JPH08163563A - Still picture data compression device - Google Patents

Abstract

PURPOSE: To suppress the errors of blocks provided with supplementary data by adding the supplementary data generated based on the data of remaining lines by a line number lacking from a stipulated line number and constituting a block. CONSTITUTION: Still picture data for eight lines are written from a still picture input circuit 20 to the picture memory 21 of eight lines by the control of a memory controller 22. Thereafter, to the still picture data of 1st-1032nd lines among 1035 lines read from the memory 21, a data compression processing by a JPEG encoder 23 is performed. Then, the data of 1032nd-1035th lines are written in the memory 21 as they are. Then, the still picture data of the 1035th line are repeatedly written in a free area for five lines. Thus, by using the still picture data of the 1035th line as the supplementary data for five lines, a large error is not generated in the block provided with the supplementary data.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION The present invention relates to an HDTV (High Definition
The present invention relates to a broadcasting system such as inition television) and, more particularly, to an improvement of a still image data compression device for performing high efficiency encoding processing on the still image data to be broadcast.

[0002]

2. Description of the Related Art As is well known, in a broadcasting system such as the one mentioned above, JPEG (Joint Photographic Coding) is used as a data compression technique applied to still image data to be broadcast.
Experts Group) method is widely adopted. This JPE
In the data compression technology by the G method, still image data is divided into blocks of 8 pixels in the horizontal direction and 8 lines in the vertical direction, and DCT (Discreat Cosine Transf
orm) processing, quantization processing, and Huffman coding processing.

By the way, since the still image data in HDTV is composed of 1920 pixels in the horizontal direction × 1035 lines in the vertical direction as shown in FIG. 4, when it is divided into blocks of 8 pixels × 8 lines, For 1
In 920/8, exactly 240 blocks are constructed,
In the vertical direction, 129 blocks are formed by 1035/8, and a remainder of 3 lines is generated.

For this reason, conventionally, as shown by hatching in FIG. 4, by adding DC (direct current) data for 5 lines in the vertical direction, that is, 1920 pixels × 5 lines, as supplementary data, lines in the vertical direction are added. Set the number to 1040, and in the end,
240 blocks in the horizontal direction × 130 blocks in the vertical direction are formed, and each block is sequentially subjected to data compression processing by the JPEG method.

FIG. 5 shows such a conventional still image data compression apparatus. That is, in the input terminal 11,
Still image data to be broadcast on HDTV is supplied. The still image data supplied to this input terminal 11 is
The one frame is written in the image memory 13 via the still image input circuit 12. Since the still image data for one frame stored in the image memory 13 has a size of 1920 pixels in the horizontal direction × 1035 lines in the vertical direction as described above, it is necessary to add DC data for 5 lines.

Therefore, by the memory controller 14,
For 1 to 1035 lines, the still image data read from the image memory 13 is guided to the JPEG encoder 15, and for 1036 to 1040 lines, the DC data supplied to the input terminal 16 is supplied to the JPEG encoder. The selector 17 is controlled so as to lead to 15. Then, the still image data and the DC data selected by the selector 17 are supplied to the JPEG encoder 15, subjected to data compression processing, and taken out from the output terminal 18.

By the way, in the conventional still image data compression apparatus as described above, the value of the DC data for supplementing 1036 to 1040 lines is 1 to 103.
If the distance is far from the still image data up to 5 lines, a high frequency component is generated when the DCT conversion process is performed, and as a result of image display after compression / expansion, 1033 lines to 1035 lines included in the same block as the DC data. Distortion will occur in the still image data.

In this case, the portion from the 1033th line to the 1035th line is displayed so as to be brightly raised on the screen, or conversely, to be darkly depressed. Then, in this case, for example, in a broadcast to which a vertical scroll effect is given, as shown in FIG.
Scrolls in the vertical direction, there is an inconvenience that the boundary between the still images 1 and 2 becomes unnatural.

More specifically, when the quantization table shown in FIG. 7 is used to quantize still image data (DCT coefficients) of 8-bit 256 gradations as shown in FIG. think of. In this case, 1036 to 104
"30" is added as DC data to the 0th line. When the compressed still image data is expanded, it can be seen that a large error has occurred in 1034 lines as shown in FIG. 8B. This error is a value that is clearly recognized on the screen.

[0010]

As described above, in the conventional still image data compression apparatus, when the still image data is compressed and expanded, a large error occurs in the block of the compression process including the supplementary data. Have the problem of doing.

Therefore, the present invention has been made in consideration of the above circumstances, and can prevent a large error from occurring in a block including supplementary data when compression / expansion processing is performed on still image data. An object is to provide a very good still image data compression device.

[0012]

A still image data compression apparatus according to the present invention divides one screen of still image data into a plurality of blocks each having a prescribed number of pixels and a prescribed number of lines in the horizontal and vertical directions. However, the data compression processing is performed in units of the blocks, and when still image data for one screen is divided in the vertical direction by a specified number of lines, the remaining lines that are less than the specified number of lines are , Adds supplementary data of the number of lines that is insufficient to the specified number of lines, and applies data compression processing in block units composed of extra lines and supplementary data, based on the data of the extra lines. It is configured to generate supplementary data.

Further, the still image data compression apparatus according to the present invention has a storage means having a storage capacity for storing one screen of still image data, and stores one screen of still image data in the storage means. At the same time, the stored still image data for one screen is divided into a plurality of blocks each having a prescribed number of pixels and a specified number of lines in the horizontal direction and the vertical direction, and is read out. In a state in which still image data for one screen is divided into blocks and read out, for a surplus line that does not reach the specified number of lines in the vertical direction, the specified number of lines is determined based on the data of the remaining lines. By generating and adding supplementary data of the number of lines lacking in the memory, and adding means for forming a block with the remaining lines and supplementary data, and the storage control means. Is obtained so as to include a compression means for performing data compression processing, respectively on the still image data from the still picture data and additional means are read in blocks from 憶 means is output block by block.

Further, the still image data compressing apparatus according to the present invention has two consecutive scrolling processes in the vertical direction.
The first and second still image data corresponding to the screen are divided into a plurality of blocks each having a specified number of pixels and a specified number of lines in the horizontal direction and the vertical direction, and data compression processing is performed for each block. , When the first still image data is divided in the vertical direction by a prescribed number of lines, the second still image data is insufficient in the prescribed number of lines for the remaining lines less than the prescribed number of lines. The data of the number of lines being extracted is extracted and added, and data compression processing is performed in block units composed of the remaining lines of the first still image data and the lines extracted from the second still image data. It was done.

[0015]

According to the above configuration, when still image data for one screen is divided in the vertical direction by a prescribed number of lines, the remaining lines less than the prescribed number of lines are Since supplementary data generated based on the data of the remaining lines is added to the specified number of lines by the number of lines that are insufficient, blocks are constructed, so compression / decompression processing is performed on still image data. When applied, it is possible to prevent a large error from occurring in the block including the supplementary data.

Further, for the first and second still image data corresponding to two consecutive screens which are scrolled in the vertical direction, the first still image data is divided in the vertical direction by a prescribed number of lines. At this time, the data of the number of lines short of the specified number of lines is extracted from the second still image data and added to the remaining lines less than the specified number of lines. Also in this case, it is possible to prevent the boundary portion between two continuous screens from becoming unnatural.

[0017]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described in detail below with reference to the drawings. In FIG. 1, the input terminal 19 is supplied with still image data of 1920 pixels in the horizontal direction and 1035 lines in the vertical direction to be broadcast by HDTV. The still image data supplied to this input terminal 19 is
It is supplied to the 8-line image memory 21 via the still image input circuit 20. The 8-line image memory 21 has a storage capacity capable of storing still image data of 1920 pixels in the horizontal direction × 8 lines in the vertical direction or more.

Under control of the memory controller 22, the 8-line image memory 21 outputs 1920 pixels in the horizontal direction × 1035 in the vertical direction output from the still image input circuit 20.
Of still image data of lines, horizontal 1920 pixels × vertical 8 lines are written. And this 8
From the line image memory 21, the memory controller 22
By this control, the written still image data of 1920 pixels in the horizontal direction × 8 lines in the vertical direction is read out in 240 units in blocks of 8 pixels in the horizontal direction × 8 lines in the vertical direction.

After that, the still image data in block units of 8 pixels in the horizontal direction × 8 lines in the vertical direction read from the 8-line image memory 21 is JPEG encoder 23.
Is supplied to the DC as described above,
Data compression processing is performed by performing T processing, quantization processing, and Huffman coding processing, and the data is taken out from the output terminal 24.

In the above manner, 1 line to 103
The still image data of up to 2 lines is normally subjected to data compression processing by the JPEG encoder 23 in block units of 8 pixels in the horizontal direction × 8 lines in the vertical direction. Then, when the data compression processing by the JPEG encoder 23 is completed for the still image data from 1 line to 1032 lines, 1033 lines to 1033 lines are completed.
Still image data for 3 lines up to 35 lines will remain.

The still image data of 3 lines from 1033 lines to 1035 lines are directly stored in the 8-line image memory 2 under the control of the memory controller 22.
Written to 1. Therefore, the 8-line image memory 21
Will result in an empty area for 5 lines. At this time, the memory controller 22 controls so that the still image data of the 1035th line is repeatedly written in the empty area of 5 lines of the 8 line image memory 21.

In this way, from 1033 lines to 10
3 lines up to 35 lines and 5 lines of still image data on the 1035th line become 8 lines, the still image data is an 8 line image in block units of 8 pixels in the horizontal direction and 8 lines in the vertical direction. The data is read out from the memory 21 and supplied to the JPEG encoder 23 for data compression processing.

According to the configuration of the above embodiment, 103
As supplementary data for 5 lines added to 3 lines from 3 lines to 1035 lines, the last line, that is, the still image data of 1035 lines is repeatedly used instead of the DC data as in the conventional case. It is possible to prevent a large error from occurring in the block including the supplementary data when the compression / expansion processing is performed without largely separating from the image data.

Next, FIG. 2 shows a second embodiment of the present invention.
It shows a configuration in which data compression processing is performed on still image data for two screens that are scrolled in the vertical direction. In other words, the input terminal 25 is supplied with still image data of 1920 pixels in the horizontal direction × 1035 lines in the vertical direction, which is to be vertically scrolled by HDTV.

The still image data supplied to the input terminal 25 is supplied to the frame image memory 27 and the 5-line image memory 28 via the still image input circuit 26. Of these, the frame image memory 27 is 1 in the horizontal direction.
It has a storage capacity capable of storing still image data of 920 pixels × 1035 lines in the vertical direction. Further, the 5-line image memory 28 has 1920 horizontal pixels × 5 vertical pixels.
It has a storage capacity capable of storing still image data of lines.

Under the control of the memory controller 29, the frame image memory 27 outputs 1920 pixels in the horizontal direction × 1035 in the vertical direction output from the still image input circuit 26.
The still image data of the line is written. Then, from the frame image memory 27, the memory controller 2
By the control of 9, the written horizontal 1920 pixels ×
The still image data for 1035 lines in the vertical direction is read in block units of 8 pixels in the horizontal direction × 8 lines in the vertical direction.

Further, the 5-line image memory 28 includes:
Under the control of the memory controller 29, the horizontal direction 1920 pixels output from the still image input circuit 26 × vertical direction 1
Horizontal direction 192 out of 035 lines of still image data
0 pixels x 5 lines in the vertical direction are written. And
Under the control of the memory controller 29, the written still image data of 1920 pixels in the horizontal direction × 5 lines in the vertical direction is read from the 5-line image memory 28 in units of 8 pixels in the horizontal direction × 5 lines in the vertical direction. .

Thereafter, still image data read from the frame image memory 27 in block units of 8 pixels in the horizontal direction × 8 lines in the vertical direction and the 5-line image memory 28.
The still image data in the unit of horizontal 8 pixels × vertical 5 lines read from is supplied to the selector 30 and selectively derived under the control of the memory controller 29.

In this case, the memory controller 29 determines the frame image memory 27 and the 5-line image memory 2 based on the scroll identification signal supplied to the input terminal 31.
The write operation and the read operation for 8 are controlled, and the selection operation of the selector 30 is controlled. This scroll identification signal indicates whether or not the still image data supplied to the input terminal 25 is given a vertical scrolling instruction on the transmitting side.

The still image data derived by the selector 30 is supplied to the JPEG encoder 32 and is subjected to DCT processing, quantization processing and Huffman coding processing on a block-by-block basis to perform data compression processing. Is performed and is taken out from the output terminal 33.

First, the input terminal 25 is supplied with the first still image data on the upper side of the screen, out of the two screens of still image data scrolled in the vertical direction. The first still image data is written in the frame image memory 27 via the still image input circuit 26, and then sequentially read in block units of 8 pixels in the horizontal direction × 8 lines in the vertical direction, and is read via the selector 30. JPE
By being supplied to the G encoder 31, data compression processing is performed.

In this way, when the data compression processing in block units by the JPEG encoder 31 is completed for the first still image data from line 1 to line 1032, the frame image memory 27 stores Three lines from 1033 lines to 1035 lines of one still image data are left unread. In this state, the memory controller 29 determines whether or not there is a scroll instruction based on the scroll identification signal supplied to the input terminal 31.

When there is a scroll instruction,
The memory controller 29 selects one of the two still image data scrolled in the vertical direction on the lower side of the screen, that is, the second still image input to the input terminal 25 after the first still image data. Control is performed so that the first to fifth lines of the image data are written in the 5-line image memory 28.

Thereafter, the memory controller 29 stores the first still image data 103 from the frame image memory 27.
Three lines from 3 lines to 1035 lines are read in units of horizontal 8 pixels × vertical 3 lines, and 5 lines from the 5 line image memory 28 to 1 line to 5 lines of the second still image data. Minutes are read in units of 8 pixels in the horizontal direction and 5 lines in the vertical direction,
By alternately selecting with the selector 30, three lines of the first still image data and five lines of the second still image data are mixed in block units of horizontal 8 pixels × vertical 8 lines. Still image data is generated, JPE
Data compression processing is performed by the G encoder 31.

On the other hand, when the scroll instruction is not issued in the state where three lines 1033 to 1035 of the first still image data remain in the frame image memory 27 without being read, the first still image data 1033
5 lines of supplemental data generated based on 3 lines from line 1035 to image line 2
Write to 8 and the first still image data for 3 lines and 5
8 pixels in the horizontal direction mixed with supplementary data for lines x
The data compression processing is performed on a block composed of 8 lines in the vertical direction.

According to the configuration of the second embodiment,
First, when there is a scroll instruction, the supplementary data is not interposed between the first still image data and the second still image data, and the last three lines of the first still image data and the second still image data are not inserted. By mixing the first 5 lines of the image data to form a block consisting of 8 pixels in the horizontal direction and 8 lines in the vertical direction for compression processing, it is possible to reduce As described above, it is possible to prevent the boundary portion of the still images for two screens from becoming unnatural.

When there is no scroll instruction, the last three lines of the first still image data and the supplementary data for five lines generated based on these three lines are mixed to perform compression processing. Horizontal 8 pixels x vertical 8
Since the block made up of lines is configured, the supplementary data does not greatly separate from the still image data, and a large error may occur in the block including the supplementary data when the compression / decompression process is performed. Can be suppressed.

Next, FIG. 3 shows a modification of the second embodiment. 3, the same parts as those in FIG. 2 are indicated by the same reference numerals. That is, instead of the 5-line image memory 28 shown in FIG.
A frame image memory 34 having a storage capacity capable of storing still image data of 1035 lines in the vertical direction is used.

In this case, the first still image data is written in the frame image memory 27, and the second still image data is written in the frame image memory 34. If there is a scroll instruction, the frame image memory 27
By mixing the last 3 lines of the first still image data stored in the first 5 lines of the second still image data stored in the frame image memory 34 with each other in the horizontal direction 8 for compression processing. A block composed of pixels × 8 lines in the vertical direction can be configured. Also, when there are two or more screens of scrolling still images, the positions of the frame image memories 27 and 34 are exchanged so that continuous compression processing can be performed. The present invention is not limited to the above-described embodiments, but can be variously modified and implemented without departing from the scope of the invention.

[0040]

As described above in detail, according to the present invention,
When compressing / decompressing still image data,
It is possible to provide a very good still image data compression apparatus capable of suppressing a large error from occurring in a block including supplementary data.

[Brief description of drawings]

FIG. 1 is a block configuration diagram showing an embodiment of a still image data compression apparatus according to the present invention.

FIG. 2 is a block diagram showing a second embodiment of the present invention.

FIG. 3 is a block diagram showing a modified example of the second embodiment.

FIG. 4 is a diagram shown for explaining still image data in HDTV.

FIG. 5 is a block configuration diagram showing a conventional still image data compression device.

FIG. 6 is a diagram for explaining a problem of the conventional device.

FIG. 7 is a diagram showing an example of a quantization table.

FIG. 8 is a diagram showing still image data compressed / decompressed by the quantization table.

[Explanation of symbols]

11 ... Input terminal, 12 ... Still image input circuit, 13 ... Image memory, 14 ... Memory controller, 15 ... JPEG encoder, 16 ... Input terminal, 17 ... Selector, 18 ... Output terminal, 19 ... Input terminal, 20 ... Still image Input circuit, 21 ...
8-line image memory, 22 ... Memory controller, 23
... JPEG encoder, 24 ... Output terminal, 25 ... Input terminal, 26 ... Still image input circuit, 27 ... Frame image memory, 28 ... 5-line image memory, 29 ... Memory controller, 30 ... Selector, 31 ... Input terminal, 32 ... JPE
G encoder, 33 ... Output terminal, 34 ... Frame image memory.

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI Technical indication H04N 7/015

Claims (4)

[Claims] 1. Still image data for one screen is divided into a plurality of blocks each having a prescribed number of pixels and a specified number of lines in the horizontal and vertical directions, and data compression processing is performed in block units. When still image data for one screen is divided in the vertical direction by a prescribed number of lines, the remaining number of lines less than the prescribed number of lines is determined by the number of lines short of the prescribed number of lines. In a still image data compression device that adds supplementary data and performs the data compression processing in block units composed of the extra line and supplemental data, the supplemental data is generated based on the data of the extra line. A still image data compression apparatus having the above-mentioned configuration. 2. A storage unit having a storage capacity for storing still image data for one screen, the still image data for one screen being stored in the storage unit, and the stored one screen. Storage control means for reading the still image data by dividing it into a plurality of blocks each having a prescribed number of pixels and a predetermined number of lines in the horizontal and vertical directions, and still image data for one screen from the storage means by the storage control means. In the state where the data is divided into the block units and read out, with respect to a surplus line that is less than the specified line number in the vertical direction, the specified line number is insufficient based on the data of the surplus line. Generating and adding supplementary data for the number of lines, and adding means for forming the block by the remaining lines and supplementary data, and the storage control means Still image data, comprising: still image data read in block units from the storage unit and compression unit for performing data compression processing on the still image data output in block units from the adding unit. Compressor. 3. The still image data compression apparatus according to claim 1, wherein the supplementary data is obtained by repeatedly using still image data of the remaining lines by the number of insufficient lines. 4. A plurality of blocks in which first and second still image data corresponding to two consecutive screens that are scrolled in the vertical direction are defined in the horizontal direction and in the vertical direction by a specified number of pixels and lines, respectively. And the data compression processing is performed for each block, and when the first still image data is divided by a specified number of lines in the vertical direction, the remaining lines that are less than the specified number of lines are Data of the number of lines short of the specified number of lines is extracted from the second still image data and added, and the remaining lines of the first still image data and the second still image data are added. A still image data compression apparatus, wherein the data compression processing is performed in a block unit composed of a line extracted from the line.