JP2001103473A - Image packet structure, page image data structure, and packet generating apparatus and method - Google Patents

Abstract

(57) [Summary] [Problem] To determine the maximum data amount to be transmitted. SOLUTION: A packet is generated in tile units of 32 pixels × 32 pixels. If the image data amount after compression is smaller than the image data amount before compression, the compressed image data is accommodated in the data body 10 of the packet, and the compression flag of the header 12 is set to “1”. If the amount of image data after compression is larger than the amount of image data before compression, uncompressed image data is stored in the data body 10 of the packet, and the compression flag of the header 12 is set to “0”.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to an image packet structure, a page image data structure, and a packet generating apparatus and method.

[0002]

2. Description of the Related Art When compressing image data, the data amount after compression may be larger than the data amount before compression depending on the image or the compression method. As described above, when the data amount may increase due to the compression processing, it is difficult to determine the maximum data amount of one page or screen, or the maximum data amount of one packet obtained by further subdividing one page. Become. Here, conventionally, the capacity of a memory or a buffer for storing the compressed image data has a margin as expected.

[0003] When the data amount after compression is larger than the data amount before compression, a configuration that employs uncompressed image data is also known. In this conventional example, since compression / non-compression is selected in units of one page, memory for one page is separately provided.
It is necessary to prepare and store uncompressed image data in its memory until it is determined whether to use compressed or uncompressed oil.

[0004]

The capacity of the memory or buffer is not only wasted unless it is used, but it is difficult to set the capacity to a minimum. There is room for this, which is wasteful.

In a conventional example in which image data is compressed in units of one page and the amount of data after compression is larger than the amount of data before compression, one page of memory is used in a conventional example in which uncompressed data is used. It must be prepared, which increases costs.

Accordingly, an object of the present invention is to provide an image packet structure, a page image data structure, and a packet generating apparatus and method which do not require such a wasteful margin in a memory or a buffer.

According to the present invention or the memory for one page,
An object of the present invention is to provide an image packet structure, a page image data structure, and a packet generation device and method which do not need to be prepared.

[0008]

According to the present invention, there is provided an image packet structure comprising a data body containing image information and a header thereof. The header is obtained by compressing the image information contained in the data body. A compression flag indicating whether or not the compression flag is present.

The page image data structure according to the present invention has the following structure.
The image data of the page is nxm pixels (where n,
m is an integer of 1 or more), and the image information of each tile is formed into a packet including a data body containing the image information and a header, and the header is contained in the data body. And a compression flag indicating whether or not the image information is compressed.

A packet generating apparatus according to the present invention compresses image data, compares the data amount before and after the compression, and outputs data indicating the smaller data amount and information indicating the smaller data amount; A header generating means for generating a header including a compression flag having a content according to whether or not the data output from the header is compressed image data, and merging the data output from the compression means with the header output from the header generating means And merging means.

A packet generation method according to the present invention includes a compression step of compressing image data, comparing the data amounts before and after compression, and outputting data with the smaller data amount and information indicating the smaller data amount; As a result, a header generation step of generating a header including a compression flag having a content according to whether or not the output data is compressed image data; a data output by the compression step; and a header output by the header generation step Merge
And steps.

[0012]

Embodiments of the present invention will be described below in detail with reference to the drawings.

FIG. 1 shows a packet structure according to an embodiment of the present invention. One packet is composed of a data body 10 containing one tile of image data of 32 × 32 pixels and a header 12. The header 12 includes a packet ID 14 indicating the serial number of the packet, a compression flag 16 indicating whether the image data contained in the data body 10 is compressed, and an image representing the amount of image data contained in the data body 10. The data length 18 is provided. In this embodiment, the packet I
D14 is 1 byte, the compression flag 16 is 1 bit, and the data length 18 is 2 bytes.

FIG. 2 shows a state where one page of image data is divided into tiles of 32 pixels × 32 pixels. One tile of image data is accommodated in one packet as shown in FIG. One tile may not be 32 × 32 pixels, for example, may be 64 × 64 pixels, and may be a rectangle instead of a square. One tile is 32 pixels x 32
In the case of a pixel, for example, when A4 has a resolution of 600 dpi, the number of tiles is 220 × 156 = 34320,
When the resolution of A3 is 1200 dpi, the number of tiles is 440.
× 624 = 274,560.

In this embodiment, when compressing image data, it is determined whether to compress image data in packet units, that is, tile units, instead of compressing one page at a time. If the data amount after compression is larger than the data amount before compression, compression is not performed.

Since the packet of this embodiment has the compression flag 16 in the header 12, it can be identified by the compression flag 16 whether the data body 10 contains compressed image data or uncompressed image data. When the compressed image data is stored in the data body 10, the compression flag is set to "1".
Is set, and when non-compressed image data is stored in the data body 10, '0' is set in the compression flag. In one page, packets having compressed image data and packets having non-compressed image data are mixed.

For example, it is assumed that there is one page of plain image data, and when the image data is compressed, the image data format of the present embodiment is adopted. First, the page image data is divided into tiles as shown in FIG. The image data for each tile is compressed by, for example, the JPEG method. The compression method may be another method. As a result of the compression, when the data amount becomes smaller than the original image data amount, the compressed image data is stored in the data body 10 of the packet, the data amount is stored in the data length 18 of the header 12, and the compression flag 16 Set '1'. Conversely, as a result of compression, when the data amount becomes larger than the original image data amount, the uncompressed image data is accommodated in the data body 10 of the packet, and the data amount is stored in the data length 1 of the header 12.
8 and the compression flag 16 is set to '0'.

In this way, data compression and packetization are repeated for each tile to generate one page of data. The data for one page created in this way is
It can be guaranteed that the data body 10 of each packet does not exceed the original image data amount. Since the data amount of the header 12 is fixed (4 bytes), the maximum value of the total data amount of one page is (uncompressed image data amount of one tile + data amount of header 4 Bytes) × (number of tiles of one page). Is calculated.

When decompressing uncompressed image data by expanding page image data in which compressed / uncompressed packets are mixed, the compression flag 16 of the header 12 of each packet is used.
, And only the data contained in the packet in which it is '1' need be expanded.

As described above, in the present embodiment, it is only necessary to temporarily hold the uncompressed image data for one tile, so that the memory capacity for that is small. After the compression of the image data of the tile is completed and the packet of the tile is generated, the image data of the tile may be deleted. That is, each time the image data of one tile is compressed and packetized, the original image data of that tile can be deleted. As a result, the work area of the memory required for compression only needs to have the capacity of image data for one tile.

FIG. 3 is a schematic block diagram of the present embodiment. 20 is a CPU, 22 is a memory control circuit, and 24 is a memory. 26 is an input terminal for inputting image data to be processed, 28 is image data input to the input terminal 26,
FIG. 4 is a packet generation circuit that forms a packet described with reference to FIGS. 1 and 2. FIG. The packet generation circuit 28 includes an image compression circuit that compresses image data in tile units. Reference numeral 30 denotes a buffer for temporarily storing image data before and after image compression. Reference numeral 32 denotes a DMA (Direct Memory Ac) output packet of the packet generation circuit 28.
cess), and is a DMA control circuit that transfers data to the memory 24 via the memory control circuit 22. 34 is a CPU
20, a system bus connecting the memory control circuit 22 and the DMA control circuit 32.

The basic data flow will be described. The image data is input terminal 2 in units of tiles of 32 pixels × 32 pixels.
Enter 6 The packet generation circuit 28 compresses the image data in tile units, compares the non-compressed image data with the compressed image data, stores the smaller data amount in the data body 10, and sets the compression flag 16 of the header 12 to “0”. 'Or'
1 'is set and output to the DMA control circuit 32 in the packet format shown in FIG. DMA control circuit 32
Represents the packet data from the packet generation circuit 28,
The data is written to the memory 24 via the system bus 34 and the memory control circuit 22.

FIG. 4 is a block diagram showing a schematic configuration of the packet generation circuit 28. The compression circuit 40 stores the image data in tile units from the input terminal 26 in the buffer 30 and also compresses and stores the compressed image data in the buffer 30.
The non-compressed image data is compared with the compressed image data, and the smaller data amount is output. That is, when the image data amount after compression is smaller than the image data amount before compression,
The compression circuit 40 outputs the compressed image data to the merge circuit 44, and also outputs the merged image data to the header generation circuit 42.
Is notified that the output data is compressed data and the amount of data. Conversely, when the image data amount after compression is larger than the image data amount before compression, the compression circuit 40
The image data before compression is output to the merge circuit 44, and the header generation circuit 42 is notified that the data output to the merge circuit 44 is uncompressed data and the amount of the data.

The header generation circuit 42 outputs the packet ID 14
Are sequentially incremented, and the data amount is set to the image data length 18.
Is set to "0" when the uncompressed data is output to the merge circuit 44, the compression circuit 40
Is set to "1" when the compressed image data is output to. Then, the header generation circuit 42 outputs the generated header information to the merge circuit 44.

The merge circuit 44 merges the header from the header generation circuit 42 and the data from the compression circuit 40 to generate a packet having the format shown in FIG. The generated packet is sent to the DMA control circuit 32.

The DMA control circuit 32 has a packet address register indicating an address for storing packet data. The DMA control circuit 32 includes a packet generation circuit 28
The packet from (the merge circuit 44) is written to the address of the memory 24 indicated by the packet address register. The packet address register adds and updates the stored value by the packet capacity. After that, the DMA control circuit 3
2 receives the next packet from the packet generation circuit 28.

[0027]

As can be easily understood from the above description, according to the present invention, a compression flag indicating whether or not image data is compressed is provided in the header of an image packet, so that each packet is compressed / compressed. You can select uncompressed. Accordingly, when the data after compression becomes larger than the original data, it is possible to select packetization without changing the uncompressed data in packet units. Also, since the maximum value of the image data amount of one packet and one page can be determined,
It is not necessary to provide a memory and a buffer for storing data with an extra margin.

Further, each time one tile of image data is compressed and packetized, the original tile image data can be deleted, so that the work area of the memory required for compression can be reduced to one tile. .

[Brief description of the drawings]

FIG. 1 is a structural diagram of a packet data format of the present embodiment.

FIG. 2 is a schematic diagram of one page of image data divided into tile units of 32 × 32 pixels.

FIG. 3 is a schematic configuration block diagram of the present embodiment.

FIG. 4 is a schematic block diagram of a configuration of a packet generation circuit 28;

[Explanation of symbols]

 10: Data body 12: Header 14: Packet ID 16: Compression flag 18: Image data length 20: CPU 22: Memory control circuit 24: Memory 26: Input terminal 28: Packet generation circuit 30: Buffer 32: DMA control circuit 34: System bus 40: Compression circuit 42: Header generation circuit 44: Merge circuit

Claims (10)

[Claims] 1. A data body containing image information, and a header thereof, wherein the header includes a compression flag indicating whether or not the image information contained in the data body is compressed. Image packet structure. 2. The image information contained in the data body is one of a plurality of tiles constituting one page of image data.
The image packet structure according to claim 1, wherein the image packet structure includes: 3. The tile has n × m pixels (where n, m
3. The image packet structure according to claim 2, wherein (a) is an integer of 1 or more. 4. An image data of one page is n × m
It is divided into a plurality of tiles composed of pixels (where n and m are integers of 1 or more), and the image information of each tile is formed into a packet composed of a data body containing the image information and its header, and the header is A page image data structure comprising a compression flag indicating whether image information contained in the data body is compressed. 5. A compression means for compressing image data, comparing the data amounts before and after compression, and outputting data of the smaller data amount and information indicating the smaller data amount, wherein the data output from the compression means is compressed. A header generating means for generating a header including a compression flag having a content corresponding to image data; and a merging means for merging data output from the compressing means and a header output from the header generating means. A packet generation device characterized by the above-mentioned. 6. The image information contained in the data body is one of a plurality of tiles constituting one page of image data.
6. The packet generation device according to claim 5, wherein the packet generation device includes: 7. The tile having a size of n × m pixels (where n, m
7. The packet generation device according to claim 6, comprising: an integer of 1 or more. 8. A compression step of compressing image data, comparing the data amounts before and after compression, and outputting data with the smaller data amount and information indicating the smaller data amount, and data output as a result of the compression step. A header generation step of generating a header including a compression flag having a content according to whether or not the data is compressed image data; a merge step of merging data output by the compression step with a header output by the header generation step A packet generation method comprising: 9. The image information contained in the data body is one of a plurality of tiles constituting one page of image data.
9. The packet generation method according to claim 8, wherein the packet generation method includes: 10. The tile in which n × m pixels (where n,
The packet generation method according to claim 8, wherein m is an integer of 1 or more.