JP2000261812A - Image coder and decoder - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent a throughput from being lowered even when a coding method encoding a picture through the use of code information of a just preceding block or other coding method is selected. SOLUTION: A block processing means 1 segments a block picture 8 being a coding object from input picture data 7 and outputs the result. A coding changeover means 2 selects a reversible coding means 3 or an irreversible coding means 5 adaptively and outputs changeover identification information 10 to a code data multiplexer means 6. A coding dependence information calculation means 4 calculates coding dependence information 12 required for coding by the reversible coding means 3 or the irreversible coding means 5 and outputs the information 12 to the code data multiplexer means 6. The code data multiplexer means 6 multiplexes the coding system identification information 10, the coding dependence information 12, reversible code data 11 or irreversible code data 13 to configure output code data 14.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image encoding apparatus and a decoding apparatus for switching an encoding method according to the characteristics of image data, and more particularly to an encoding method suitable for PDL (page description language) image data and a decoding method. This is to switch between an image encoding method suitable for image data and to suppress a decrease in encoding efficiency due to the switching.

[0002]

2. Description of the Related Art (1) Reversible coding method for PDL images JP-A-9-224253 discloses a method in which a plurality of predictive coding units (for example, for predicting a drawing direction) are provided, and a predictive code whose prediction matches a target pixel is provided. A technique for outputting an identifier of a conversion unit as code data is disclosed. This image coding technology (drawing direction prediction coding method) efficiently losslessly codes a PDL image described in an opaque model by utilizing the fact that the probability of the same pixel value continuing is extremely high. . However, since the same pixel value is hardly continuous in the natural image, the natural image cannot be efficiently encoded.

(2) Lossy coding method for natural images JPEG (Joint photographic C)
JPEG Baseline (Baseline) standardized in the Oding Experts Group
The encoding method (hereinafter, simply referred to as JPEG method) efficiently encodes a natural image. The JPEG method uses a discrete cosine transform (hereinafter, referred to as digital cosine transform) of digitized input image data.
DCT transformation), quantizes the transformed coefficient (hereinafter, DCT coefficient) to create a quantization index, and encodes the image data by entropy-encoding the quantization index. In general, the power of a low frequency component is large and the power of a high frequency component is small in a natural image. On the other hand, PDL
Since an image has higher power of high-frequency components than a natural image, encoding a PDL image by the JPEG method deteriorates edges. In order to reduce the deterioration, the quantization level of the DCT coefficient may be reduced to reduce the removal of the redundant component, but the coding efficiency is reduced.

(3) Encoding method for PDL / natural mixed image When encoding an image in which a PDL image and a natural image are mixed, the above two methods may be combined. For example, as in JP-A-7-212601, there is a conventional example in which there are two or more encoding systems that can be processed in matrix units, and an encoding system suitable for an image is applied for each encoding unit.

(4) Coding switching method A conventional example of switching the coding method for each coding unit will be described with reference to FIG. In FIG. 18, an input signal is subjected to a scaling process in a scaling unit 1701, and then, in an image area separation unit 1702, a portion having a large change in luminance (character portion) such as character information and a portion having a large change in luminance such as a photograph It is separated into small parts (photographs). The separated result is output as a determination signal. The data of the character part is sent to the binarization unit 1704 via the interpolation unit 1703, and the binarization unit 1704
, And losslessly encoded by the compression unit (A) 1705. Further, the data of the photograph part is stored in the halftone processing part 170.
After being processed by pseudo halftone processing such as an error diffusion method in 6, the compression unit (B) 1707 performs compression processing such as JPEG using a matrix. When the attention matrix is determined to be a photograph portion, the matrix adjacent to the right is set as the next attention matrix. In the JPEG system, a block of 8 × 8 pixels is used as a matrix. The 8 × 8 pixel data of each block is converted into 8 × 8 D data by two-dimensional DCT.
The DCT coefficients are converted into CT coefficients, and the 64 DCT coefficients are quantized with a different step size for each coefficient by using a quantization table that defines a quantization step size for each coefficient.
DC component of quantized DCT coefficient (hereinafter, DC coefficient)
Is obtained by calculating the difference between the DC coefficient and the immediately preceding block, and the DC differences are grouped. By this grouping, the DC difference is divided into a group number and an additional bit indicating a DC difference value.
The difference is encoded by arithmetic coding. AC components of DCT coefficients (hereinafter, AC coefficients) are rearranged one-dimensionally by zigzag scanning. Coefficients of continuous 0 are run-length encoded, and AC coefficients other than 0 are grouped and Huffman encoded.

[0006] The decision signal is sent to a compression section (C) 1708 and compressed.

The compression section (A) 1705 and the compression section (B) 17
07, the data compressed by the compression unit (C) 1708 is stored in the storage unit 1709.

[0008] The data stored in the storage unit 1709 is read and decoded for image print processing and the like. That is, the compression unit (A) 1705, the compression unit (B) 1707,
Expansion unit (A) corresponding to compression unit (C) 1708
1710, extension part (B) 1711, extension part (C) 171
2, the data is decoded, and then the image is synthesized by the synthesizing unit 1713 to obtain the original image.

[0009]

However, when the coding data is configured by switching between a coding method such as the JPEG method depending on the coding information of the immediately preceding block and another coding method, the coding target block is If the immediately preceding block is encoded by a different encoding method, an appropriate value cannot be referred to as the encoding-dependent information, which causes a problem of lowering the encoding efficiency.

Referring to FIGS. 15 to 17, a description will be given of a decrease in coding efficiency when switching between the JPEG method and the lossless coding method. FIG. 15 shows three consecutive blocks as JP
It is an example of a DC coefficient in the case of encoding by the EG method. FIG.
6 and FIG. 17 are diagrams showing coding in which switching is performed between the JPEG method and the lossless coding method, and the DC
This is an example in which the coefficient is a fixed value. In FIGS. 15 to 17, # indicates a block number. For example, the #N block is the N-th block to be encoded. 16 and 17
Is that the # N-1 block is encoded by the JPEG method,
The N blocks are switched to lossless encoding and encoded, and the # N + 1 blocks are switched to JPEG and encoded. FIG. 16 shows an example in which the fixed value of the DC coefficient is 0, FIG.
7 is an example in which the fixed value of the DC coefficient is 255.

[0011] As shown in FIG.
When coding by the EG method, the DC difference between the respective blocks is 60−54 = 6 for the # N−1 block and the #N block, and is 102−60 = 42 for the #N block and the # N + 1 block. . Therefore, according to the DC difference grouping table of FIG. 11 in the coding of the DC coefficient, the DC differences between the # N-1 block and the #N block are grouped by the group number 3 and three additional bits are added. . The DC difference between the #N block and the # N + 1 block is grouped by a group number 6, and the additional bits are 6 bits. Accordingly, 3 + 6 = 9 bits are added in the coding of the DC coefficient in these three blocks.

On the other hand, when a fixed value 0 is used as a DC value at the time of coding switching as shown in FIG. 16, the DC difference between the blocks is 0 for the # N-1 block and the #N block. −54 = −54, #N block and # N +
In one block, 102-0 = 102. According to the DC difference grouping table of FIG.
The DC differences of the N blocks are grouped by a group number 6, and 6 additional bits are added. The DC difference between the #N block and the # N + 1 block is the group number 7
And the additional bits are 7 bits. Therefore, in the coding of the DC coefficient in these three blocks, 6 + 7 = 1
Three bits are added. Therefore, in the example of FIG. 16 in which the coding is switched, the additional bits are increased by 13−9 = 4 bits in contrast to the case of not switching the coding in FIG.

Also, as shown in FIG.
When the fixed value is set to 255 as the C value, 255-54 = 201 and #N for the # N-1 block and the #N block.
For the block and the # N + 1 block, 102-255 =-
153. Accordingly, the DC difference between the # N-1 block and the #N block is grouped by the group number 8, and the additional bits are 8 bits.
The DC difference of the block is represented by group number 8 and additional bits 8
Becomes Accordingly, 8 + 8 = 16 bits are added in the coding of the DC coefficient in these three blocks. In the example in which the coding is switched, the number of additional bits is increased by 16-9 = 7 bits as compared with the case where the coding in FIG. 15 is not switched.

As described above, DC between blocks
When coding data is configured by switching between a coding method in which coding information depends on the immediately preceding block and another coding method, such as a JPEG method for coding a difference between values, when coding is switched, Since coding information cannot be referred to, coding efficiency is reduced.

A problem to be solved by the present invention is to prevent the coding efficiency from being reduced when the coding system is switched.

[0016]

According to the present invention, in order to solve the above-mentioned problems, encoding is performed in units of blocks,
And an image encoding device that converts input image data into output encoded data by a first encoding method in which codes of adjacent blocks depend on each other and a second encoding method different from the first encoding method. Blocking means for dividing the input image data into blocks; first encoding means for encoding the input image data in block units using the first encoding method; and Second encoding means for encoding in block units by the encoding method,
Encoding switching means for selectively switching between the encoding by the encoding means and the encoding by the second encoding means, and the first encoding means when the first encoding is not performed. Means for generating encoding-dependent information used for encoding, and multiplexing means for multiplexing the code output by the second encoding means and the encoding-dependent information to form an output code, Further, the first encoding means performs the encoding using the encoding-dependent information when the encoding is restarted.

In this configuration, even when the first encoding is not performed, that is, when the second encoding is performed, the dependency information of the first encoding is generated. When the first encoding is restarted, encoding can be performed efficiently using this dependency information. Further, since the dependency information is sent to the decoding side together with the second code, the decoding can be reliably performed even when the first decoding is restarted.

Further, in this configuration, one of the first encoding method and the second encoding method may be a lossy encoding method, and the other may be a lossless encoding method.

Further, the above-mentioned first encoding method may be such that encoding is performed by utilizing the dependence of the DC component of image data. Further, the first encoding method is JPEG
It can be a baseline coding technique. A natural image may be encoded in the first encoding, and a PDL image may be encoded in the second encoding.

If the second encoding method also requires inter-block dependence information, the corresponding inter-block dependence information is similarly multiplexed at the time of the first encoding.

According to the present invention, in order to solve the above-mentioned problems, a decoding device for decoding an output code is provided with the first device.
First decoding means for decoding the code coded by the coding means, second decoding means for decoding the code coded by the second coding means, and multiplexing by the multiplexing means. Means for separating the encoded code of the second encoding technique and the encoding-dependent information of the first encoding technique, image information obtained by decoding by the first decoding means, and the second decoding Means for synthesizing image information obtained by decoding by the means, and a code of the first coding method using the coding dependence information when decoding by the first decoding means is restarted. Is to be decrypted.

Also in this configuration, decoding can be efficiently performed using the encoding-dependent information sent redundantly.

[0023]

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, the present invention will be described in detail.
First, an example of the principle configuration of the present invention will be described.

FIG. 1 shows an example of the basic configuration of an image encoding apparatus according to the present invention. In this figure, the image encoding apparatus comprises a blocking unit 1, an encoding switching unit 2, a lossless encoding 3, It is configured to include an encoding-dependent information calculating unit 4, an irreversible encoding unit 5, and a code data multiplexing unit 6.

The blocking means 1 receives the input image data 7, cuts out the block image 8 to be coded, and outputs it. The encoding switching means 2 includes a lossless encoding means 3 as means for encoding the block image 8,
Alternatively, any one of the irreversible encoding means 5 is selected and switched, and the encoding target image 9 is output to the lossless encoding means 3, the irreversible encoding means 5, and the encoding-dependent information calculating means 4, and further switched. The identification information 10 is transmitted to the code data multiplexing means 6
Is output to

The lossless encoding means 3 is disclosed in Japanese Unexamined Patent Publication No. 9-2242.
In this technique, a PDL image is efficiently encoded by a drawing direction predictive encoding method proposed in Japanese Patent Publication No. 53-53. The encoding-dependent information calculating means 4 calculates the encoding-dependent information 12 necessary for encoding in the lossless encoding means 3 and the irreversible encoding means 5 and outputs the information to the encoded data multiplexing means 6. It is. The irreversible encoding means 5 uses JPEG
It encodes a natural image such as a method efficiently.

The code data multiplexing means 6 includes coding method identification information 10 for identifying the coding method selected by the coding switching means 2 and coding dependency information 12 calculated by the coding dependency information calculating means 5. And lossless code data 11 of the lossless encoding means 3 or lossy code data 13 of the lossy encoding means 5 to multiplex to form output code data 14.

Next, the operation of the present invention will be described with reference to FIGS. 2 and 3 in addition to FIG. FIG. 1 is a block diagram of the present invention as described above, and FIG. 2 is an operation flowchart of the present invention. FIG. 3 shows an operation example at the time of encoding switching.

[Step S1]: The blocking unit divides the input image 7 into block images 8. Proceed to step S2. [Step S2]: Initialize or update the block to be encoded. Proceed to step S3. [Step S3]: It is determined whether the encoding target image 9 has the characteristics of a PDL image. If it has the characteristics of the PDL image, the process proceeds to step S4; otherwise, the process proceeds to step S7. [Step S4]: The lossless encoding unit 3 losslessly encodes the encoding target image 9 and outputs lossless encoded data 11.
Proceed to step S5. [Step S5]: The encoding dependence information calculation means 4 outputs the inter-block encoding dependence information 12 of the lossy encoding.
Proceed to step S6. [Step S6]: The code data multiplexing unit 6 multiplexes the lossless code data 11, the coding dependency information 12, and the coding method identification information 10, and outputs output code data 14.
Proceed to step S10. [Step S7]: The irreversible encoding unit 5 irreversibly encodes the encoding target image 9 and outputs irreversible encoded data. Proceed to step S8. [Step S8]: The encoding dependency information calculation means 4 calculates the inter-block dependency information 12 of the lossless coding. Proceed to step S9. [Step S9]: The output code data 14 is output by multiplexing the irreversible code data 13, the coding dependency information 12, and the coding method identification information 10. Proceed to step S10. [Step S10]: It is determined whether or not all blocks have been processed. If the processing has been completed, the routine processing ends. If the processing has not been completed, the process returns to step S2.

In the above description, the case where the encoding dependent information is used in both the lossless encoding and the irreversible encoding has been described. However, when one of them does not use the encoding dependent information,
For one of them, there is no need to calculate the encoding dependency information.

As shown in FIG. 3, when the (N-1) th block is encoded by the lossless encoding means 3 and the Nth block is switched to the lossy encoding means 5, Since the output coded data 14 includes the coding-dependent information 12 between the blocks in the irreversible coding means 5, even when switching from the lossless coding means to the irreversible coding means, the output code data is N-1th. The lossy encoding can be performed with reference to the lossy encoding dependent information between the Nth blocks.

Next, the decoding apparatus of the present invention will be described. In FIG. 4, the decoding device includes code switching means 21,
It includes a lossless decoding unit 22, an irreversible decoding unit 23, an encoding-dependent information extracting unit 24, and an image synthesizing unit 25. The switching unit 21 extracts the encoding system identification information from the code, and sends the code to the lossless decoding unit 22 or the irreversible decoding unit 23 based on the encoding system identification information.
The encoding-dependent information extracting means 24 extracts the encoding-dependent information from the code. Lossless decoding means 22 or lossy decoding means 2
3 decodes the code. When the encoding system is switched, the lossless decoding unit 22 or the irreversible decoding unit 23 performs decoding using the encoding-dependent information. The decoding result is sent to the image synthesizing means 22 and synthesized into a final image.

[0033]

Next, specific embodiments of the present invention will be described.

FIG. 5 shows the overall configuration of the image encoding apparatus of this embodiment. In this figure, the image encoding apparatus comprises a blocking circuit 101, a PDL / natural image discriminating circuit 201, an adaptive predictive lossless circuit. It is configured to include an encoding circuit 301, a DC coefficient encoding circuit 401, a DCT conversion circuit 501, a DCT coefficient quantization circuit 502, an AC coefficient encoding circuit 503, and a code data multiplexing circuit 601.

The blocking circuit 101 converts the input image data 701 input one pixel at a time in the raster scan
The image is configured as a block unit image of × 8 pixels and output as a block image 801. The PDL / natural image identification circuit 201 identifies whether the pixel image is a natural image in which pixel values fluctuate locally or a PDL image in which the same pixels are continuous based on the frequency characteristics and the number of connected pixels of the block image 801. . The adaptive predictive lossless encoding circuit 301 performs lossless encoding by a drawing direction predictive encoding method described in Japanese Patent Application Laid-Open No. 9-224253. That is, predicting the neighboring pixel whose pixel value matches the pixel value to be coded, and using the statistical bias of the coded symbol using the rank of the predictor that has been predicted correctly or the prediction error as the coded symbol. It compresses the quantity.

The DC coefficient encoding circuit 401 encodes the inter-block difference information of the DC coefficient of the DCT transform in the JPEG system to form inter-block coded data 1202. The DCT transform circuit 501 transforms the encoding target image 901 into a DCT coefficient 1301.
The CT coefficient quantization circuit 502 quantizes the DCT coefficient 1301 based on a predetermined quantization table.
The AC coefficient encoding circuit 503 uses DC in the JPEG system.
It encodes the AC coefficient of the T conversion.

The code data multiplexing circuit 601 selects the reversible code data 1101 or the reversible code data 1303 based on the switching identification signal 1001 and multiplexes it with the inter-block code data 1202, and outputs the output code data 1401.
1 is output.

Next, the operation of the embodiment will be described with reference to FIGS.

First, the blocking circuit 101 divides the input image data 701 into 8 × 8 pixel blocks by the flow processing shown in FIG.
/ Output to the natural image identification circuit 201. The operation shown in FIG. 6 can be easily understood from the contents described in FIG. 6 and will not be particularly described.

The PDL / natural image discriminating circuit 201 outputs the block image 801 as an encoding target image 901 to the adaptive predictive lossless encoding circuit 301 and the DCT transform circuit 501, and the block image 801 is subjected to the flow processing shown in FIG. It is determined whether the image is a PDL image or a natural image, and the result is output to the code data multiplexing circuit 601 as a switching identification signal 1001. The operation shown in FIG. 7 can be easily understood from the description content of FIG. 7, and therefore will not be particularly described.

The adaptive predictive lossless coding circuit 301 predicts a neighboring pixel whose pixel value matches that of the pixel to be coded, and uses the rank of the predictor that has achieved prediction or the prediction error as a coded symbol. The data amount is compressed using the statistical bias of the symbols to form the lossless code data 1101,
Output to the code data multiplexing circuit 601.

The DCT transform circuit 501 transforms the image 901 to be encoded into an 8 × by the two-dimensional DCT transform shown in FIGS.
The signal is converted into eight DCT coefficients 1301 and output to the DCT coefficient quantization circuit 502.

The DCT coefficient quantization circuit 502 quantizes the DCT coefficient 1301 using a predetermined quantization table that defines a quantization step size for each coefficient. Further, of the quantized coefficients, the DC coefficient 1201 that is a DC component is output to the DC coefficient encoding circuit 401, and the other AC coefficients 1302 are output to the AC coefficient encoding circuit 503.

As shown in FIG. 10, the DC coefficient encoding circuit 401 includes a block delay unit 401a and an adder 40.
1b, the DC difference between the blocks is obtained, and the grouping unit 401c groups the DC differences according to the table of FIG. Further, the grouped DC differences are associated with the group number and additional bits indicating the value of the DC difference, and the group number is assigned to the one-dimensional Huffman coding unit 40.
One-dimensional Huffman coding is performed by 1d and the DC code table 401e, and an additional bit is added to output the data as a DC code.

The AC coefficient encoding circuit 503 rearranges the two-dimensionally arranged AC coefficients into one dimension by the zigzag scan unit 503a as shown in FIG. 12, and determines whether each coefficient is 0 or not. If it is 0, the length of a continuous run is counted by the multi-run length counting unit 503b. If it is not 0, grouping is performed by the grouping unit 503c according to the table in FIG. The grouped AC coefficients are associated with a group number and an additional bit indicating the value of the AC coefficient, and the run length and the group number are two-dimensionally Huffman-coded by the two-dimensional Huffman coding unit 503d and the AC code table 503e. And output as an AC code.

When the switching identification signal 1001 indicates a PDL image, the code data multiplexing circuit 601 forms output code data 1401 by multiplexing the lossless code data 1101 and the inter-block code data 1202. Therefore, even when the immediately preceding block is encoded by the lossless encoding method and the next block is switched to the irreversible encoding method, the DC difference information between the blocks of the irreversible encoding method is output code data 1401. , Lossless coding can be performed between the lossless coding block and the lossy coding block with reference to DC difference information regarding lossy coding.

Further, the code data multiplexing circuit 601
When the switching identification signal 1001 indicates a natural image, irreversible code data and inter-block code data 1202 are multiplexed to form output code data 1401. In this manner, efficient coding can be performed even when interblock dependent information is required for the lossless coding method. Of course, if the lossless encoding method does not require inter-block dependency information,
There is no need to output only irreversible data and multiplex inter-block coded data for lossless encoding.

It should be noted that the decoding device performs the reverse operation of the coding device, and has not been described in detail since it has already been schematically described in FIG.

As described above, according to this embodiment, the lossy coding method in which the coding information depends on the immediately preceding block, such as the JPEG method for coding the DC value difference between blocks, and the lossless coding method In an image coding method in which code data is formed by switching coding methods, coding dependent information between blocks is multiplexed to form output code data, and when coding is switched, reference is made to the coding dependent information of the output code data. Since the encoding is performed, the encoding efficiency does not decrease even when the encoding method is switched between adjacent blocks.

For example, as shown in FIG. 14, when the #N block is coded by the lossless coding method, an output code obtained by multiplexing the DC coefficient for the #N block with reference to the DC coefficient of the JPEG method. Organize the data. In this example, #
In the (N + 1) -th block, the switching from the lossless encoding method to the JPEG encoding method is performed.
Since the DC coefficient itself of the PEG method is held in the output code data of the block #N, the DC coefficient of the JPEG method is used.
Coefficients can be encoded with the same efficiency as if the encoding were not switched.

In the above description, an example has been described in which lossless coding and irreversible coding are switched and used.
The present invention is not limited to such a configuration,
In short, when there are two or more encoding units, these are switched and used, and when at least one of the encoding units uses the encoding dependence information, the dependence information is encoded when the other encoding units are encoded. Are multiplexed. Of course, the dependency information may not be always stored, but may be selectively stored in consideration of other efficiency factors.

[0052]

As described above, according to the present invention,
Even when a plurality of coding methods are used and at least one coding method has dependency information between blocks, the coding efficiency is not reduced by switching the coding method.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating an example of a principle configuration of an image encoding device according to the present invention.

FIG. 2 is a flowchart illustrating the operation of the present invention.

FIG. 3 is a diagram illustrating an operation example at the time of encoding switching.

FIG. 4 is a block diagram illustrating an example of a basic configuration of a decoding device according to the present invention.

FIG. 5 is a block diagram showing a specific configuration example of the present invention.

FIG. 6 is a flowchart showing the processing of the blocking circuit 101 of FIG.

FIG. 7 is a flowchart showing processing of the PDL / natural image identification circuit 201 in FIG. 5;

8 is a diagram illustrating an example of two-dimensional DCT transform in the DCT transform circuit 501 of FIG.

9 is a diagram illustrating an example of DC coefficients and AC coefficients in a transform coefficient block of the DCT transform circuit 501 of FIG.

10 is a block diagram illustrating a configuration example of a DC coefficient encoding circuit 401 in FIG.

11 is a diagram illustrating an example of grouping of DC differences in the DC coefficient encoding circuit 401 in FIG.

12 is a block diagram illustrating a configuration example of an AC coefficient encoding circuit 503 in FIG.

13 is a diagram illustrating an example of grouping of AC coefficients in the AC coefficient encoding circuit 503 in FIG. 5;

FIG. 14 is a diagram illustrating an example in which a DC coefficient is referred to in coding switching.

FIG. 15 is a diagram illustrating an example of a DC coefficient when encoding is performed by the JPEG method.

FIG. 16 is a diagram illustrating an example in which a DC coefficient is set to a fixed value 0 in encoding switching.

FIG. 17 is a diagram illustrating an example in which a DC coefficient is set to a fixed value 255 in encoding switching.

FIG. 18 is a block diagram showing a configuration of a conventional example.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Blocking means 2 Encoding switching means 3 Lossless encoding means 4 Encoding dependent information calculating means 5 Lossy encoding means 6 Code data multiplexing means 7 Input image data 8 Block image 9 Encoding target image 10 Switching identification information 11 Lossless code data 12 Coding dependency information 13 Lossy code data 14 Output code data 101 Blocking circuit 201 PDL / natural image discriminating circuit 301 Adaptive predictive lossless coding circuit 401 DC coefficient coding circuit 501 DCT transform circuit 502 DCT coefficient quantization Circuit 503 AC coefficient encoding circuit 601 Code data multiplexing circuit 701 Input image data 801 Block image 901 Image to be encoded 1001 Switching identification signal 1101 Reversible code data 1201 DC coefficient 1202 Inter-block dependent code data 1301 DCT coefficient 1 02 AC coefficients 1303 lossy encoding data 1401 output coded data

 ────────────────────────────────────────────────── ─── Continued on the front page (72) Inventor Kenichi Kawachi 430 Sakai Nakaicho, Ashigara-gun, Kanagawa Prefecture Green Tech Inside Fuji Xerox Co., Ltd. F-term (reference) 5C059 KK01 MA00 ME01 PP18 RC11 RC12 SS28 TA17 TB08 TC01 TC24 TC27 TC42 TD01 TD09 UA02 UA05 5C078 BA21 BA42 CA02 DA01 DA02 5J064 AA02 BA08 BA16 BB05 BB11 BB13 BD06 BD07

Claims (6)

[Claims] 1. An input image data is encoded by a first encoding method in which encoding is performed on a block basis and codes of adjacent blocks depend on each other, and a second encoding method different from the first encoding method. An image coding apparatus for converting input image data into output code data, comprising: blocking means for dividing input image data into blocks; and first coding for coding the input image data in block units using the first coding method. Means, second encoding means for encoding the input image data in block units by the second encoding method, encoding by the first encoding means, and encoding by the second encoding means Encoding switching means for selectively switching between encoding and encoding; and means for generating encoding-dependent information used by the first encoding means for encoding when the first encoding is not performed; 2 sign Multiplexing means for multiplexing the code output by the encoding means and the encoding-dependent information to form an output code. Further, the first encoding means is configured to An image encoding device, wherein encoding is performed using the encoding-dependent information. 2. The image coding apparatus according to claim 1, wherein one of the first coding method and the second coding method is a lossy coding method, and the other is a lossless coding method. 3. The image encoding apparatus according to claim 1, wherein said first encoding method performs encoding by utilizing a DC component dependency of image data. 4. The image encoding apparatus according to claim 3, wherein said first encoding method is a JPEG baseline encoding method. 5. The coding method according to claim 1, wherein the second coding method is also a coding method in which codes of adjacent blocks depend on each other.
5. The image coding apparatus according to 2, 3, or 4. 6. A decoding device for decoding an output code output from the image encoding device according to claim 1, wherein: a first decoding unit for decoding a code encoded by the first encoding unit; Second decoding means for decoding the code coded by the second coding means, code of the second coding method multiplexed by the multiplexing means, and coding of the first coding method Means for separating dependent information, image information obtained by decoding by the first decoding means,
Means for synthesizing image information obtained by decoding by the decoding means, and when the decoding by the first decoding means is restarted, the first encoding is performed by using the encoding-dependent information. A decoding device for decoding a method code.