification unit identifies a video stream from which each video packet originates. The GOP header extractor analyzes the video stream and looks for a GOP header including a time code for the image group. The time code extraction unit 703 removes the time code from the GOP header,
This data is processed by the resynchronization control unit 702 to resynchronize the video data, and this is stored in the data store 405. The timecode is only included in the GOP header (thus only being added to the first frame of the GOP), and the individual timecode is derived from each subsequent frame of the GOP.

FIG. 6 is a diagram showing how the data store 405 is arranged. A number of slice stores 600 are arranged in the data store. Each slice store is indexed by a time code 601. The slice store 600 is further divided into individual slice stores for each of the elements A to F. Each element's slice store 602 is used to store all slices, all of which form one complete frame of element 101.

In this example, the HDTV image is divided into six elements, but there are six element slice stores 602 and the complete image 100 is stored. Thus, each slice store stores one complete image formed of multiple slices from each of the individual video streams. The reason for this is that, due to the nature of the distribution of the transport stream, the slices arrive asynchronously and some video streams are out of synchronization by several frames. By storing all slices for a given frame along with all other slices from other video streams for the same frame, the HD
Synchronization is restored when the TV data stream is finally formed.

Each of the slice stores 601, 603, and 6
An exponent is assigned by the time code identification unit such as 04. When the first time code is extracted by the time code extraction unit 703, each slice store 60
Each slice store 60 is initialized by initializing the timecode identifier so that 0 has an incremental timecode identifier.
0 is created. As each slice arrives, the time code is extracted and the slice is stored in the slice store 602 corresponding to each time code. Each slice is stored in the appropriate store depending on the element from which it originated. As more video packets arrive, each slice store is filled until it contains all slices for all elements for a given image. Video store 405 may store a number of complete images.

As a result of this process, all slices from all elements associated with the image are stored together, thereby mitigating the problems caused by the distribution of video packets in the multiplexed data stream. The stored data is eventually used as the stream recombining unit 302 in FIG.
Removed by As data is removed from the slice store, the timecode index 601 is updated with the next available timecode so that future slices can be stored. The process of filling and removing data from the slice store is a cyclic process.

Also, slice data may require additional processing to correctly spatially reposition within a full HDTV image. The slice repositioner can derive the required position of the slice in the full HDTV image by identifying to which element the slice belongs. Returning to FIG. 1, it can be seen that since the element A is in the normal display position, that is, on the upper left edge, the slice 102 of the element A does not need a horizontal offset. Slice 1 of elements B and C, such that slice 103 is displayed adjacent to slice 102 and slice 104 is displayed adjacent to slice 103
A horizontal offset needs to be added to 03 and 104 respectively. Also, the slices 102, 103, and 104 must be repositioned to form a single continuous line of the original image 100. Slices D, E and F from each element forming the lower half of image 100
Requires a vertical offset, and elements E and F require a horizontal offset.

The offset of each slice is changed by changing the slice header. To obtain the required horizontal offset, add the horizontal offset to the slice header by inserting additional bits. The vertical offset includes a change in the current offset. The repositioning is performed before storing the data in the slice store, or by post-processing resynchronization data already stored in the slice store.

The video data is stored in resynchronization order,
When the spatial positioning offset is changed, the data
You are ready to rejoin the HDTV compliant data stream.

FIG. 8 is a diagram of a system for recombining data stored in memory 301 to form an HDTV data stream. FIG. 8 shows a packet generator 800 for removing control data from the control store 404 and removing video data from the video store 405 to generate HDTV packets. Some of the control data, including data about image size, image format, and temporal order, may need to be changed, especially image size data that needs to reflect the size of the image. The control store also contains control data from individual data streams. Much of this data is duplicated in each data stream, so only a small portion of the stored control data is in the HDTV data stream.

The video data in the data store 405 is arranged as described above. Thus, the packet generator can extract the video data in a synchronized manner and create packets that remain synchronized, unlike a normal multiplexed data stream that is constructed in a distributed manner. The decoder buffer management unit 801 models the buffer of the decoder, and adds padding data to the video data as necessary so that the decoder buffer does not underflow or overflow. Packet insertion unit 8
02 takes the generated packets and inserts these packets into the stripped stream previously stored in the packet store 403 by the stream stripper 400 described above. The stripped data stream stored in the packet store 403 indicates how many packets have been removed from the stream during the stripping process. This causes the packet inserter to reinsert the same number of packets as originally removed into various portions of the stripped pit stream. The advantage of inserting the same number of packets that have been removed is that the position of important packets, such as voice and timecode, is unchanged in the bitstream. As a result, lip synchronization (synchronization of audio and image) is maintained. Another solution is to remove all data from the data stream and reconstruct an entirely new data stream from scratch, but this is much more complex and preserves all important time packet relationships. Requires much more processing power to do.

The repositioning of the slices and the stream reassociation produce additional data which must be included in the HDTV data stream. A convenient way to achieve this is to add multiplexer stuffing packets to the original multiplexed transport stream.
Is to adapt. Since the stuffing packets are removed by the stream stripper 400, additional space is created in the stripped stream, where extra data can be inserted. The present invention further simplifies the present invention because it eliminates the need to process significant time packet locations, such as voice and time codes, which are handled by the multiplexer. Thus, there is no duplication of functions. The amount of extra stuffing packets inserted by the multiplexer is
It can be calculated in advance so that sufficient space is secured.

The effect of the reinsertion process is more clearly shown in FIG. The original transport stream, including the six MP @ ML video streams, is processed and recombined to form a single MP @ HL transport stream. The position of the audio and time code in the data stream is the same as in the original multiplexed data stream of FIG. 5a. In FIG. 5c, instead of video packets from individual elements, only HDTV video packets are included and an HDTV compliant data stream is shown.

It should be noted that in the present invention, the above-described process is performed in real time with very little delay despite performing complicated processing. One major advantage of the present invention is that it provides broadcasters with flexibility. Generates a standard television bitstream using a number of standard encoders, primarily by using standard equipment, or using the invention to
You can choose to generate a TV bitstream. This allows the broadcaster to generate both standard and high definition television streams without the need for dedicated and costly high definition television encoders. With the same settings, even a low-resolution HDTV can encode both an HDTV and many SDTV channels using the same device. On the broadcast station side, for example, H
In order to broadcast a movie on a DTV and later broadcast news and simultaneous programs, the use of the broadcasting system can be changed by switching to broadcasting on several SDTV channels.

[Brief description of the drawings]

FIG. 1 is a diagram showing a high-definition television image divided into a number of elements.

FIG. 2 is a diagram showing an outline of a system according to the present invention.

FIG. 3 shows a post processor 2 included in the system of FIG. 2;
It is a figure which shows the outline of 06.

FIG. 4 is a diagram illustrating an embodiment of a resynchronization unit and a repositioning unit included in the system of FIG. 3;

FIG. 5 shows a number of packet structures of a digital data stream.

FIG. 6 is a detailed diagram of an example of the video store (storage unit) in FIG. 4;

FIG. 7 is a diagram illustrating an embodiment of a slice resynchronization unit included in the system of FIG. 4;

FIG. 8 is a diagram showing an embodiment of a stream recombining unit included in the system of FIG. 3;

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Kevin Alistair Marie UK, S.O.E.510 J.R., Hunts, Romsey, Rotzkerleigh, Holbery Lane, Markway 4 (72) Inventor Anthony Richard Hadget UK, S.O.50, 5 Pizzette, Hampshire Eastleigh, Market Street 205 (72) Inventor John Stuart Funnel UK, S.O.15 2T Southampton, Northlands Road, Court Royal Mews 9 (72) Inventor Richard John Warburton, S.O.30 2 R.P., Southampton Hedgeend, Britannia Garden 84 (72) Inventor Cleave Anthonys Binguzu UK, Esuo 50 7 N. W. chromatography, Ha Npusha, Eastleigh, Hotonhi over scan Zadorobu 2 (72) inventor Alois Martin Bock UK, Esuo 53 4 ares etch, Han Pusha, Eastleigh, Chandler Zufodo, Tyne Close

Claims (16)

[Claims] An apparatus for processing a plurality of individually encoded image elements to produce a single encoded image that can be reconstructed to represent the image, comprising: An apparatus comprising: a coordinator for selectively arranging each element selectively; and a combining unit for combining each of the selectively arranged elements to form a single encoded image. 2. The apparatus of claim 1, wherein the coordinator further comprises a separator for separating each element into components including video and control data. 3. Apparatus according to claim 1, wherein a coordinator is arranged to arrange each element so that temporally relevant video data from each element can be combined. 4. The coordinator further includes a repositioner for spatially positioning each element by selectively changing offset data in the video data.
Apparatus according to claim 1, 2 or 3. 5. The combining unit combines the temporally and spatially arranged video data with previously separated and temporally related components to form a single encoded image. The apparatus according to any one of 1, 2, 3, and 4. 6. At least one of the individually encoded elements to provide space used by the combiner.
6. Inserting redundant data into one of the two.
An apparatus according to any one of the above. 7. A method for processing a plurality of individually encoded image elements to produce a single encoded image that can be reconstructed to represent the image, the method comprising: Selectively adjusting each element for temporal and temporal positioning, and combining the selectively positioned elements to form a single encoded image. 8. The method of claim 7, wherein adjusting comprises separating each element into components including video and control data. 9. The method according to claim 7, further comprising the step of arranging each element such that temporally related video data from each element can be combined. 10. The method of claim 7, 8 or 9, wherein adjusting comprises repositioning each element by selectively changing offset data in the video data.
The method described in. 11. The combining step wherein the temporally and spatially arranged video data is combined with previously separated and temporally related components to form a single encoded image. Item 11. The method according to Item 7, 8, 9 or 10. 12. The method of claim 7, further comprising inserting redundant data into at least one of the individually encoded elements to provide space used in the combining step. The method described in. 13. Apparatus for transmitting a signal processed according to any of the preceding claims. 14. A method for transmitting a signal processed according to any one of claims 7 to 12. 15. Elements of a plurality of individually encoded images to create a single encoded image that can be reconstructed to represent the images as described herein with reference to the accompanying drawings. Equipment for processing. 16. A plurality of individually encoded image elements to create a single encoded image that can be reconstructed to represent the image as described herein with reference to the accompanying drawings. Way to handle.