JP3603099B2 - Method and apparatus for lossless encoding of data - Google Patents

Description

[0001]
[Industrial applications]
The present invention relates to a method and an apparatus for losslessly encoding data.
[0002]
[Prior art]
In recent years, data compression techniques have been widely used to cope with a rapid increase in the amount of data processed by a computer. Data compression techniques include lossless encoding, in which data before compression is restored as before, and irreversible encoding, in which data is not restored as before. Lossy coding is mainly used for compression of image data. On the other hand, lossless encoding is used for compressing various data such as program data, text data, dictionary data, and the like.
[0003]
Font data is one type of data to be subjected to lossless encoding. Font data is data representing the shape of characters included in one set of fonts. In recent years, in many cases, several types of font data are loaded on a personal computer so that a user can select and use a desired font.
[0004]
[Problems to be solved by the invention]
However, when the font data is compressed by the same reversible encoding method as the text data and the program data, the compression ratio may not be so high. This is considered to be because the font data has a specific data array different from the text data and the program data.
[0005]
SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problems in the related art, and has as its object to provide a lossless encoding method and apparatus capable of efficiently compressing font data.
[0006]
Means and action for solving the problem
In order to solve the above-mentioned problem, a lossless encoding method for data according to claim 1 of the present invention comprises the steps of: (A) examining two consecutive bytes of input data; Determining whether the encoding condition that one byte immediately before is equal to each other and the first byte of the two bytes is equal to or less than a predetermined value is satisfied; and (B) determining whether the encoding condition is satisfied. Encoding the two bytes in a predetermined encoding mode if satisfied.
[0007]
In font data, the trailing byte of two bytes and the one byte immediately before the two bytes are equal to each other, and the leading byte of the two bytes is often equal to or smaller than a predetermined value. Therefore, when such an encoding condition is satisfied, font data can be efficiently compressed by encoding in a predetermined encoding mode suitable for the font data.
[0008]
In the data lossless encoding method according to claim 2, the step (B) includes a step (C) of encoding the two bytes with a code word of less than 16 bits.
[0009]
In this case, the compression ratio can be set to 1 or more.
[0010]
4. The data lossless encoding method according to claim 3, wherein the step (C) includes: a prefix of 8 bits or less indicating that the data is encoded in the predetermined encoding mode; Creating encoded data by combining the bits.
[0011]
In this way, 2 bytes can be encoded with 15 bits or less.
[0012]
According to a fourth aspect of the present invention, (A) two consecutive bytes of input data satisfy a first encoding condition for encoding in a predetermined first encoding mode. And (B) if the first encoding condition is not satisfied, the last byte of the two bytes is equal to the one byte immediately before the two bytes, and Determining whether or not a second encoding condition that the first byte of the two bytes is equal to or smaller than a predetermined value is satisfied; and (C) determining whether the first determination condition is satisfied. Encoding bytes in the first encoding mode having the highest compression ratio, and encoding the two bytes in the second encoding mode having a medium compression ratio when the second encoding condition is satisfied. , When neither the first nor the second encoding condition is satisfied Characterized in that it comprises a step of at least the compression ratio of the first byte is encoded in the lowest third encoding mode.
[0013]
If the encoding mode is selected based on the encoding condition, an appropriate encoding mode can be selected according to the content of the 2-byte data. In particular, when the two bytes satisfy a second encoding condition specific to font data, the font data is efficiently compressed by encoding in a second encoding mode suitable for the font data. Can be.
[0014]
8. The lossless encoding method of data according to claim 7, wherein the encoding step according to the first encoding mode includes the step of encoding a longest second byte sequence that matches a first byte sequence including at least the two bytes. Searching in the input data, a first codeword indicating that the encoding is performed in the first encoding mode, and a second codeword indicating a distance between the first and second byte strings. And generating a coded data by combining the third byte with the third codeword indicating the length of the first byte sequence.
[0015]
In the first encoding mode, since the longest matching byte string is encoded, the compression ratio can be increased.
[0016]
9. The data lossless encoding method according to claim 8, wherein the third encoding mode includes a prefix indicating that encoding is performed in the third encoding mode, and lower 7 bits of the first byte. Obtaining the encoded data by combining the first byte with the first byte. The prefix in the third encoding mode is a 1-bit first byte used when the hexadecimal notation of the first byte is 7F or less. It is selected from one of a prefix and a multi-bit second prefix used when the hexadecimal notation of the first byte is 80 or more.
[0017]
In the third encoding mode, when the hexadecimal notation of the first byte in two bytes is 7F or less, encoding is performed with 8 bits, so that the compression ratio does not become excessively low.
[0018]
The lossless encoding method for data according to claim 9, wherein it is determined whether or not the appearance rate of the first type of byte data whose hexadecimal notation is 7F or less in input data satisfies a predetermined condition. If the appearance rate of the first type of byte data satisfies a predetermined condition, (A) two consecutive bytes in the input data are encoded in an encoding mode in which the compression rate exceeds 1. (B) determining whether the encoding condition is satisfied, and if the encoding condition is not satisfied, sets at least the first byte of the two bytes to an encoding mode having a compression ratio of 1 or less. And encoding.
[0019]
When the encoding including the steps (A) and (B) is executed when the appearance rate of byte data whose hexadecimal notation is 7F or less satisfies a predetermined condition, when byte data of 7F or less is encoded Can be set so that the compression ratio of the code word does not become excessively low.
[0020]
11. The data lossless encoding method according to claim 10, wherein in the step (B), a prefix indicating that encoding has been performed in the encoding mode having a compression ratio of 1 or less, and a lower 7 bytes of the first byte. Obtaining encoded data by combining the first and second bits with each other, wherein the prefix is a 1-bit first prefix used when the hexadecimal notation of the first byte is 7F or less, and It is selected from either one of the multi-bit second prefix used when the hexadecimal notation of the first byte is 80 or more.
[0021]
Since byte data of 7F or less can be encoded with 8 bits, the compression ratio at the time of encoding does not become excessively low.
[0022]
12. The data lossless encoding apparatus according to claim 11, wherein two consecutive bytes of the input data are checked, and a trailing byte of the two bytes and a byte immediately before the two bytes are equal to each other, and Determining means for determining whether or not an encoding condition that the first byte of the byte is equal to or less than a predetermined value is satisfied; and, if the encoding condition is satisfied, encoding the two bytes in a predetermined encoding mode. Encoding means for encoding.
[0023]
According to a fourteenth aspect of the present invention, in the data lossless encoding apparatus, two consecutive bytes of the input data satisfy a first encoding condition for encoding in a predetermined first encoding mode. First determining means for determining whether or not the first encoding condition is not satisfied; and a trailing byte of the two bytes and a byte immediately before the two bytes are equal to each other, and Second means for determining whether or not a second encoding condition that the first byte of the two bytes is equal to or less than a predetermined value is satisfied; and if the first determination condition is satisfied, at least the second condition is satisfied. Encoding a byte in the first encoding mode, and encoding the two bytes in a second encoding mode if the second encoding condition is satisfied; If none of the above holds, at least the first byte Characterized in that it comprises encoding means for encoding the third encoding mode, a.
[0024]
The lossless data encoding apparatus according to claim 19, determines whether or not the appearance rate of the first type of byte data whose hexadecimal notation is 7F or less in input data satisfies a predetermined condition. And encoding means for performing encoding when the appearance rate of the first type of byte data satisfies a predetermined condition, wherein two consecutive bytes in the input data have a compression rate of First means for determining whether or not an encoding condition for encoding in an encoding mode exceeding 1 is satisfied; and, when the encoding condition is not satisfied, at least the first byte of the two bytes is Encoding means including: a second means for encoding in an encoding mode with a compression ratio of 1 or less.
[0025]
22. The lossless encoding method of data according to claim 21, wherein in the input data, it is determined whether or not a rate at which byte data at both ends of three consecutive bytes match each other satisfies a predetermined condition. Satisfies a predetermined condition, it is determined whether (A) two consecutive bytes in the input data satisfy an encoding condition for encoding in an encoding mode in which the compression ratio exceeds 1. And (B) encoding at least the first byte of the two bytes in an encoding mode having a compression ratio of 1 or less when the encoding condition is not satisfied. And
[0026]
In the font data, since the ratio at which the byte data at both ends of the three bytes match is high, if this matching ratio satisfies a predetermined condition, if the encoding including the steps (A) and (B) is executed, It is possible to perform encoding suitable for font data.
[0027]
23. The method according to claim 22, wherein the step (B) is performed by combining a prefix indicating that encoding is performed in the encoding mode with a compression ratio of 1 or less and the lower 7 bits of the first byte. Obtaining encoded data, wherein the input data has a first byte data group whose hexadecimal notation is 7F or less and a second byte data group whose hexadecimal notation is 80 or more. In the encoding of each byte data of the byte data group having a relatively high appearance rate, a 1-bit first prefix is used, and the encoding of each byte data of the byte data group having a relatively low appearance rate is performed. The encoding uses a multi-bit second prefix.
[0028]
Among the byte data group of 7F or less and the byte data group of 80 or more, each byte data of the byte data group having a relatively high appearance rate can be encoded with 8 bits, so that the compression rate at the time of encoding is excessively low. Nothing.
[0029]
【Example】
A. Equipment configuration:
FIG. 1 is a block diagram showing a hardware configuration of an information processing apparatus to which an embodiment of the present invention is applied. This information processing apparatus is configured as a personal computer system, and includes the following units mutually connected by a bus around a CPU 101 as shown in the figure.
[0030]
ROM 104: read-only memory for storing monitor programs and the like
RAM 105: read / write memory constituting main memory
PIC112: Interrupt controller that controls priorities of various interrupts
Mouse interface 115: an interface that controls data exchange with the two-button mouse 114
Keyboard interface 118: an interface for controlling key input from the keyboard 117
FDC 121: Flexible disk controller for controlling flexible disk drive (FDD) 120
HDC 125: a hard disk controller that controls a hard disk drive (HDD) 124
CRTC 129: CRT controller for controlling signal output to CRT 128 for displaying necessary data and the like
Printer interface 131: an interface that controls output of data to the printer 130.
[0031]
FIG. 2 is a functional block diagram showing the configuration of a compression / decompression device driver 200 that performs compression and decompression of data by lossless encoding. The compression / decompression device driver 200 is realized by the CPU 101 executing a program stored in the RAM 105.
[0032]
The device driver 200 for compression and decompression includes a cluster buffer 202 for storing data for one cluster to be compressed, a compression unit 204 for compressing data by lossless encoding, a compressed data buffer 206 for storing compressed data, A write control unit 208 that writes the compressed data to the hard disk 123. One cluster has a predetermined data size, for example, 8 Kbytes.
[0033]
The compression unit 204 includes an encoding type determination unit 210 that determines an encoding type described later, an LZ encoding unit 212 that performs LZ (Lempel-Ziv) encoding, and a font encoding that performs font type encoding. And a non-compression encoding unit 216 for performing non-compression type encoding. First, three types of encoding performed by the LZ encoding unit 212, the font encoding unit 214, and the non-compression encoding unit 216 will be described. In this embodiment, "encoding type" and "encoding mode" have the same meaning.
[0034]
FIG. 3 is an explanatory diagram showing an LZ type encoding method. The LZ type coding used in this embodiment is a Lempel-Ziv coding, and particularly uses a coding method called a slide dictionary method. In the example shown in FIG. 3A, the data to be compressed includes a byte string “71h 3Ah 3Bh...”. In addition, in this specification, the additional symbol "h" indicates that it is a hexadecimal notation.
[0035]
In the LZ type encoding, it is checked whether or not the same byte sequence as a continuous byte sequence of 2 bytes or more has appeared before. If there is the same byte sequence before, the longest matching byte sequence is checked. Then, the current byte sequence is encoded by the distance (offset) from the current byte sequence to the previous byte sequence and the number of bytes of the longest matching byte sequence (match length). For example, in the example of FIG. 3A, the two-byte byte sequence “3Ah 3Bh” underlined with a solid line is the same as the two-byte byte sequence underlined with a broken line. Therefore, the compressed data of the 2-byte byte string “3Ah 3Bh” underlined with a solid line is a prefix indicating that it is LZ type encoded data, and a code word OFFSET (which indicates the distance between the matched byte strings. 3) and a codeword LENGTH (2) representing the match length. The numerical value in parentheses of OFFSET (3) is the number of bytes indicating the distance, and the numerical value in parentheses of LENGTH (2) is the number of bytes indicating the matching length. The format of the compressed data of each encoding type will be described later.
[0036]
In the example of FIG. 3B, since the offset is 3 bytes and the matching length is 3 bytes, the solid, underlined 3 bytes are formed by a combination of the prefix, the code word OFFSET (3), and the code word LENGTH (3). Is encoded.
[0037]
In the example of FIG. 3C, the 5 bytes after the second byte are the same. In this case, the four bytes after the third byte, which are underlined with a solid line, match the four bytes after the second byte. Therefore, the offset is 1 byte and the match length is 4 bytes.
[0038]
As described above, in the LZ type encoding method, when a matching byte sequence has previously existed, data is compressed by encoding the longest matching byte sequence. It is suitable for compressing data that appears repeatedly.
[0039]
FIG. 4 is an explanatory diagram showing a font type encoding method. The font data is characterized in that both ends of three consecutive bytes are equal to each other, and the central byte is often less than 80h (that is, 7F or less). This feature is particularly remarkable in TrueType fonts (TrueType is a trademark of Apple Computer).
[0040]
In FIG. 4, the leading one byte “71h” and the underlined two bytes “3Ah 71h” have the above-described characteristics, so the two underlined bytes are compressed by font type encoding. Is done. One byte “71h” immediately before the two bytes “3Ah 71h” to be encoded by the font type may be encoded by non-compression type encoding, or may be encoded by LZ type encoding. It may be.
[0041]
The font-type compressed data includes a prefix indicating the font type and the lower 7 bits of the first byte "3Ah" in the two bytes to be encoded. The reason why only the lower 7 bits of “3Ah” should be used is that the font type must be encoded in the hexadecimal notation of less than 80h when the font type is encoded. This is because the most significant bit (MSB) is always 0. When a prefix indicating the font type is added, it can be understood that the immediately preceding byte and the last byte of the two bytes encoded by the font type are the same. Therefore, information indicating the trailing byte is unnecessary.
[0042]
In the LZ type encoding in this embodiment, encoding is performed when a continuous byte sequence of two or more bytes is the same as a previous byte sequence. Also, in the case of font-type encoding, encoding is performed when two consecutive bytes satisfy the above-mentioned predetermined characteristics. Therefore, when determining which encoding type is used for encoding, it is only necessary to examine two consecutive bytes and select one of the three encoding types.
[0043]
FIG. 5 is an explanatory diagram showing the format of compressed data of each encoding type used in this embodiment. Type 1 (FIG. 5 (A)) and type 2 (FIG. 5 (B)) exist in the compressed data by the non-compression type encoding. Uncompressed type 1 is compressed data of byte data (00h to 7Fh) whose most significant bit (MSB) before compression is 0. The uncompressed type 2 is compressed data of byte data (80h to FFh) whose most significant bit (MSB) before compression is 1. As shown in FIGS. 5A and 5B, the prefix of the non-compression type compressed data is changed according to the value of the most significant bit of the byte data before compression, so that the data added after the prefix is , Only the lower 7 bits of the data before compression need be used. Note that, in the compressed data of the non-compression type 1, one byte before compression is encoded by 8 bits including the prefix. Therefore, the compression ratio of the non-compression type 1 is 1.0. Since the prefix of uncompressed type 2 is 4 bits, its compression ratio is about 0.7.
[0044]
In a normal encoding method, encoded data obtained by encoding 1-byte data without compression is composed of a prefix of 1 bit or more indicating non-compression and the original 8 bits. Therefore, the encoded data has 9 or more bits, and the compression ratio is less than 1. On the other hand, in the above-mentioned non-compression type 1 encoding, since the compression ratio is 1.0, the compression ratio is higher than that of ordinary uncompressed encoded data. Therefore, if the encoding method of this embodiment is applied to data including bytes less than 80h to some extent or more, it is possible to increase the compression ratio as compared with a normal encoding method.
[0045]
Type 1 (FIG. 5 (C)) and type 2 (FIG. 5 (D)) also exist in compressed data obtained by LZ type encoding. LZ type 1 is compressed data used when the offset value is 255 (FFh in hexadecimal notation), and LZ type 2 is used when the offset value is 256 (100 h 2 in hexadecimal). It is compressed data. Note that the LZ type 1 and the LZ type 2 not only have different prefixes, but also have different codewords OFFSET representing offsets. For example, an LZ type 2 offset codeword is represented by 8 bits, while an LZ type 1 offset codeword is represented by a predetermined number of 9 bits or more. Note that the codeword LENGTH having a matching length of LZ types 1 and 2 is encoded according to an encoding table such as Huffman encoding or Weyl encoding.
[0046]
As shown in FIG. 5E, the font-type compressed data includes a prefix “1110” indicating the font type and the lower 7 bits of the first byte.
[0047]
Each encoding type shown in FIG. 5 has a different prefix, and the prefix is a codeword that can be uniquely and instantaneously decoded. Therefore, when decoding the compressed data, each coding type can be instantaneously determined only by examining the prefix, and decoding can be performed according to each coding type.
[0048]
FIG. 6 is a flowchart showing the procedure of the encoding process. In step S1, two consecutive bytes are extracted from the data to be compressed stored in the cluster buffer 202 as a coding type determination target. The reason why two bytes are to be determined as the encoding type is that, as described above, in the LZ type and font type encoding, a continuous byte string of two or more bytes is encoded.
[0049]
In step S2, it is determined whether or not the LZ type encoding condition is satisfied. The LZ type encoding condition C1 is as follows.
Encoding condition C1: A byte string identical to the two bytes to be determined previously exists.
[0050]
If the above condition C1 is satisfied, LZ type encoding is performed in step S4. In the font type encoding, a continuous byte string of 2 bytes or more is encoded as described with reference to FIG. If the encoding condition C1 is not satisfied, it is determined in step S3 whether the encoding condition of the font type is satisfied. The encoding condition C2 of the font type is as follows.
Coding condition C2: The last byte of the two bytes to be determined is equal to the one byte immediately before the two bytes, and the first byte of the two bytes is less than 80h.
[0051]
If the above-described encoding condition C2 is satisfied, font-type encoding is performed in step S5. In the font type encoding, two bytes are encoded (FIG. 4). If the encoding condition C2 is not satisfied, in step S6, the first byte of the two bytes to be determined is encoded in the non-compression type. In non-compression type coding, only one byte is coded.
[0052]
In step S7, it is determined whether or not all of the data to be compressed stored in the cluster buffer 202 has been encoded. If not, the process returns to step S1. Steps S1 to S3 and S7 are executed by the encoding type determination unit 210 (FIG. 2), and steps S4, S5 and S6 are performed by the LZ encoding unit 212, the font encoding unit 214, and the non-compression encoding unit 216. Respectively.
[0053]
The procedure of FIG. 6 can be reworded as follows. That is, since the LZ type encoding has the highest compression rate among the three encoding types, the LZ type encoding is performed when the LZ type encoding condition C1 is satisfied. In the font type encoding, since the compression ratio is medium among the three encoding types, the LZ type encoding condition C1 is not satisfied. However, when the font type encoding condition C2 is satisfied, the font is not encoded. Type encoding is performed. Since the non-compression type encoding has the lowest compression rate, it is encoded by the non-compression type only when encoding cannot be performed by the LZ type and the font type. Note that the compression ratio of the LZ type and the font type is greater than 1, whereas the compression ratio of the non-compression type is less than 1.
[0054]
7 and 8 are explanatory diagrams showing an example of encoding input data by applying the embodiment. FIG. 7A shows a byte sequence of the data to be compressed. Since the first byte of the byte string cannot be compressed by either the LZ type or the font type, it is encoded by the non-compression type as shown in FIG. Since the first byte "85h" is 80h or more, it is compressed by the non-compression type 2.
[0055]
Since the second and third bytes “3Ah 85h” satisfy the above-described font type encoding condition C2, they are encoded in the font type as shown in FIG. 7C.
[0056]
The fourth and fifth bytes “3Bh E0h” do not satisfy the LZ type encoding condition C1 and do not satisfy the font type encoding condition C2. Therefore, as shown in FIG. 7 (D), only the fourth byte “3Bh” is encoded by the non-compression type 1.
[0057]
Since the fifth and sixth bytes “E0h 3Bh” do not satisfy either of the two encoding conditions C1 and C2, only the fifth byte “E0h” is encoded by the uncompressed type 2 as shown in FIG. Be transformed into Also, since the sixth and seventh bytes “3Bh 3Ah” do not satisfy both of the two encoding conditions C1 and C2, only the sixth byte “3Bh” is the uncompressed type 1 as shown in FIG. .
[0058]
Since the seventh and eighth bytes “3Ah 85h” are the same as the second and third bytes, they are encoded in the LZ type. In LZ type encoding, the longest matching byte sequence is examined. In the example of FIG. 8C, since the longest matching byte string is 3 bytes, the 3rd to 7th to 9th bytes are encoded in the LZ type. Note that the offset in the case of FIG. 8C is 5 bytes, which is 255 bytes or less, so that LZ type 1 encoding is performed.
[0059]
Since the tenth byte “FFh” does not satisfy any of the two encoding conditions C1 and C2, it is encoded by the non-compression type 2 as shown in FIG. When the encoding of the data to be compressed is completed, a predetermined end code (not shown) is added to the end of the compressed data.
[0060]
In the above embodiment, the encoding type is determined according to whether two consecutive bytes of the compressed data satisfy the determination condition C1 or C2. Therefore, the encoding type is determined according to the local data structure of the compressed data. An appropriate coding type can be set, and as a result, there is an advantage that the compression ratio can be improved.
[0061]
In the above embodiment, the case where the data to be compressed is encoded by the lossless encoding method for font data has been described. However, the given data to be compressed is compressed by the lossless encoding method for font data, It is also possible to determine in advance whether to perform compression by the lossless encoding method.
[0062]
For example, the extension of the file name of font data is often limited to several types such as "TTF". Therefore, when data compression is performed, the extension of the compressed data file is checked, and if the extension matches a previously registered extension name such as “TTF”, the above-described lossless encoding method for font data is used. You may make it apply.
[0063]
Note that the above-described uncompressed type 1 encoding requires only 8 bits of encoded data, and thus has an advantage that the compression rate of the uncompressed type encoded data is higher than that obtained by a normal encoding method. Therefore, when the encoding method of the font data according to the above-described embodiment is applied to the input data having the appearance rate of byte data of less than 80h less than a certain level, the compression of the encoded data is smaller than that of the normal encoding method. It is possible to increase the rate. There are various methods for determining whether to apply the font data encoding method according to such a concept. For example, a fixed number of bytes (that is, at least a part of the input data) is checked from the head of the data to be compressed, and the encoding method of the font data is applied when the number of bytes less than 80h is equal to or more than a predetermined value. Good. Further, a fixed number of bytes may be checked while thinning out the data to be compressed, and the font data encoding method may be applied when the number of bytes less than 80h is equal to or more than a predetermined value.
[0064]
When the byte data of 80h or more is relatively large, the prefix for the data of 80h or more is 1 bit (for example, "0"), and the data of less than 80h is 2 bits or more prefix (for example, "0"). 1111 "). In other words, by examining at least a part of the input data, examining the number (or appearance rate) of the byte data in hexadecimal notation less than 80h and the byte data of 80h or more (or appearance rate), A prefix may be used, and a prefix of two or more bits having a relatively low appearance rate may be used.
[0065]
Note that the font data has a feature that the bytes at both ends of three consecutive bytes are the same, and this feature is used to determine the sign of the font data if the ratio of coincidence of every other two bytes is equal to or greater than a certain value. It is also possible to apply the conversion method. In this case, the rate at which every other two bytes match at least a part of the input data may be checked, and an encoding method to be applied may be selected according to the result.
[0066]
The present invention is not limited to the above-described embodiment, and can be implemented in various modes without departing from the gist of the present invention. For example, the following modifications are possible.
[0067]
(1) FIG. 9 is obtained by adding steps S10 and S11 to the processing procedure of FIG. In step S10, if the offset at the time of encoding with the LZ type is 256 or more and the matching length is equal to 2, the process proceeds to step S11, where 2 bytes to be encoded are determined as the font type. It is determined whether condition C2 is satisfied. If the font type determination condition C2 is satisfied, the two bytes to be encoded are encoded with the font type, otherwise, the two bytes are encoded with the LZ type. The reason for this is that when both the conditions of steps S10 and S11 are satisfied, the compression rate of the encoded data is higher when encoding is performed using the font type than when encoding is performed using the LZ type. .
[0068]
As described above, even when a 2-byte byte string that satisfies the LZ type encoding determination condition C1 is encoded with the font type and the compression rate of the encoded data becomes higher, the encoding with the font type is performed. You may make it.
[0069]
(2) Various lossless encoding methods such as Huffman encoding and arithmetic encoding can be used in place of the LZ encoding method in the above embodiment.
[0070]
【The invention's effect】
As described above, according to the first and eleventh aspects of the present invention, when input data satisfies a predetermined encoding condition, encoding is performed in a predetermined encoding mode suitable for font data. There is an effect that font data can be efficiently compressed.
[0071]
According to the second, fifth, twelfth, and fifteenth aspects, the compression ratio can be made 1 or more.
[0072]
According to the third, sixth, thirteenth, and sixteenth aspects, there is an effect that two bytes can be encoded with 15 bits or less.
[0073]
According to the fourth and fourteenth aspects of the present invention, it is possible to select an appropriate encoding mode in accordance with the content of two-byte data. In particular, the two bytes are the second code unique to font data. There is an effect that the compression can be efficiently performed when the conversion condition is satisfied.
[0074]
According to the seventh and 17th aspects of the present invention, in the first encoding mode, the longest matching byte string is encoded, so that the compression rate can be increased.
[0075]
According to the eighth and eighteenth aspects of the present invention, in the third encoding mode, when the hexadecimal notation of the first byte in the two bytes is 7F or less, encoding is performed by 8 bits, so that the compression rate is excessively high. The effect is that it does not decrease.
[0076]
According to the ninth and nineteenth aspects of the present invention, the encoding including the steps (A) and (B) is executed when the appearance rate of byte data whose hexadecimal notation is 7F or less satisfies a predetermined condition. , 7F or less, it is possible to set a code word so that the compression ratio does not become excessively low.
[0077]
According to the tenth and twentieth aspects of the present invention, since byte data of 7F or less can be encoded with 8 bits, the compression rate at the time of encoding does not become excessively low.
[0078]
According to the twenty-first aspect of the present invention, the encoding including the steps (A) and (B) is executed when the ratio of coincidence of the byte data at both ends of the three bytes satisfies a predetermined condition. This makes it possible to perform encoding suitable for.
[0079]
According to the invention described in claim 22, each byte data of a byte data group having a relatively high appearance rate among the byte data group of 7F or less and the byte data group of 80 or more can be encoded with 8 bits. This has the effect that the compression ratio at the time of encoding does not become excessively low.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a hardware configuration of an information processing apparatus to which an embodiment of the present invention is applied.
FIG. 2 is a functional block diagram showing the configuration of a compression / decompression device driver 200 that performs compression and decompression of data by lossless encoding.
FIG. 3 is an explanatory diagram showing an LZ type encoding method.
FIG. 4 is an explanatory diagram showing a font type encoding method.
FIG. 5 is an explanatory diagram showing a format of compressed data of each encoding type.
FIG. 6 is a flowchart showing a procedure of an encoding process.
FIG. 7 is an explanatory diagram showing an example of encoding input data by applying the embodiment.
FIG. 8 is an explanatory diagram showing an example of encoding input data by applying the embodiment.
FIG. 9 is a flowchart illustrating a modification of the procedure of the encoding process.
[Explanation of symbols]
101 ... CPU
104 ... ROM
105 ... RAM
112 ... PIC
115 ... Mouse interface
117 ... Keyboard
118 ... Keyboard interface
121 ... FDC
123 ... Hard disk
124 ... Hard disk drive
125 ... HDC
128 ... CRT
129… CRTC
130 ... Printer
131 ... Printer interface
200: Device driver for compression and decompression
202: Cluster buffer
204: compression unit
206: compressed data buffer
208: Write control unit
210: coding type determination unit
212 ... LZ encoding unit
214: Font encoding unit
216: Non-compression encoding unit

Claims (22)

A method of reversibly encoding data,
(A) Checking two consecutive bytes of input data, if the last byte of the two bytes is equal to the one byte immediately before the two bytes and the first byte of the two bytes is equal to or less than a predetermined value Determining whether or not the encoding condition of
(B) encoding the two bytes in a predetermined encoding mode when the encoding condition is satisfied;
A lossless encoding method for data. The lossless encoding method of data according to claim 1, wherein
In the step (B),
(C) encoding the two bytes with a code word of less than 16 bits. 3. The data lossless encoding method according to claim 2, wherein
The step (C) includes:
A method of creating encoded data by combining a prefix of 8 bits or less indicating that encoding has been performed in the predetermined encoding mode and the lower 7 bits of the first byte. . A method of reversibly encoding data,
(A) determining whether two consecutive bytes of input data satisfy a first encoding condition for encoding in a first encoding mode;
(B) when the first encoding condition is not satisfied, the last byte of the two bytes is equal to the one byte immediately before the two bytes, and the first byte of the two bytes is a predetermined value. Determining whether or not a second encoding condition that:
(C) at least the two bytes are encoded in the first encoding mode having the highest compression ratio when the first determination condition is satisfied, and when the second encoding condition is satisfied, Two bytes are encoded in a second encoding mode having a medium compression ratio, and if neither the first nor the second encoding condition is satisfied, at least the first byte is converted to a third encoding having the lowest compression ratio. Encoding in an encoding mode;
A lossless encoding method for data. The data lossless encoding method according to claim 4, wherein
The encoding step according to the second encoding mode includes:
Encoding the two bytes with a codeword of less than 16 bits. The data lossless encoding method according to claim 5,
In the encoding step in the second encoding mode, the encoding is performed by combining a prefix of 8 bits or less indicating that encoding is performed in the second encoding mode, and the lower 7 bits of the first byte. Creating a data, a method of reversible encoding of data. A lossless encoding method for data according to any one of claims 4 to 6, wherein:
The encoding step according to the first encoding mode includes:
Searching the input data for a longest second byte sequence that matches a first byte sequence containing at least the two bytes;
A first codeword indicating that encoding has been performed in the first encoding mode, a second codeword indicating a distance between the first and second byte strings, and Creating encoded data by combining with a third codeword indicating the length. A lossless encoding method for data according to any one of claims 4 to 7,
The third encoding mode includes a step of obtaining encoded data by combining a prefix indicating that encoding is performed in the third encoding mode and the lower 7 bits of the first byte,
The prefix in the third encoding mode is a 1-bit first prefix used when the hexadecimal notation of the first byte is 7F or less, and the hexadecimal notation of the first byte is 80 or more. A lossless encoding method for data selected from any one of a plurality of bits and a second prefix used when. A lossless encoding method for data,
In the input data, it is determined whether the appearance rate of the first type of byte data whose hexadecimal notation is 7F or less satisfies a predetermined condition,
When the appearance rate of the byte data of the first type satisfies a predetermined condition, (A) two consecutive bytes in the input data are encoded in an encoding mode in which the compression rate exceeds 1. Determining whether or not the encoding condition of
(B) encoding the at least the first byte of the two bytes in an encoding mode having a compression ratio of 1 or less when the encoding condition is not satisfied. Encoding method. The data lossless encoding method according to claim 9, wherein
The step (B) includes a step of obtaining encoded data by combining a prefix indicating that encoding has been performed in the encoding mode with a compression ratio of 1 or less and the lower 7 bits of the first byte. Including
The prefix is a 1-bit first prefix used when the hexadecimal notation of the first byte is 7F or less, and a plurality of prefixes used when the hexadecimal notation of the first byte is 80 or more. A lossless encoding method for data selected from one of a bit and a second prefix. A lossless encoding device for data,
Checking two consecutive bytes of the input data, encoding that the last byte of the two bytes and the one byte immediately before the two bytes are equal to each other and the first byte of the two bytes is equal to or less than a predetermined value Determining means for determining whether a condition is satisfied;
Encoding means for encoding the two bytes in a predetermined encoding mode when the encoding condition is satisfied;
A lossless encoding device for data, comprising: The data lossless encoding device according to claim 11, wherein
The encoding means,
A first means for encoding the two bytes with a code word of less than 16 bits, the data lossless encoding apparatus. The data lossless encoding device according to claim 12,
The first means includes:
A lossless encoding apparatus for encoding the data, comprising: means for creating encoded data by combining a prefix of 8 bits or less indicating that encoding is performed in the predetermined encoding mode and the lower 7 bits of the first byte. . A lossless encoding device for data,
First determining means for determining whether two consecutive bytes of input data satisfy a first encoding condition for encoding in a predetermined first encoding mode;
If the first encoding condition is not satisfied, the last byte of the two bytes is equal to the one byte immediately before the two bytes, and the first byte of the two bytes is equal to or less than a predetermined value. Second means for determining whether or not a second encoding condition of
When the first determination condition is satisfied, at least the two bytes are encoded in the first encoding mode having the highest compression ratio. When the second encoding condition is satisfied, the two bytes are encoded. Encoding is performed in a second encoding mode having a medium compression rate, and if neither the first nor second encoding condition is satisfied, at least the first byte is encoded in a third encoding mode having the lowest compression rate. Encoding means for encoding with
A lossless encoding device for data, comprising: The data lossless encoding device according to claim 14,
The encoding means,
A reversible coding apparatus for data, comprising: first means for coding the two bytes with a code word of less than 16 bits in the coding in the second coding mode. The data lossless encoding device according to claim 15, wherein
The first means includes:
Lossless encoding of data including means for creating encoded data by combining a prefix of 8 bits or less indicating that encoding is performed in the second encoding mode and the lower 7 bits of the first byte. apparatus. A data lossless encoding device according to any one of claims 14 to 16,
The encoding means,
Searching means for searching in the input data for a longest second byte string that matches the first byte string including at least the two bytes in the encoding in the first encoding mode;
A first codeword indicating that encoding has been performed in the first encoding mode, a second codeword indicating a distance between the first and second byte strings, and A coded data generating means for generating coded data by combining the third codeword indicating the length with the third codeword. The data lossless encoding device according to any one of claims 14 to 17, wherein
The encoding means,
Means for creating encoded data by combining a prefix indicating that encoding has been performed in the third encoding mode and the lower 7 bits of the first byte in encoding in the third encoding mode. When,
As the prefix in the third encoding mode, the 1-bit first prefix used when the hexadecimal notation of the first byte is 7F or less, and the hexadecimal notation of the first byte is 80 or more. Means for selecting from either one of a plurality of bits and a second prefix used in the case of. A lossless encoding device for data,
Determining means for determining whether or not the appearance rate of the first type of byte data whose hexadecimal notation is 7F or less in input data satisfies a predetermined condition;
Encoding means for encoding when the appearance rate of the first type of byte data satisfies a predetermined condition,
First means for determining whether or not two consecutive bytes in the input data satisfy an encoding condition for encoding in an encoding mode in which a compression ratio exceeds 1;
An encoding unit including, if the encoding condition is not satisfied, a second unit that encodes at least a first byte of the two bytes in an encoding mode with a compression ratio of 1 or less;
A lossless encoding device for data, comprising: 20. The data lossless encoding device according to claim 19,
The second means includes:
Means for generating encoded data by combining a prefix indicating that encoding has been performed in the encoding mode with a compression ratio of 1 or less and the lower 7 bits of the first byte;
As the prefix, a 1-bit first prefix used when the hexadecimal notation of the first byte is 7F or less, and a plurality of prefixes used when the hexadecimal notation of the first byte is 80 or more. Means for selecting from one of a bit and a second prefix. A lossless encoding method for data,
In the input data, it is checked whether or not the rate at which the byte data at both ends of three consecutive bytes match each other satisfies a predetermined condition.
When the matching ratio satisfies a predetermined condition,
(A) determining whether two consecutive bytes in the input data satisfy an encoding condition for encoding in an encoding mode in which a compression ratio exceeds 1;
(B) encoding the at least the first byte of the two bytes in an encoding mode having a compression ratio of 1 or less when the encoding condition is not satisfied. Encoding method. The data lossless encoding method according to claim 9 or 21,
In the step (B),
Obtaining encoded data by combining a prefix indicating that encoding is performed in the encoding mode with a compression ratio of 1 or less and lower 7 bits of the first byte,
The appearance rate of the first byte data group whose hexadecimal notation is 7F or less in the input data is compared with the appearance rate of the second byte data group whose hexadecimal notation is 80 or more, and the appearance rate is relatively high. In encoding each byte data of the byte data group, a 1-bit first prefix is used, and in encoding each byte data of the byte data group having a relatively low appearance rate, a second prefix of a plurality of bits is used. A lossless encoding method for data that uses words.

Images