JP2003087573A - Image compressing device and image expanding device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image compressing device which separates the image data in a word unit and compresses the image data by a compressing method fitted to the kind of the image data of the word unit, and an image expanding device which separates the encoded data in a word unit and expands the encoded data by an expanding method fitted to the kind of the encoded data of the word unit. SOLUTION: This image compressing device 1 separates the input image data with 32 bits as one word unit in each of different kinds (1) to (3) in an image separation block 2, compresses the separated image data of the one word unit in encoding blocks 3 to 5 fitted to the kinds (1) to (3) of the image data of the one word unit among a plurality of encoding blocks 3 to 5 for performing compression respectively by specialized compressing methods, and outputs the image data as encoded data with 8 bits as one word unit. Then, the image data are compressed with a simple algorithm, making it easy to build into hardware, a processing speed can be improved and compressibility can also be improved.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image compression apparatus and an image decompression apparatus, and more specifically, to 32 bit image data.
The data capacity is surely reduced by compressing by a compression method adapted to the type of the image data in the word unit of t, and the code data is separated in the word unit of 8 bit to calculate the word unit of the word unit. The present invention relates to an image compression apparatus and an image decompression apparatus that decompress by a decompression method adapted to the type of coded data and restore image data without deterioration.

[0002]

2. Description of the Related Art Recently, image data has come to be handled not only by facsimile machines but also by digital cameras, computers, etc., in order to improve the transfer rate of image data and reduce the storage capacity in the storage memory. In addition, image data is encoded and data is compressed.

There are various types of data compression methods, such as JPEG DCT (Joint P).
photographic Expert Group,
Discrete Cosine Transform
m) system and JPEG Spatial system. In the field of facsimile, for example, MH (Modified
Huffman) encoding method, MR (Modufied Relative eleme)
nt address designate) Encoding method or MMR
(Modified MR) encoding method and the like.

[0004] Conventionally, a device for image compression has been
An appropriate compression method is fixedly set for the image data handled by the device.

[0005]

However, in such a conventional image compression apparatus, an appropriate compression method is fixedly set for the image data handled by the apparatus, so that various image data can be obtained. Nowadays we are dealing with, and still need improvement.

That is, today, image data is handled in various fields, and the types of image data are diversified. In such diversified image data, when the image data is compressed by a certain compression method, the data capacity does not decrease but rather increases, but when compressed by another compression method, the data capacity is reduced. There is also.

There is also image data that is difficult to compress, and even if such image data that is difficult to compress is encoded, the data is not compressed, and rather the capacity may increase to several times the capacity of the original image data. is there. In reality, even when encoding image data that is difficult to compress, there is a method that can be performed with the same data capacity as the original image data, but it is difficult to implement in hardware because the algorithm is complicated. However, there is a problem that the processing speed is slow when processed by software.

On the other hand, there are various image data compression methods,
Code data compressed by various methods by the image compression apparatus of the present invention may be deteriorated unless it is expanded by a decompression method suitable for the code data.

Therefore, according to the first aspect of the invention, the input image data is separated by type by the image separating means with 32 bits as one word unit, and the separated one-word unit image data is compressed by a specialized compression method. The compression means adapted to the type of the image data in 1-word units out of the plurality of compression means to output 8 bits as coded data in 1-word units, thereby compressing by a simple algorithm, and To make it easy
It is an object of the present invention to provide an image compression device that improves the processing speed and has a good compression rate.

According to a second aspect of the present invention, when the serial code data having 8 bits as one word unit output from the compression means do not have the same value, the code data is directly used as the output code data and the same value is used. , The number of consecutive same values is counted and the counted number is used as the output code data, and the feature of the output code data is used to further increase the data amount. (EN) Provided is an image compression device capable of performing better compression by performing reduction compression to further increase the compression rate and preventing the result that the compression rate is low even when the image data is data that is easily compressed. Is intended.

According to a third aspect of the present invention, when the value of serial code data in units of 8 bits output from the compression means is not 0xFF consecutively, the code data is directly used as output code data and 0xFF is used. When 2 times continue, the number of times 0xFF continues after that is counted and the count number is used as the output code data, and the feature of the output code data is used to further reduce the data amount. (EN) An image compression device capable of further improving the compression rate by performing compression to prevent the result of a low compression rate even when the image data is data that can be easily compressed, and performing better compression. It is an object.

According to a fourth aspect of the present invention, the input code data, which is compressed by different compression methods and has 8 bits as one word unit, is separated by the code separating means according to the type, and the separated code data is separated for each type of code data. Inputting to a decompressing unit adapted to the type of the coded data among a plurality of decompressing units that respectively decompress by a specialized decompression method, decompressing by the decompressing unit and outputting 32 bits as image data in 1-word units. Provide an image decompression device for decompressing coded data into image data without degrading the image while decompressing the coded data with a simple algorithm to facilitate hardware implementation, improving the processing speed. It is an object.

According to a fifth aspect of the present invention, when the serial input code data having 8 bits as one word unit does not have the same value, the code data is directly input to the code separating means as the input code data, When the same value continues twice, the pre-expansion means for counting the number of times the same value continues after that and inputting the count number to the code separation means as the input code data is provided. An object of the present invention is to provide an image decompression device that can decompress code data into image data while further preventing the deterioration of the image by a simple algorithm using the characteristics of the data.

According to a sixth aspect of the present invention, the value of serial input code data in units of 8 bits is 0xF.
When F is not continuous, the input code data is directly input to the pre-expansion means as the input code data, and 0xFF
When 2 consecutive times, the number of times 0xFF continues after that is counted, and the second pre-expansion means for inputting the count number to the pre-expansion means as input code data is provided. It is an object of the present invention to provide an image decompressing device capable of decompressing code data into image data while further preventing the deterioration of the image with a simple algorithm by utilizing the above feature.

[0015]

According to another aspect of the present invention, there is provided an image compression device for separating input image data by type, with 32 bits as one word unit, and one word separated by the image separation device. An image compression apparatus comprising: a plurality of compression means for compressing by a compression method specialized for each type of image data of a unit, wherein the input image data is separated by the image separation means, and the separation is performed. The 1-word unit image data is input to a compression means of the plurality of compression means that is adapted to the type of the 1-word unit image data, and is compressed by the compression means to obtain 8 bits.
The above object is achieved by outputting as coded data in word units.

According to the above construction, the input image data is converted into 32 bytes.
The bit is divided into 1 word unit by the image separating means, and the separated 1 word unit image data is compressed by a specialized compression method. Since the data is compressed by an adaptive compression means and is output as coded data in units of 8 bits, the compression can be performed by a simple algorithm to facilitate hardware implementation and improve the processing speed. The rate can be improved.

In this case, for example, as described in claim 2, in the image compression apparatus, when the serial code data in which the 8-bit unit output by the compression unit is one word unit does not have the same value, Is to use the code data as the output code data as it is, and when the same value continues twice, counts how many times the same value continues after that and uses the count number as the output code data. May be.

According to the above structure, the compression unit outputs 8
Serial code data with bit as 1 word unit,
When the same value does not continue, the code data is used as it is as output code data, and when the same value continues twice, the number of times the same value continues after that is counted,
Since the count number is used as the output code data, the compression rate can be further increased by performing compression to further reduce the data amount by utilizing the characteristics of the output code data, and the image data is easy to compress. However, it is possible to prevent the result that the compression ratio is low, and to perform even better compression.

Further, for example, as described in claim 3, the image compression device outputs 8b output from the compression means.
When the value of serial code data with it as a word unit is not 0xFF, the code data is used as it is as output code data, and when 0xFF is continuous twice, the number of times 0xFF is continuous after that. Whether or not to do so may be counted and the counted number may be used as output code data.

According to the above configuration, the compression unit outputs 8
When the serial code data value in which the bit is one word unit is not 0xFF consecutively, the code data is directly used as the output code data, and when 0xFF is consecutive twice, 0xFF consecutively thereafter. The output code data is characterized by using the characteristics of the output code data to further reduce the data amount and further increase the compression rate. Even if the data is easy to compress, it is possible to prevent the result of a low compression ratio and perform better compression.

In the image decompressing device according to the fourth aspect of the present invention, the code demultiplexing means for demultiplexing the input code data which is compressed by different compression methods and has 8 bits as one word unit and the code demultiplexed by the code demultiplexing means. An image decompression device comprising a plurality of decompression means for decompressing data by a decompression method specialized for each type of the code data, wherein the input code data is separated by the code separation means according to the type. By inputting the separated code data to a decompressing unit adapted to the type of the code data among the plurality of decompressing units, decompressing by the decompressing unit and outputting 32 bit as one-word unit image data, You have achieved your purpose.

According to the above construction, the input code data which is compressed by different compression methods and has 8 bits as one word unit is separated by the code separating means by the type, and the separated code data is specified for each type of code data. Of the plurality of decompression means that decompress by the decompressed decompression method, the decompression means adapted to the type of the code data is input, decompressed by the decompression means, and 32 bits are output as image data in 1-word units. It is possible to expand the code data by using such an algorithm to facilitate hardware implementation, improve the processing speed, and decode the code data into image data without degrading the image.

In this case, for example, as described in claim 5, the image decompressing device, when the serial input code data in which the 8-bit unit is one word is not continuous in the same value, Is input to the code separating means as the input code data as it is, and when the same value continues twice, the number of times the same value continues after that is counted and the count number is used as the input code data. Pre-expansion means for inputting to the code separation means may be provided.

According to the above configuration, when the serial input code data having 8 bits as a unit of word do not have the same value, the code data is directly input to the code separating means as the input code data and the same value is input. In the case where the two consecutive numbers are consecutive, the pre-expansion means for counting how many times the same value continues after that and inputting the count number to the code separation means as the input code data is provided. By making use of the feature (3), the code data can be expanded into the image data with a simple algorithm while further preventing the deterioration of the image.

Further, for example, as described in claim 6, when the value of serial input code data in which the 8-bit unit is one word is not 0xFF consecutive, the image decompressing device sets the input code. The data is directly input to the pre-expansion means as the input code data, and 0xF
When F continues twice, it is possible to include a second pre-expansion means for counting how many times 0xFF continues after that and inputting the count number to the pre-expansion means as the input code data. Good.

According to the above construction, when the value of serial input code data with 8 bits as a unit of 1 word is not continuous with 0xFF, the input code data is directly input as input code data to the pre-expansion means, 0xFF is 2
In the case of consecutive times, the second pre-expansion unit that counts the number of times 0xFF continues after that and inputs the count number to the pre-expansion unit as the input code data is provided. By utilizing the feature, the code data can be expanded into the image data with a simple algorithm while further preventing the deterioration of the image.

[0027]

BEST MODE FOR CARRYING OUT THE INVENTION Preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings. The embodiments described below are preferred embodiments of the present invention, and therefore have various technically preferable limitations. However, the scope of the present invention refers to the present invention particularly in the following description. Unless otherwise stated, the present invention is not limited to these embodiments.

1 to 3 are views showing a first embodiment of an image compression apparatus and an image expansion apparatus of the present invention.
FIG. 1 is a block diagram of a main part of an image compression device 1 to which an image compression device and an image expansion device according to a first embodiment of the present invention are applied.

In FIG. 1, the image compression apparatus 1 is provided with an image separation block (image separation means) 2 and three coding blocks (compression means) 3 to 3 of first to third.

The image separation block 2 contains 3 pieces of image data.
The input image word having 2 bits as one word unit is input, and the image separation block 2 classifies the input image word into the following three types. However, since the number of main scanning pixels of the input image data is not always a multiple of 32, there may be a portion where there is no image data in the final image word of the line. In that case, "0x00" or Dummy data such as "0xFF" is added to fill the remaining portion.

All 32 bits are "0" or "1". Also, if it applies to 5 specific patterns.

When the input image word has a low frequency component which is lower than the specified value.

In cases other than the above.

The image separation block 2 classifies the input image word into any one of the above three cases, and the first coding block 3, the second coding block 4 and the third coding which are suitable for the respective cases. Output to block 5.

The first coding block 3, the second coding block 4, and the third coding block 5 are the image separation block 2
Is encoded (compressed) by an encoding method (compression method) adapted to the image block separated and input by
Output as an output code word in word units. The number of output code words for one input image word will be described later.

In the first encoding block 3 to the third encoding block 5, the input image word adapted to the first encoding block 3 is the input image word of the above type for the sake of explanation, and the second encoding The input image word adapted to block 4 is an input image word of the above type,
It is assumed that the input image word adapted to the coding block 5 is an input image word of the above type.

The first encoding block 3 classifies the input image words to be adapted into types, and converts the input image words of 32 bits into output code data of 4 bits as shown in FIG. Packs two pieces of output code data and outputs them as an 8-bit code word. The first encoding block 3 processes as described above when the type input image words are continuous, but outputs the type 4 bit output when the type or type input image word comes after the type input image word. When the code data is odd, the pass code of "0110" is packed and output as a code word of 8 bits.

It should be noted that the patterns ~ of the types of input image words in Fig. 2 correspond to the five patterns ~ shown in Fig. 3. When the pattern is the same as the preceding input image word, the pattern is Is shifted to the input image word side 8 bits before. Similarly, a pattern ~ is defined.

The first encoding block 3 encodes the input image word by 32b except the pass code.
The input image word of it becomes a code of 4 bits.

Next, the second coding block 4 classifies the input image data to be adapted into types, and the input image words that did not apply in the case of the type have the same value from the MSB side in the LSB direction. The input image word is divided into a plurality of run-length groups by counting whether or not they are successively continued. Then, since the second coding block 4 defines the value of the first set as 0, the leading bit is "1".
In the case of, the first group is the group with the count number “0”. In addition, the second coding block 4 sets the maximum number of counts in the set to 15 (4 bits), and if the same value continues for more than that, the next set is also set as the same value and the subsequent count number is inserted. . Further, in the second coding block 4, even if the count number ends with just “15”, the next group also has the same value group, so the count number “0” is put in the next group and the last group is added. Since the information of (1) can be decrypted without it, it is omitted. Then, when the count number of the immediately preceding group is “15”, the second coding block 4 also omits the information about the immediately preceding group, and outputs the last 1
When the count number of the immediately preceding set and the count number of the immediately preceding set are both “15”, the last and last two sets are also omitted.

When the number of sets thus created is 7 or less, since the input image word is applicable to the type, the second coding block 4 outputs the output code word as shown in FIG. , "1yyyzzzz", "yy
Put the number of pairs in the bit (3 bits) of "y", and
Put the count number of the first set into the bit (4 bits) of "z". Then, the second coding block 4 packs the count numbers of the second and subsequent groups for every two groups and outputs them as 8-bit code words.

Therefore, when the number of sets is 1, it is 1 word, when the number of sets is 2 or 3, it is 2 words, when it is 4 or 5, it is 3 words, and when it is 6 or 7, it is 4 words. Is the output code of. As a result, when the type of the input image word is, at worst, the input image and the output code have the same data amount.

Next, the input image data applicable to the third coding block 5 is classified into types, and the types are
This is the case when neither the type nor the type is applicable.
As shown in FIG. 2, the third coding block 5 is "100".
"0xxxx" is used as the output code word of the header, and 32 bits of the input image word are divided into 8 bits and are sequentially output. In this case, with respect to the input image of 32 bits, the output code of 8 bits is 5 words and the total is 40 bits.
The amount of data increases 1.25 times.

It should be noted that "x" of "1000xxxx" in FIG.
The bit of “xxx” is don't care.

As described above, the image compression apparatus 1 according to the present embodiment separates the input image data by the image separation block 2 in units of 32 bits per word, and separates the separated image data in units of 1 word. 1 out of a plurality of coding blocks 3 to 5 which are compressed by a specialized compression method
The data is compressed by the coding blocks 3 to 5 corresponding to the type of image data in word units, and 8 bits are output as code data in 1 word units.

Therefore, it is possible to improve the processing speed and the compression rate by using a simple algorithm for easy hardware implementation.

In the above processing, the image compression device 1
Is the same as the output code word (8 bits), if the same value does not continue,
In the case of consecutive times, the number of times the same value continues after that may be counted and the counted number may be used as the code word.

However, since the output code word is 8 bits, the maximum count number is set to "255", and if the same value continues for more than that, the next count number is put in the next output code word. Also, the count number is exactly "255"
If it ends with, put "0" in the next output code word.

In this way, the feature of the output code data can be used to perform the compression for further reducing the data amount to further increase the compression rate, and the compression rate can be achieved even when the image data is data that can be easily compressed. Can be prevented and a better compression can be achieved.

Further, in this case, when the output code word value (8 bits) is not continuous with "0xFF", the image compression apparatus 1 regards it as the output code word as it is, and when it continues twice, how many times thereafter. Count whether it continues,
The count number may be used as the code word.

However, since the output code word is 8 bits, the maximum count number is set to "255", and if it continues beyond this, the subsequent count number is put in the next output code word.
Also, if the count number ends with just "255",
Place a "0" in the next output codeword.

By doing so, the feature of the output code data can be used to perform the compression for further reducing the data amount to further increase the compression rate, and even if the image data is the data which can be easily compressed, the compression rate is increased. Can be prevented and a better compression can be achieved.

FIGS. 4 to 6 are views showing the second embodiment of the image compression apparatus and the image expansion apparatus of the present invention.
FIG. 9 is a block diagram of the main parts of an image decompression device 10 to which the second embodiment of the image compression device and the image decompression device of the present invention has been applied.

In FIG. 4, an image decompression device 10 decompresses the data encoded by the image compression device 1 of the first embodiment, and includes a code separation block (code separation means) 11 and a first code separation block. To 3rd expansion block (expansion means) 12-14.

As described above, the image decompressing device 10 is configured such that the code data encoded by the image compressing device 1 is composed of 8 bits in units of 1 word, and therefore the code data sequentially input in the raster format is used. , 8 bit units.

That is, 8 bits of code data are input to the code separation block 11 on a word-by-word basis. In the code separation block 11, the input code word is one of the types of the input code words shown in FIG. The type and the code word of the type are identified and separated.
The code separation block 11 inputs the separated code data to the first decompression block 12 for the type, to the second decompression block 13 for the type, and to the third decompression block 14 for the type, respectively. , Each extension block 12
Decompress to image data at ~ 14. However, the code separation block 11 divides 8-bit code data into 4-bit units for the type of code data, and inputs the higher-order code data in two.

The code data of the type is 8 bits.
However, since the code data classified into types and types is one code data that spans a plurality of code words, the code separation block 11 has the type and type previously. If there is a code word of, the code is not separated until the code data passes, and the code word is input to the same decompression blocks 12-14.

Each of the decompression blocks 12 to 14 decompresses the input code data and restores 32-bit image data for one code data.

Next, the processing in the code separation block 11 and the decompression blocks 12 to 14 will be described for each type of code data.

First, the code separation block 11 has 8 bits.
If the most significant bit of the input code word is 0, the input code word is input to the first decompression block 12 as the code data belonging to the type shown in FIG. The first decompression block 12 decompresses this type of code data and outputs an output image word.

In the code data of the type, a 4-bit code is divided into an upper bit and a lower bit in one 8-bit code word.
The data is packed in one, and in the order, the upper data is first and the lower data is later. These code data are
When one is decoded and expanded into image data, 32 bits of image data are obtained. Depending on the code word, the high-order code can be expanded into a 32-bit image, but the low-order code is a pass code of "0110", and there is a case where there is no image to be expanded. This is because the code data next to the higher-order code data is a type or a type that is an 8-bit code described later, and cannot be packed with the type code.

In this type of code data, as shown in FIG. 5, when the 4-bit code data is "0000", the decompressed 32-bit image data are all "0".
Is. Similarly, the 4-bit code data is “111
If it is "1", all 32 bits are "1" and 4b
If the code data of it is “0110”, it is a pass code, and thus the first decompression block 12 does not perform decompression processing. In addition, the code data is "0001" to "01".
In the case of "01", the pattern is shown in FIG. 5 but is as shown in FIG. That is, as shown in FIG. 6, the pattern has a length of 32 bi that has been expanded one before.
This is a pattern for expanding the same image data as the image data of t.
t This is a pattern for expanding the same image as the image data that has been shifted in the past.
It is a pattern for expanding the same image as the image data shifted in the past.

Next, the code separation block 11 has 8 bits.
If the most significant bit of the input code word of 1 is 1 and the next 3 bits are not “000”, the input code word is regarded as code data belonging to the type of FIG. input. The second expansion block 13 expands this type of code data and outputs an output image word. However, the type code data is 1
The number of code words is not limited to one code word and may extend to a maximum of four code words, and the leading code word is set as the header code word of the type.

In the input code word, the length of the code data is determined by the next 3 bits of the most significant bit of the header code word, and this value represents the number of sets in the code data. Take a value. The set is a 4-bit code, and the first set is in the lower 4 bits of the header code word. The next set is the upper 4 bits of the next code word
And the next set is in the lower 4 bits. In this way, every set other than the first set is packed into two sets to form an 8-bit code word.

The maximum of code data is when the number of sets is 7,
The code data is spread over four code words. However, when the number of pairs is an even number, the lower 4 bits of the last code word are left over.
on't care.

The meaning of the set is to represent run lengths of "1" and "0" in order from the MSB side of the image data of 32 bits representing up to 15 (4 bits) in the LSB direction. The value of the first set is defined as "0", and the first b
When it is "1", the count number of the first set is "0"
It is a group. In addition, the number of counts in the set is
Since the maximum number is 15 (4 bits), if the same value continues for more than that, the next set is also a set of the same value, and the subsequent count number is included. Even if the count number ends with just “15”, the next set also has the same value, so the count number “0” is entered in the next set. The information about the last set is omitted because it can be decoded (decompressed) without it. The count number of the last group from the last,
In the case of "15", the information about the group immediately before the last is also omitted. Further, when the count number of the last set and the count number of the last set are both “15”, the last set and the last set are also omitted. The code data of the type is configured as described above, and the second decompression block 13 has one code data and 32 bits.
Expand the image data of.

Further, the code separation block 11 has 8 bi
When the upper 4 bits of the input code word of t is “1000”, the input code word is input to the third expansion block 14 as the code data belonging to the type of FIG. The third expansion block 14 expands this type of code data and outputs an output image word. However,
The type code data is code data that spans five code words, and the leading code word is the type header code word.

In the code data of the type, the image data before encoding is contained in every 4 bits in the four code words other than the header code word.
32 is obtained by arranging the code words in order from the higher order.
The output image word is output by restoring the bit image data.

As described above, the image expansion apparatus 10 of the present embodiment separates the input code data, which is compressed by different compression methods and has 8 bits as one word unit, by the code separation block 11 according to the type and the separated code. The data is input to the decompression blocks 12 to 14 that are adapted to the type of the code data among the plurality of decompression blocks 12 to 14 that are decompressed by the decompression method specialized for each type of the code data. Extend from 12 to 4 to 32
Bits are output as image data in units of 1 word.

Therefore, the code data is decompressed by a simple algorithm to facilitate hardware implementation,
The processing speed can be improved, and the coded data can be decoded into image data without degrading the image.

FIG. 7 is a diagram showing a third embodiment of the image compression apparatus and the image expansion apparatus of the present invention, and FIG. 7 is a third embodiment of the image compression apparatus and the image expansion apparatus of the present invention. 3 is a block diagram of a main part of an image expansion device 20 to which a form is applied. FIG.

In FIG. 7, the image decompression device 20 is provided with an image decompressor 21 in the preceding stage of the image decompression device 10 according to the second embodiment. This is similar to the second embodiment.

The image decompressor (predecompression means) 21 has a first
In the image compression device 1 according to the embodiment of the present invention, the code word generated by encoding the image data is used as the output code word as it is when the code words (8 bits) are not consecutive in the same value, and the same value is used. If is repeated twice,
After that, the number of times the same value continues is counted, and the code word having the count number as the code word is input.

The image decompressor 21 further encodes the code word generated by encoding the image data by the image compression apparatus 1 of the first embodiment, and further the code word (8 bits).
If the same value does not continue, it is used as the output code word as it is, and if the same value continues twice, the number of times the same value continues after that is counted and the count number is taken as the code word. Since the code word is configured in units of 8 bits, the code words sequentially input in this raster format are referred to in units of 8 bits and expanded without degrading the image data.

Then, the image decompressor 21 sets 1 to 8 bits.
The code word input serially as a word unit,
Unless the same value continues, it is output as it is to the code separation block 11 of the image expansion device 10 as output data, that is, as a code word. In addition, the image decompressor 21
If code words with the same value are consecutive, the next code word is set to the count number of the number of times the value of the previous code word is repeated in the future, and the code word is expanded to expand the image. Code separation block 11 of device 10
Output to.

For example, when a code word "..., 44, 44, 05, ..." Is input, the image decompressor 21
"..., 44,44,44,44,44,44,44,
"...", the code data of "05" is changed to the preceding "4".
The code word is expanded assuming that the value "4" is data (count number) repeated 5 times later.

However, since the input code word is 8 bits, the maximum count number is "255". Therefore, for higher counts, the following counts are put in the next codeword, and if the count ends exactly with "255", the next codeword will contain "0". However, since this means the end of the count number, the image decompressor 21 decompresses in consideration of these.

Then, the image expansion device 10 expands the code data expanded by the image expansion device 21 as in the second embodiment.

As described above, the image decompressing device 20 of the present embodiment uses the code data as it is as the input code data when the serial input code data having 8 bits as one word unit does not have the same value. When the same value is input twice to the code separation block 11, the number of times the same value continues after that is counted, and the count number is used as input code data.
An image decompressor 21 is provided as a pre-decompression means for inputting the data into the 1.

Therefore, by utilizing the characteristics of the input code data, the code data can be expanded into the image data with a simple algorithm while further preventing the deterioration of the image.

FIG. 8 is a diagram showing the fourth embodiment of the image compression apparatus and the image expansion apparatus of the present invention, and FIG. 8 is the fourth embodiment of the image compression apparatus and the image expansion apparatus of the present invention. FIG. 3 is a block diagram of a main part of an image expansion device 30 to which a mode is applied.

In FIG. 8, the image decompression device 30 is provided with an image decompression device 31 before the image decompression device 21 of the third embodiment, that is, the image decompression device 10 of the second embodiment. The image decompressor 21 is arranged in the preceding stage, and the image decompressor 31 is further arranged in the preceding stage of the image decompressor 21, and the processing in the image decompressing device 10 is the same as that of the second embodiment. The same processing is performed by the image decompressor 21 as in the third embodiment.

The image decompressor (second front decompression means) 31 includes
In the image compression device 1 of the first embodiment, the code word generated by encoding the image data is used as the output code word as it is when the code word value (8 bit) “0xFF” is not continuous. When the number of consecutive times is two, the number of consecutive times is counted, and the code word having the count number as the code word is input.

Then, the image decompressor 31 outputs the code word compressed by the image compression device 1 of the first embodiment,
Since 8 bits are configured in units of 1 word, code words sequentially input in this raster format are converted into 8 bits.
The image data is expanded without being deteriorated by referring to the t unit.

That is, the image decompressor 31 converts the data serially input in units of 8 bits into one word,
Unless the value "FF" continues, it is output as it is to the image decompressor 21 as output data, that is, as a code word. Further, when the code word having the value “FF” is consecutive twice, the image decompressor 31 sets the next code word to the count number indicating how many times the value of the code word “FF” is repeated. Assuming that the count has been performed, the count number is output to the image decompressor 21 as a code word.

For example, if a code word "..., FF, FF, 05, ..." Is input, the image decompressor 31 is input.
Is "..., FF, FF, FF, FF, FF, FF, F
The code data of "05" such as "F, ..." is assumed to be data (count number) obtained by repeating the previous value of "FF" five times later, and the code word is expanded.

However, since the input code word is 8 bits, the maximum count number is "255". Therefore, for higher counts, the following counts are put in the next codeword, and if the count ends exactly with "255", the next codeword will contain "0". However, since this means the end of the count number, the image decompressor 31 decompresses in consideration of these.

Then, the image decompressor 21 uses the image decompressor 3
The code data decompressed by 1 is decompressed as in the third embodiment and output to the image decompressor 21.
Is to expand the code data expanded by the image expander 21 as in the second embodiment.

As described above, the image decompressing device 30 of the present embodiment, if the value of the serial input code data whose 8-bit unit is 8 bits is not "0xFF" consecutively, the input code data is left as it is. It is input as input code data to the image decompressor 21 as the predecompression means,
When “0xFF” continues twice, the number of times “0xFF” continues after that is counted, and the count number is input to the image decompressor 21 as input code data. An image decompressor 31 is provided.

Therefore, by utilizing the characteristics of the input code data, the code data can be expanded into the image data with a simple algorithm while further preventing the deterioration of the image.

Although the invention made by the present inventor has been specifically described based on the preferred embodiments, the present invention is not limited to the above, and various modifications can be made without departing from the scope of the invention. It goes without saying that it is possible.

[0092]

According to the image compressing apparatus of the present invention, the input image data is separated by the image separating means in units of 32 bits per word, and the separated image data in units of 1 word is specified. Of the plurality of compression means for compressing by the compressed compression method, the compression means adapted to the type of the image data of the 1-word unit is used to output 8 bit as the code data of 1-word unit, so the compression is performed by a simple algorithm. Then, it is possible to improve the processing speed by making it easy to implement hardware,
The compression rate can be improved.

According to the image compression apparatus of the second aspect of the present invention, when the serial code data having the 8-bit unit output by the compression means does not have the same value, the code data is output as it is. In the case of code data, if the same value continues twice, the number of times the same value continues after that is counted and the count number is used as output code data. , It is possible to further reduce the amount of data by performing compression to further improve the compression rate, and prevent the result of low compression rate even when the image data is data that is easily compressed, and perform better compression. You can

According to the image compression apparatus of the third aspect of the invention, when the value of the serial code data in which one bit unit is 8 bits output from the compression means is 0xFF, the code data is left as it is. If the output code data is 0xFF twice consecutively, then 0
Since it is counted whether xFF is continuous and the counted number is used as the output code data, it is possible to perform the compression for further reducing the data amount by using the feature of the output code data and further increase the compression rate. Even if the image data is data that can be easily compressed, it is possible to prevent the result that the compression rate is low and perform more favorable compression.

According to the image decompressing device of the invention described in claim 4, the input code data, which is compressed by different compression methods and has 8 bits as one word unit, is separated by the code separating means by the type, and the separated code data is Of a plurality of decompression means that decompress by a decompression method specialized for each type of code data, it is input to a decompression means adapted to the type of the code data and decompressed by the decompression means to make 32 bits in 1-word units. Since it is output as image data, the code data can be decompressed with a simple algorithm to facilitate hardware implementation, improving the processing speed and decoding the code data into image data without degrading the image. can do.

According to the image decompressing device of the fifth aspect of the invention, when serial input code data having 8 bits as one word unit does not have the same value, the code data is directly coded as the input code data. When the same value is input twice to the separating means, the pre-expansion means that counts how many times the same value continues after that and inputs the count number to the code separating means as the input code data. Since it is provided, the characteristic of the input code data can be used to expand the code data into the image data with a simple algorithm while further preventing the deterioration of the image.

According to the image decompressing device of the sixth aspect of the invention, when the value of serial input code data having 8 bits as one word unit is not 0xFF consecutively, the input code data is directly input code data. When the 0xFF continues twice, the number of times 0xFF continues after that is counted, and the count number is input to the pre-expansion means as input code data. Since the means is provided, the characteristic of the input code data can be used to expand the code data into the image data with a simple algorithm while further preventing the deterioration of the image.

[Brief description of drawings]

FIG. 1 is a first image compression device and image expansion device of the present invention.
FIG. 3 is a block diagram of a main part of an image compression apparatus to which the embodiment of FIG.

FIG. 2 is a diagram showing the types of input image words of the image compression apparatus shown in FIG.
The figure which shows output code data and its description.

FIG. 3 is an explanatory diagram of the pattern of FIG.

FIG. 4 is a second image compression device and image expansion device of the present invention.
FIG. 3 is a block diagram of a main part of an image expansion device to which the embodiment of FIG.

FIG. 5 is a kind of input codeword of the image compression apparatus of FIG.
The figure which shows code data and its description.

FIG. 6 is an explanatory view of the pattern of FIG.

FIG. 7 is a third image compression device and image expansion device according to the present invention.
FIG. 3 is a block diagram of a main part of an image expansion device to which the embodiment of FIG.

FIG. 8 is a fourth image compression apparatus and image expansion apparatus of the present invention.
FIG. 3 is a block diagram of a main part of an image expansion device to which the embodiment of FIG.

[Explanation of symbols]

1 Image compression device 2 image separation block 3 First coded block 4 Second coded block 5 Third coded block 10 Image decompression device 11 code separation block 12 First extension block 13 Second expansion block 14 Third extension block 20 Image decompression device 21 Image Decompressor 30 image decompressor 31 Image decompressor

Claims (6)

[Claims] 1. An image separation means for separating input image data by type in units of 32 bits, and a compression method specialized for each type of image data in 1 word units separated by the image separation means. And a plurality of compression means for compressing the input image data according to the type by the image separation means, and the separated one-word-unit image data among the plurality of compression means. An image compression apparatus characterized by inputting to a compression unit adapted to the type of image data in units of 1 word, compressing by the compression unit and outputting as coded data in units of 8 bits. 2. The image compressing device, when serial code data having 8 bits as one word unit output from the compressing means does not continuously have the same value, the code data is directly used as output code data, and the same. 2. The image compression apparatus according to claim 1, wherein when the value continues twice, the number of times the same value continues after that is counted and the counted number is used as output code data. 3. The image compression apparatus, when the value of serial code data in which 1 bit unit is 8 bits output from the compression means is not 0xFF consecutively, the code data is directly used as output code data, The image compression apparatus according to claim 1, wherein, when 0xFF continues twice, the number of times 0xFF continues after that is counted and the counted number is used as output code data. 4. Code separating means for separating input code data compressed by different compression methods and having 8 bits as one word unit, and code data separated by the code separating means for each type of the code data. An image decompression device comprising a plurality of decompression means for decompressing with a specialized decompression method, wherein the input code data is separated by the code separation means by type, and the separated code data is divided into the plurality of expansion means. An image decompression device, wherein the image decompression device inputs the decompression means adapted to the type of the code data, decompresses the decompression means, and outputs 32 bits as image data in 1-word units. 5. The image decompressing device, when serial input code data having the 8-bit unit as a word unit does not have the same value, the code data is directly input to the code separating means as the input code data. When the same value is input twice and the same value continues twice, a pre-expansion means is provided to count how many times the same value continues after that and input the count number to the code separation means as the input code data. The image expansion device according to claim 4, wherein 6. The value of serial input code data in which the image decompressing device has 8 bits as one word unit is 0x.
When the FFs are not continuous, the input code data is directly input to the pre-expansion means as the input code data, and when 0xFF is continuous twice, it is 0 times thereafter.
The image decompression device according to claim 5, further comprising a second pre-expansion unit that counts whether xFFs are continuous and inputs the count number to the pre-expansion unit as the input code data.