JP3212393B2 - Encoding device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an encoding apparatus for arbitrarily selecting one of image data and character code data and compressing the data.

[0002]

2. Description of the Related Art When transmitting and storing data, data compression is often performed. As a coding method for performing such data compression, an arithmetic code and a Ziv-Lempel universal code are known.

[0003] Arithmetic codes are one type of predictive coding method, and are suitable for data of Markov information sources and can be adapted according to the characteristics of the data. I have. On the other hand, a universal code has a high compression effect on data in which long data of the same symbol pattern repeatedly appears, and is therefore often used for data compression of character code data.

In some cases, it is desired to select image data and character code data arbitrarily and compress the data.

Conventionally, in such a case, two types of coding methods, such as an arithmetic code and a universal code, have been selectively used. For this reason, two types of encoding means must be provided in the device, and the device is complicated.

On the other hand, as disclosed in, for example, Japanese Patent Application Laid-Open No. 3-70268, there has been proposed a technique in which both image data and character data are compressed by a universal code.

In this proposal, character code data is compressed by a universal code in byte units. The image data is MH (Modified H).
uffman), MR (Modified Relat)
live Element Address Design
nate) or MMR (Modified MR)
After encoding by a method, a fixed length code is allocated to a run length code and mode information obtained by the encoding, and the fixed length code is data-compressed by a universal code.

Accordingly, in the above proposal, not only the encoding means for the universal code but also another encoding means such as MH, MR or MMR is required.
The equipment was complicated.

[0009]

As described above, conventionally,
In the case of encoding image data and character code data, a plurality of encoding units are required, and there is a problem that the apparatus becomes complicated.

An object of the present invention is to solve the above-mentioned problems and to provide an encoding apparatus capable of compressing image data and character code data with a simple apparatus configuration.

[0011]

According to the present invention, image data or character code data is arbitrarily input, and the character code data is developed into bitmap data corresponding to one bit per character. Then, the expanded bitmap data or image data is compressed by an arithmetic coding method.

When character code data is developed into bitmap data, search character codes are generated one by one in a predetermined order, and each character code of the character code data is collated with the search character code. Bitmap data is generated by matching / mismatching.

[0013]

Since both image data and character code data are encoded by arithmetic codes, only one encoding means is required, so that the apparatus configuration is simplified. After expanding the character code data into bitmap data corresponding to one bit per character,
Since encoding is performed using arithmetic codes, a higher data compression effect can be obtained than when character code data is directly encoded.

[0014]

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

First, a first embodiment of the present invention will be described.
FIG. 1 is a block diagram illustrating the configuration of an encoding apparatus according to the present embodiment. In the figure, a memory 1 temporarily stores input image data or character code data.
The switching circuit 2 switches a signal route according to the type of input data. The bitmap developing unit 3
It converts character code data into bitmap data corresponding to one bit per character. The encoding unit 4 compresses data by an arithmetic code encoding method.

In the encoding unit 4, a reference symbol acquisition unit 41 focuses on each bit of data to be processed and reads a symbol of a plurality of bits at a fixed position with respect to the bit. The prediction unit 42 predicts the symbol of the target bit with reference to the read bits. The arithmetic code generation unit 43 generates an arithmetic code based on the prediction result.

FIG. 2 is a block diagram showing the configuration of the decoding apparatus according to the present embodiment. In the figure, a decoding unit 5 restores code information of an arithmetic code to original image data and bitmap data. The switching circuit 6 switches the signal route according to the type of data. The character data restoring unit 7 restores bitmap data to character code data. The memory 8 temporarily stores image data and character code data.

In the decoding unit 5, the arithmetic code decoding unit 5
Numeral 1 restores the code information of the arithmetic code to the data before encoding one bit at a time. Reference symbol acquisition unit 52
Focuses on each bit of the restored data, and reads each bit at a fixed position, similarly to the reference symbol acquisition unit 41 of FIG. The prediction unit 53 predicts a target bit with reference to the read symbol of each bit.

With the above configuration, image data or character code data is input to the encoding device of the present embodiment.
The image data is, for example, black-and-white image data obtained by reading a document image line by line by a scanner device. Character code data is, for example, ASCII
(American National Standa
rd Code for Information In
This is data of character information of a (change) code.

The input data is temporarily stored in the memory 1 and then read out bit by bit.

When processing image data, the image data read from the memory 1 is directly input to the encoding unit 4 by switching of the switching circuit 2.

The reference symbol obtaining means 41 pays attention to each bit of the input image data and reads each bit data in a certain range around the bit of interest. The prediction unit 42
The symbol of the target bit is predicted based on the read symbol of each bit. The arithmetic code generation unit 43 generates an arithmetic code based on the predicted value. As a result, the input image data is subjected to data compression by encoding.

On the other hand, when processing character code data,
The character code data read from the memory 1 is input to the bitmap developing unit 3 by switching of the switching circuit 2.

FIG. 3 shows the operation of the bitmap developing section 3. That is, when activated, the bitmap developing unit 3 sets the variable CODE for character code search to “0”.
0 ”(process 101). Then, one byte of the input character code data is read (process 102). This one
A byte corresponds to one character and represents one character code.

Next, the 1-byte data is collated with the variable CODE (step 103). Here, if they match (Y in processing 103), 1-bit data “1” is output to the encoding unit 4 (processing 104), and if they do not match (N in processing 103), the data “0” is output. Output (process 105).

Then, the input of the character code data is checked (step 106), and if data is still present (step 1).
(N of 06) The same process is repeated for each byte (to process 102). In this way, when all the bits of the character code data are completely processed (Y in process 106), it is determined whether or not the variable CODE has reached “FFh” (“h” indicates a hexadecimal number). (Step 107).

If the variable CODE has not reached "FFh" (N in process 107), the variable CODE is incremented by 1 (process 108), and the read position of the character code data is returned to the beginning (process 109).

Then, the same processing is repeated (processing 10
2). When the variable CODE reaches “FFh” (Y in step 107), the operation ends.

Now, as a simple example, assume that the character code data is six-character data "abcabc". In the ASCII code, the character "a" is changed to "61h",
“B” is “62h” and “c” is “63h”. Therefore, in this case, as shown in FIG.
Are set to 61h to 63h, a match between the character codes is detected. That is, in this example, it can be considered that one surface of the character data is developed into three surfaces corresponding to each character. Then, bitmap data is generated for each surface.

Next, assume that the character code data is data of a C language source program as shown in FIG. In this case, as shown in FIG.
When it becomes "20h" indicating a space, "00000"
.0000 "is obtained. The variable CODE is set to “23” indicating the symbol “#”.
h ", bitmap data" 100... "is obtained. Also, “69” indicating the character “i”
When "h" is reached, bit map data "010..." is obtained. The bitmap data thus obtained is input to the encoding unit 4.

The encoding unit 4 encodes the input bitmap data in the same manner as the image data. In addition,
In this case, individual bitmap data corresponding to each character is processed as one page of image data. Further, the reference symbol acquisition means 41 arranges the input bitmap data for each line of the image data,
Each bit at a certain position around the target bit is read.

Next, when decoding the code information generated as described above, the code information is transmitted to the decoding unit 5 of the decoding device.
Is input to The arithmetic code decoding unit 51 restores the data one bit at a time based on the input code information and the prediction result of the symbol of the pixel of interest output from the prediction unit 53. The reference symbol acquisition unit 52 pays attention to each bit of the data to be restored, and reads a symbol of a plurality of bits at a fixed position with respect to the bit. The prediction unit 53 predicts the symbol of the target bit with reference to the read symbol of each bit. The arithmetic code decoding unit 51 receives the prediction result and executes the decoding.

When the image data is restored by the decoding unit 5, the image data is directly input to the memory 8 and temporarily stored by a predetermined switching of the switching circuit 6.

When the bitmap data as the character information is restored, it is input to the character data restoring unit 7. The character data restoring unit 7 generates character codes one by one from the input bitmap data, arranges the generated character codes, and restores the original character code data.
The restored character code data is input to the memory 8 and is temporarily stored.

The image data and character code data stored in the memory 8 are output to another device as needed.

As described above, in this embodiment, the image data is encoded by the encoding unit 4 as it is, while the character code data is converted by the bitmap development unit 3 into bitmap data corresponding to one bit per character. After that, the encoding unit 4 performs encoding. Thus, the encoding circuit can compress the binary image data and the character code data only by the encoding unit 4, thereby simplifying the device configuration.

Also, the decoding device is also composed of only the decoding unit 5,
Since binary image data and character code data can be restored,
Similarly, the device configuration is simplified.

When character code data is developed into bitmap data, search character codes are generated one by one in a fixed order, and the character codes of the character code data are compared with the search character codes. , Bitmap data is generated.

In this case, bit "1" in the obtained bitmap data corresponds to a position where the same character appears. In general, when attention is paid to one character in one document, there is a high probability that the same character is found at a specific position around the one character. Therefore, since such bitmap data is decoded by the arithmetic coding method, a relatively high data compression effect can be obtained.

Next, a second embodiment of the present invention will be described.

This embodiment is a partial modification of the character code data processing method described with reference to FIG. That is,
In the present embodiment, as shown in FIG. 7, the value of the variable CODE matches the character code (Y in process 103), and the encoding unit 4
When the data "1" is output to the corresponding character code, the corresponding one byte in the character code data is deleted (process 110).

Now, for example, as shown in FIG.
It is assumed that the character code data “abcabc” is processed in this manner. Then, in the processing of the character “a”, “100”
The bitmap data “100100” is
Is output to In this process, the character code data of “a” is deleted, so that data of “bcbcbc” remains. Next, bitmap data “101010” is output in the processing of the character “b”. At this time, since the character code data of "b" is deleted, data of "ccc" remains. Then, in the processing of the character "c", "111"
Is output, and the process ends.

The encoding unit 4 encodes the input bitmap data into predetermined code information in the same manner as in the above embodiment.

On the other hand, when decoding the code information, the decoding unit 5 restores the code information to bitmap data as in the above-described embodiment. Then, the character data restoration unit 7
The bitmap data is restored to the original character code data.

Here, it is assumed that the character data restoration unit 7 restores the bitmap data created in the above example. In this case, the character data restoration unit 7 first sets “1001
"00100" is input. In this case, it is determined that the total number of characters is 9 based on the number of bits. Then, as shown in FIG. 9, of the nine characters, the character corresponding to bit “1” is determined to be “a”. Next, when the bitmap data “101010” is input, the 6 bits are made to correspond to positions of six characters that have not been determined yet. Then, the character corresponding to the bit “1” of the bitmap data is determined to be “b”. Next, when the bitmap data “111” is input, the three bits are made to correspond to the positions of three characters that have not been determined. Then, the character corresponding to the bit “1” of the bitmap data is determined to be “c”. In this way, the original nine-character character code data "abcab
cabc "is restored.

As described above, in the present embodiment, when one character is converted into 1-bit bitmap data when the character code data is developed into bitmap data, the corresponding character code data is deleted. ing. As a result, the character code data to be processed decreases as the processing proceeds, so that the data amount of the generated bitmap data can be reduced. Therefore, the encoding process can be performed in a short time, and the generated code information is reduced.

Next, a third embodiment of the present invention will be described.

This embodiment is a modification of the character code data processing method described with reference to FIG. That is, in this embodiment, as shown in FIG. 10, first, all the used character codes are extracted from the character code data stored in the memory 1. Here, assuming that n character codes are extracted, the extracted codes are CODE1 to C
ODEn (process 201).

Then, the codes CODE1 to CODE
For only n, the expansion processing from character code data to bitmap data is executed. That is, the difference from the processing in FIG. 3 is that in FIG.
Although the expansion processing from character code data to bitmap data is executed for all characters up to “FFh”, in this embodiment, the expansion processing is executed only for characters included in the character code data. (Process 20
2).

In the processing of FIG. 3, when a character not included in the character code data is expanded, bitmap data of all “0” is generated. In the present embodiment, such bitmap data is generated. Can be generated in a short time. When the decoding device restores the bitmap data to the original character code data, the bitmap data having all “0” s may not be present.

Next, a fourth embodiment of the present invention will be described.

This embodiment is a further modification of the processing method described above with reference to FIG. That is, in the present embodiment, as shown in FIG. 11, all the character codes CODE1 to CODEn in the character code data are
Is extracted (process 301). Then, the extracted character codes CODE1 to CODEn are rearranged in the order of the appearance frequency in the character code data (process 203). Then, in the rearranged order, expansion processing from character code data to bitmap data is executed (process 204).

Here, for example, consider the case of processing character code data of 13 characters "aaaababaababa" as shown in FIG. When this character code data is developed into bitmap data in the order of characters "a" and "b", as shown in FIG. 2B, 13 bits for character "a" and 9 bits for character "b" Bitmap data is obtained respectively, for a total of 22 bits.

On the other hand, when the data is expanded into bitmap data in the order of the characters "b" and "a", as shown in FIG.
13-bit bitmap data is obtained for the character "a" and 4-bit bitmap data is obtained for the character "b".

In the case of the above character code, the appearance frequency of the character "a" is high. Therefore, it can be seen that processing the character having the higher appearance frequency first can reduce the number of bits of the bitmap data.

In this embodiment, a plurality of characters are processed in descending order of appearance frequency, so that the number of bits of bitmap data can be reduced most.

Next, a fifth embodiment of the present invention will be described.

This embodiment is a further modification of the processing method described with reference to FIG. That is, in this embodiment, as shown in FIG. 13, character codes CODE1 to CODE256 of various characters are stored in advance in various character code data in general in descending order of appearance probability.

Then, as shown in FIG. 14, expansion processing from character code data to bitmap data is executed in the order of the stored appearance probabilities (step 205).

According to this processing method, FIG.
As in the first embodiment, it is not necessary to extract the character code used in the character code data. In addition, it is also possible to simultaneously process the character code data to be processed while sequentially inputting the character code data without temporarily storing the data in the memory 1.

Next, a sixth embodiment of the present invention will be described.

In this embodiment, the prediction unit 42 of the encoding unit 4
As shown in FIG. 15, when predicting the target bit “*” of the bitmap data, the immediately preceding bit position # 1 and the preceding bit position # 2 are referred to. However, in this case, when predicting bitmap data corresponding to one character, reference is made to bitmap data corresponding to another character. For this reference, the correspondence between the characters of the bitmap data to be processed and the characters of the bitmap data to be referred to is stored in advance.

For example, in the character code data to be processed, for example, as shown in FIG.
It is assumed that various character strings “YZ” frequently appear. These character strings are, for example, character strings constituting words in a sentence.

In this case, as shown in FIG. 17, when the processing target is "c", it indicates that bit position # 1 is "b" and bit position # 2 is "a". When the processing target is “c”, it indicates that bit position # 1 is “b” and bit position # 2 is “a”. Further, when the processing target is “Z”, it indicates that bit position # 1 is “Y” and bit position # 2 is “X”.

The prediction unit 42 predicts a symbol of each bit of the bitmap data under the set conditions. For example, when predicting the target bit “*” of the bitmap data corresponding to the character “c”, the bit position # 1 of the bitmap data corresponding to the character “b” and the bitmap data corresponding to the character “c” Refer to the 2-bit symbol at bit position # 2. If the two bits are both “1”, it is predicted that the target bit “*” becomes “1” with a certain high probability. That is, the fact that the two bits are both "1" means that two characters of "ab" appear continuously, and the probability that "c" appears next is high.

Similarly, when the target bit “*” of the bitmap data corresponding to the character “Z” is predicted, the bit position # of the bitmap data corresponding to the character “Y”
1 and a 2-bit symbol at bit position # 2 of the bitmap data corresponding to the character "X". And
If the two bits are both “1”, the bit of interest “*”
Is predicted to be “1” with a certain high probability.

In general, when a target bit of various data is predicted from surrounding bits, the surrounding bit refers to the same data as the target bit. In this embodiment, however, the target bit is referred to by referring to another data. Prediction can be performed with relatively high prediction accuracy.

In each of the above embodiments, the character code is an ASCII code and is represented by one byte per character. However, it goes without saying that the number of bits can be set arbitrarily. The present invention is not limited to character code data representing character information, and can be applied to a case where various code data and image data are arbitrarily encoded / decoded.

[0069]

As described above, according to the present invention, character code data is developed into bitmap data corresponding to one bit per character, and the bitmap data or image data is encoded by an arithmetic coding method. In this case, since only one encoding unit is required, character data and image data can be compressed with a simple device configuration. When character code data is expanded into bitmap data, search character codes are generated one by one in a fixed order, and bitmap data is generated based on the result of matching each character code of the character code data with the search character code. Is generated, so that a higher data compression effect can be obtained than when character code data is directly encoded.

[Brief description of the drawings]

FIG. 1 is a block diagram of an encoding device according to a first embodiment of the present invention.

FIG. 2 is a block diagram of a decoding device in the embodiment.

FIG. 3 is an operation flowchart illustrating a process performed on character code data by a bitmap developing unit.

FIG. 4 is a conceptual diagram of the processing for the character code data.

FIG. 5 is a character information diagram showing an example of character code data.

FIG. 6 is an explanatory diagram of bitmap data obtained from the character code data.

FIG. 7 is a flowchart showing points different from FIG. 3 in the processing of the second embodiment of the present invention.

FIG. 8 is an explanatory diagram showing bitmap data expansion processing in the embodiment.

FIG. 9 is an explanatory diagram of a process of restoring original character code data from bitmap data.

FIG. 10 is a flowchart of a bitmap data expansion process according to a third embodiment of the present invention.

FIG. 11 is a flowchart of a bitmap data expansion process according to a fourth embodiment of the present invention.

FIG. 12 is an explanatory diagram of the operation and effect of the embodiment.

FIG. 13 is an explanatory diagram of information stored in advance in a fifth embodiment of the present invention.

FIG. 14 is an operation flowchart of a bitmap data expansion process in the embodiment.

FIG. 15 is an explanatory diagram showing bit positions referred to during a prediction operation in a sixth embodiment of the present invention.

FIG. 16 is an explanatory diagram showing an example of a character string that frequently appears in character code data to be processed.

FIG. 17 is an explanatory diagram of correspondence information of each character stored in advance.

[Explanation of symbols]

 Reference Signs List 1, 8 Memory 2, 6 Switching circuit 3 Bitmap expansion unit 4 Encoding unit 5 Decoding unit 7 Character data restoration unit 41 Reference symbol acquisition unit 42, 53 Prediction unit 43 Arithmetic code generation unit 51 Arithmetic code decoding unit 52 Reference symbol Acquisition unit

Claims (7)

(57) [Claims] 1. Data input means for arbitrarily inputting image data or character code data, bitmap expanding means for expanding input character code data into bitmap data corresponding to one bit per character, A coding means for compressing the data or the expanded bitmap data by an arithmetic coding method. 2. The bitmap developing means according to claim 1, wherein said code generating means generates search character codes one by one in a predetermined order, and checks each character code of said input character code data with said search character code. 2. The encoding apparatus according to claim 1, further comprising bitmap data generating means for generating bitmap data by matching / mismatching. 3. A means for creating the bitmap data for one search character code, erasing a corresponding character code in the character code data, and reducing a data amount of bitmap data generated thereafter. 3. The encoding device according to claim 2, wherein the encoding device is provided. 4. The encoding apparatus according to claim 2, wherein said code generating means is means for generating only a character code included in the input character code data. 5. The encoding apparatus according to claim 2, wherein said code generation means is means for generating each character code in the order of appearance frequency in the input character code data. 6. The encoding apparatus according to claim 2, wherein said code generation means is means for generating each character code in an order of appearance probability generally in various character code data. 7. The encoding means includes a prediction means for referring to bitmap data corresponding to another search character code when predicting each bit of bitmap data corresponding to one search character code. The encoding device according to claim 1, wherein the encoding device is provided.