JPH08111781A - Image coding device/method - Google Patents

Abstract

PURPOSE: To provide an image coding device/method which can perform the estimated coding with high coding efficiency from the beginning of conversion. CONSTITUTION: A peripheral element condition calculation part 6 calculates an optimum AT pixel position n' based on the optimum AT pixel position information D12 stored in a memory and outputs the optimum AT pixel position information D6 that prescribes the present position n' to a coding processing data generation part 2. At the same time, the information D12 is related to the image data D1 and serves as the information on the optimum AT pixel position of the coded reference image data. The part 2 decides an AT pixel based on the information D6 and extracts the coding processing data D2 based on the data D1 to output the data D2 to an arithmetic coder 3.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image coding apparatus and an image coding method for performing predictive conversion using surrounding elements.

[0002]

2. Description of the Related Art Since image data has a very large amount of information, it is not practical to directly process the image data in terms of memory capacity and communication speed. Therefore, image data compression technology is required.

One of the encoding methods used for image data compression is JBIG (Joint Bi-level Image Coding Experts).
Group) and JPEG (Joint Photographic Expert Grou)
There is an arithmetic code that is adopted in international standards such as p). First, this arithmetic code will be explained. The arithmetic code is also called a number line code, and the corresponding section on the number line between the real numbers 0 and 1 is divided into unequal lengths according to the occurrence probability of each symbol,
The coordinates of the points included in the section obtained by repeating the allocation of the symbol sequence to the corresponding section are represented as binary numbers by at least a binary fraction that can be distinguished from other sections. For example, symbol series 01
FIG. 10 shows a conceptual diagram in the case of encoding 00.

The section A (0) divided and selected by the occurrence probability of the first symbol 0 is further divided into the section A (01) by the occurrence probability when the second symbol 1 is encoded. Such division selection is repeated and finally the section A (0
100) is generated. Coordinates included in the resulting section are represented by a binary number which can be expressed by being distinguished from other sections. Therefore, the code length can be considered as the required expression precision.

Predictive conversion is one of the methods suitable for the arithmetic code. This is, for example, in encoding a pixel in an image, predicts the value of the pixel by referring to a plurality of pieces of pixel information preceding it, which is a surrounding element, and encodes the value.

This predictive conversion will be described by taking the above-mentioned JBIG as an example. FIG. 11 shows a part of the binary image, which is one of the so-called model templates. In the figure, “?” Is a coded pixel of interest, “X” and “A” are reference pixels when the pixel of interest “?” Is coded, and binary (“1”, “0”, white, black, etc.). ) Takes one value.

The encoder has the predicted value of the coded pixel "?" And its estimated probability for all combinations of the values of the respective symbols of the pixels "X" and "A". Then, using the predicted value and the actual value of the coded pixel, prediction conversion is performed and the estimated probability is changed. In JBIG, a symbol in arithmetic code that has been subjected to the predictive conversion process is used, and the estimated probability is used as the occurrence probability of the symbol.

[0008]

The coding method (especially the arithmetic coding method) using the predictive conversion by the conventional image coding apparatus improves the coding efficiency, that is, the data compression rate by performing accurate prediction. Is known to do.

For this accurate prediction, it is better to change the position of the reference pixel depending on the situation without fixing it. Even in JBIG, the pixel “A” in FIG. 11 is AT (Adaptive Templat).
es) A reference pixel that is called a pixel and is allowed to be changed to another position. However, when the AT pixel is moved, the coding efficiency temporarily decreases, so it is important to immediately determine the optimum position.

However, in the conventional method by the image coding apparatus, the coding process is advanced as an arbitrary (practically fixed) AT pixel position at the beginning of the coding process, and the optimum AT is executed in parallel with the AT pixel position.
The pixel position is examined, and after the processing of a certain amount of pixels is completed, it is changed to the examined optimum AT pixel position, and the remaining pixels are encoded at the changed AT pixel position.

The search for the optimum AT pixel position referred to here is
For example, it is a survey by a method of examining the number of times each candidate AT pixel and the encoded pixel match. Therefore, if any AT pixel position before the investigation is not optimal, there is a problem that the coding efficiency is lowered.

The present invention has been made to solve the above problems, and an object thereof is to obtain an image coding apparatus and an image coding method capable of predictive coding with high coding efficiency from the beginning of conversion.

[0013]

According to a first aspect of the present invention, there is provided an image coding apparatus, which predicts and transforms a value of each pixel defined by image data to be coded based on a first surrounding element and codes the coded value. An encoding device, which is based on information of a second surrounding element for predictive conversion of reference image data which is related to the image data to be encoded and is already encoded,
Surrounding element determining means for determining surrounding elements of the image data and the image data to be encoded, and predictively transforms the value of each pixel defined by the image data based on the first surrounding element to obtain an encoded image. And an encoded data generation unit that generates data.

Further, as in the image coding apparatus according to the second aspect, when the surrounding element determining means has different resolutions between the image data to be coded and the reference image data,
The first surrounding element may be determined so as to match the resolution of the image data to be encoded.

Further, as in the image coding apparatus according to the third aspect, the pixel of interest on the image data to be coded is used as a base point to selectively select from the pixel group defined by the image data to be coded. The extracted pixel may be the first surrounding element.

Further, as in the image coding apparatus according to the fourth aspect, a reference image data which is closely related to the image data to be coded and which stores the reference image data in a state before being encoded. Image data information storage means is further provided, and a pixel on the reference image data corresponding to a pixel of interest on the image data to be encoded is used as a base point to selectively select from a pixel group defined by the reference image data. The extracted pixel may be the first surrounding element.

An image coding method according to a fifth aspect of the present invention is a method of predictively converting the value of each pixel defined by the image data to be coded based on the first surrounding element and coding the value. , (A) determining the first surrounding element based on information of the second surrounding element for predictive conversion of the reference image data that has already been encoded and is related to the image data to be encoded. And (b) a step of predictively converting the value of each pixel of interest of the image data to be encoded based on the first surrounding element to generate encoded image data.

[0018]

According to the present invention, the surrounding element determining means in the image coding apparatus according to claim 1 is related to the image data to be coded, and is the second surrounding for predictive conversion of already coded reference image data. The first surrounding element is determined based on the element information.

The second peripheral element has been used for predictive conversion of already encoded reference image data and has sufficiently high prediction accuracy, so that the image data to be encoded and the reference image data are related to each other. If so, by determining the first surrounding element based on the information of the second surrounding element, it is possible to predict the pixel of interest of the image data to be encoded with relatively high prediction accuracy from the beginning of encoding.

Further, the surrounding element determining means of the image coding apparatus according to claim 2 is adapted to match the resolution of the image data to be coded when the image data to be coded and the reference image data have different resolutions. Since the first surrounding element is determined, if there is some correlation between the image data to be encoded and the reference image data even if the resolutions are different, the image to be encoded with relatively high prediction accuracy from the beginning of encoding. The pixel of interest in the data can be predicted.

The image coding apparatus according to claim 3 is
Since the pixel selectively extracted from the pixel group defined by the image data to be encoded is used as the first peripheral element with the pixel of interest on the image data to be encoded as the base point, the image data to be encoded The encoding can be achieved by using only.

An image coding apparatus according to a fourth aspect is:
A pixel selectively extracted from a pixel group defined by the reference image data is used as a first surrounding element, with a pixel on the reference image data corresponding to a target pixel on the image data to be encoded as a base point. Therefore, if the reference image data and the image data to be encoded are highly related, the pixel of interest of the image data to be encoded can be predicted from the beginning of conversion with high prediction accuracy.

According to a fifth aspect of the present invention, in the image encoding method according to the fifth aspect, in the step (a), the second encoding for predictive conversion of the already encoded reference image data is associated with the image data to be encoded. The first surrounding element is determined based on the information on the surrounding elements.

The second peripheral element has been used for predictive conversion of already encoded reference image data and has sufficiently high prediction accuracy, so that the image data to be encoded and the reference image data are related to each other. If so, by determining the first surrounding element based on the information of the second surrounding element, it is possible to predict the pixel of interest of the image data to be encoded with relatively high prediction accuracy from the beginning of encoding.

[0025]

【Example】

<First Embodiment> FIG. 1 is a block diagram showing the arrangement of an image coding apparatus according to the first embodiment of the present invention. As shown in the figure, a binary data input unit 1 sequentially takes in image data D1 obtained by binarizing an image taken in by an image sensor or the like, and outputs it to an encoding processing data generation unit 2.

The encoding process data generation section 2 extracts the encoding process data D2 consisting of the pixel data of interest to be encoded and the reference pixel data for predictive conversion of the pixel data of interest, based on the image data D1. And outputs it to the arithmetic encoder 3. At this time, the encoding processing data generation unit 2 determines the AT pixel data that is a part of the reference image data based on the optimum AT pixel position information D6 from the surrounding element condition calculation unit 6.

The arithmetic encoder 3 is provided with the encoding processing data D2.
In response to the reference pixel data, the pixel data of interest is predictively converted and arithmetically encoded to output the encoded image data D3 to the outside through the data output unit 7, and the predicted correct / incorrect data D
13 is output to the optimum surrounding element condition inspection unit 5, and
The predicted value and the occurrence probability of the pixel of interest are output and stored in the memory 4.

The optimum surrounding element condition examining unit 5 examines the optimum AT pixel position n, which is determined to have the highest prediction accuracy, based on the prediction correctness data D13, and performs a certain amount of encoding processing on the image data D1. After the end, the optimum AT pixel position information D5 defining the optimum AT pixel position n for the image data D1 currently being encoded is output to the surrounding element condition calculation unit. Then, the optimum surrounding element condition examining unit 5 outputs the optimum AT pixel position n at the end to the memory 12 when the encoding process of the image data D1 is completed.

The memory 12 stores the final optimum AT pixel position information D5 output from the optimum surrounding element condition examining unit 5 as memory storage optimum AT pixel position information D12.

At the beginning of the encoding process, the surrounding element condition calculation unit 6 calculates the optimum AT pixel position n'on the basis of the memory storage optimum AT pixel position information D12 to define the current optimum AT pixel position n '. The optimum AT pixel position information D6 is output to the encoding processing data generation unit 2. At this time, the memory storage optimum AT pixel position information D12 is related to the image data D1 and is the optimum A of the already encoded reference image data.
It becomes the information of the T pixel position.

Then, as the coding process of the image data D1 progresses, the optimum surrounding element condition examining unit 5 makes the optimum AT.
When the pixel position information D5 is output, the surrounding element condition calculation unit 6 again calculates the optimum AT pixel position n'on the basis of the optimum AT pixel position information D5, and the optimum AT pixel position that defines the current optimum AT pixel position. The position information D6 is output to the encoding processing data generation unit 2.

FIG. 2 is a flow chart showing the operation of the image coding apparatus according to the first embodiment.

Referring to FIG. 3, in step S1, the binary data input unit 1 takes in the image data D1 and sequentially outputs the image data D1 to the encoding processing data generation unit 2.

In step S2, the surrounding element condition calculation unit 6
Is the memory storage optimum AT pixel position information D from the memory 12.
Take in 12. At this time, in the memory 12, in the previous image data encoding process, the reference image data that has already been encoded and is related to the image data D1 is finally determined to be optimal by the optimal surrounding element condition inspection unit 5. Memory storage optimum AT pixel position information D12 defining the optimum AT pixel position n is stored.

If the image data to be encoded this time and the image data previously encoded previously have a high relevance, the optimum A specified by the memory storage optimum AT pixel position information D12 is obtained.
The T pixel position n is very useful as information for determining the optimum AT pixel position of the image data to be encoded this time.

Then, in step S3, the surrounding element condition calculating unit 6 calculates the optimum AT pixel position n'on the basis of the memory storage optimum AT pixel position information D12, and in step S4 the current optimum AT pixel position n '. The specified optimum AT pixel position information D6 is output to the encoding processing data generation unit 2.

For example, consider a case where the image data to be encoded has a high relevance to the already encoded image data and the resolution is double. The resolution doubled means that the resolution is doubled both horizontally and vertically, that is, the number of pixels is quadrupled. In addition,
Double-resolution transmission is one of the JBIG standard encoding procedures.

At this time, with respect to the optimum AT pixel position n indicated by the memory storage optimum AT pixel position information D12, the optimum AT pixel position n'at this time is calculated as follows, and the calculation processing is performed in consideration of the resolution. To do.

N ′ = 2 × n (n ≦ 8) n ′ = n (n> 8) Then, the surrounding element condition calculation unit 6 codes the optimum AT pixel position information D6 defining the optimum AT pixel position n ′. The data is output to the data processing unit 2 for conversion processing.

FIG. 3 is an explanatory diagram showing the positional relationship between the pixel of interest and the reference pixel in the image data to be encoded. In the figure, “?” Is the pixel of interest to be encoded,
“X” is a fixed reference pixel whose relative position is fixed when the pixel of interest “?” Is encoded.

Then, in P5 to P16, the pixel of interest "?"
Corresponding to the horizontal distance from the pixel of interest. That is, if i = n ', then Pi
Is the AT pixel position. Therefore, the reference pixel for predictive conversion of the target pixel “?” Is composed of nine fixed reference pixels “X” and one AT pixel.

In the image data D1, as shown in FIG. 4, the pixels gi, j arranged in a matrix form "g1,1, g".
2, 1, ..., G13,1, g1,2, g2,2 ,. However, i of gi, j means the x-direction coordinate of FIG. 4, and j means the y-direction coordinate. In the example of FIG. 4, g8,4
Indicates the case of the target pixel, and in this case, g5,3 to g
9,3 and g4,4 to g7,4 are fixed reference pixels, and g2,4 are AT pixels.

Returning to FIG. 2, in step S5, the encoding process is executed by the encoding data generator 2 and the arithmetic encoder 3. That is, the encoding processing data generation unit 2
Is the optimum AT pixel position information D6 based on the image data D1.
The encoding processing data D2 is extracted based on the above, and is output to the arithmetic encoder 3. The arithmetic encoder 3 uses the encoding processing data D.
2, the pixel data of interest is predictively converted based on the reference pixel data, arithmetically encoded, and the encoded image data D3 is output to the outside through the data output unit 7, and the predicted correct / incorrect data D13 is searched for the optimum surrounding element condition. The prediction value and the occurrence probability of the pixel of interest are output to the unit 5 and stored in the memory 4.

On the other hand, during this period, the optimum surrounding element condition investigation unit 5
Is the optimum AT pixel position n for which the prediction accuracy is determined to have the highest value based on the prediction correctness data D13, and the optimum AT pixel position for the image data D1 currently being encoded after a certain amount of encoding is completed. The optimum AT pixel position information D5 defining the position n is output to the surrounding element condition calculation unit 6.

Thereafter, in step S6, the encoding process data generator 2 determines whether or not all the encoding of the image data D1 is completed. If YES, the encoding end process is executed in step S8. The final optimum AT pixel position information D5 is output from the optimum surrounding element condition examining unit 5 to the memory 12, and the optimum AT pixel position information D5 is stored in the memory 12 as the memory storage optimum AT pixel position information D12.

On the other hand, if NO in step S6, the process proceeds to step S7. In step S7, the surrounding element condition computing unit 6 sends the optimum surrounding element condition examining unit 5 to the optimum AT.
If the pixel position information D5 is output, the optimum AT pixel position n is recognized from the optimum AT pixel position information D5, the optimum AT pixel position information D6 is changed so as to define the optimum AT pixel position n, and the code The data is output to the data processing unit 2 for conversion processing.

Thereafter, steps S4 to S7 are repeated until YES is determined in step S6.

As described above, the image coding apparatus of the first embodiment performs the coding based on the AT pixel position at the end of the coding process of the already coded image data which is related to the image data to be coded. Since the AT pixel position with respect to the target image data is determined and the encoding process of the image data to be encoded is started, it is possible to improve the prediction accuracy of the pixel of interest from the beginning of encoding, and the conversion from the arithmetic encoder 3 is performed. From the beginning, it is possible to output the encoded image data D3 having high encoding efficiency.

When the image data to be encoded and the already encoded image data for AT pixel position determination have a very strong relationship, as shown in the flowchart of FIG.
The step of S7 may be omitted and the transition step in the case of NO of S6 may be changed to S5.

That is, based on the optimum AT pixel position n designated by the memory storage optimum AT pixel position information D12, the image data D1 is fixed by predicting conversion at the optimum AT pixel position n'determined by the surrounding element condition calculating unit 6. Even if the encoding process is performed for, since the reliability of the optimum AT pixel position n ′ is high, the encoding efficiency can be significantly improved as compared with the conventional case.

When the image data to be coded and the reference image data have different resolutions, the surrounding element condition computing unit 6 determines the optimum AT pixel position n'in consideration of the resolutions, so that the resolutions are different. However, if there is any correlation between the image data to be encoded and the reference image data, the pixel of interest of the image data to be encoded can be predicted with relatively high prediction accuracy from the beginning of encoding.

As a result, it is possible to obtain an image coding apparatus having high coding efficiency even if the image data to be coded and the reference image data have different resolutions.

In the predictive conversion of the image data D1,
Since the reference pixels extracted from the pixels defined by the image data to be encoded are used as peripheral elements, it is possible to perform encoding using only the image data to be encoded.

<Second Embodiment> FIG. 6 is a block diagram showing the arrangement of an image coding apparatus according to the second embodiment of the present invention. As shown in the figure, the binary data input unit 1 takes in image data D1 obtained by binarizing an image taken in by an image sensor or the like, and stores the image data D1 in the frame memory 8
And output to and stored in the data generation unit 2 for encoding processing.
Or output to.

The frame memory 8 stores the image data D1 and outputs it to the encoding processing data generating section 2 as the stored image data D8 which becomes the reference image data.

The encoding processing data generating section 2 uses the image data D1 and the stored image data D8 to generate the image data D.
The pixel data of interest is extracted from 1 and the reference pixel data is extracted from the stored image data D8, and these extracted data are output to the arithmetic encoder 3 as encoding process data D2.

FIG. 7 is an explanatory diagram showing details of the operation of the encoding processing data generating section 2. FIG. 7 shows image data D1 directly output from the binary data input unit 1 to the encoding processing data generation unit 2 and stored image data D8 stored in the frame memory 8.

For example, when the color image data is finally obtained from the R, G, and B image data, when the G image data and the B image data are image data D1, the G image data and the B image data are closely related. It is expected that the prediction accuracy can be improved by using the R image data that is related to each other and is not encoded yet as the stored image data D8.

Since the other structure is the same as that of the image coding apparatus of the first embodiment shown in FIG. 1, the description thereof will be omitted.

The operation of the image coding apparatus of the second embodiment is the same as the above-described operation of the coding processing data generating section 2 in the coding processing step of step S5 of FIG. 2 (FIG. 5). Only the changes are made, and the other contents are the same as the operation of the first embodiment.

As described above, even when the reference pixel including the AT pixel is extracted from the reference image data different from the image data to be encoded, the prediction accuracy of the pixel of interest is improved from the beginning of the encoding as in the first embodiment. The coding efficiency can be improved from the beginning of conversion.

If the reference image data and the image data to be coded have a strong relationship, the pixel of interest of the image data to be coded can be predicted with higher prediction accuracy than in the first embodiment.

In the second embodiment, the pre-encoding image data which is not the encoding target is stored as it is in the stored image data D.
However, the difference data between the image data before encoding and the image data to be encoded may be the stored image data D8.

<Third Embodiment> FIG. 8 is a block diagram showing the arrangement of an image coding apparatus according to the third embodiment of the present invention. As shown in the figure, a binary data input unit 1 takes in image data D1 obtained by binarizing an image taken in by an image sensor or the like and outputs it to an encoding processing data generation unit 2.

The data table 9 outputs the fixed image data D9, which is the previously stored reference image data, to the encoding data generator 2.

The encoding processing data generating section 2 uses the image data D1 and the fixed image data D9 to generate the image data D.
The pixel data of interest is extracted from 1, the reference pixel data is extracted from the fixed image data D9, and these extracted data are output to the arithmetic encoder 3 as the encoding processing data D2.
The operation of the encoding data generator 2 is to store the stored image data D
The operation is the same as that of the encoding processing data generation unit 2 of the second embodiment except that 8 is replaced with the fixed image data D9.

Since the other structure is the same as that of the image coding apparatus of the first embodiment shown in FIG. 1, its explanation is omitted.

Further, the operation of the image coding apparatus of the third embodiment is the same as the above-described operation of the coding processing data generating section 2 in the coding processing step of step S5 of FIG. 2 (FIG. 5). Only the changes are made, and the other contents are the same as the operation of the first embodiment.

As described above, even if the reference pixel including the AT pixel is extracted from the fixed image data different from the image data to be encoded, as in the first and second embodiments, the pixel of interest is relatively focused from the beginning of encoding. It is possible to improve the prediction accuracy of, and to improve the coding efficiency from the beginning of conversion.

In addition, by using the data table 9 that has a read function, compared with the second embodiment, the degree of integration which does not require the frame memory 8 having a relatively large circuit scale that requires reading and writing. Can be improved. Further, since the fixed image data D9 is stored in the data table 9 in advance, when compared with the second embodiment, the processing time can be improved because the time for storing the stored image data D8 can be omitted.

<Fourth Embodiment> FIG. 9 is a block diagram showing the arrangement of an image coding apparatus according to the fourth embodiment of the present invention. As shown in the figure, the multi-valued data input unit 10 takes in multi-valued image data D10 obtained by performing multi-valued processing on an image taken in by an image sensor or the like, and outputs it to the orthogonal transformation unit 11.

The orthogonal transformation unit 11 uses the multivalued image data D10.
Is divided into, for example, a plurality of 8 × 8 pixel blocks, subjected to a discrete cosine transform process, subjected to orthogonal transform, and outputs orthogonal transform image data D11 to the encoding process data generation unit 2.

The encoding processing data generating section 2 is based on the orthogonal transformation image data D11 and then uses the orthogonal transformation image data D11.
The pixel data of interest and the reference pixel data are extracted from, and the encoding processing data D2 is output to the arithmetic encoder 3. The operation of the encoding processing data generation unit 2 is the same as the operation of the encoding processing data generation unit 2 of the first embodiment except that the image data D1 is replaced with the orthogonal transformation image data D11.

Since the other structure is the same as that of the image coding apparatus of the first embodiment shown in FIG. 1, its explanation is omitted.

Further, when the operation of the image coding apparatus of the fourth embodiment is compared with the operation of the first embodiment shown in FIG. 2 (FIG. 5), the step S1 is performed by the multi-valued data input unit 10. By capturing the multi-valued image data D10 and outputting it to the orthogonal transformation unit 11, the orthogonal transformation unit 11 orthogonally transforms the multi-valued image data D10 and outputs the orthogonal transformation image data D11 to the encoding processing data generation unit 2, The process is changed to a process of capturing image data.

Further, the operation of the encoding processing data generating section 2 in the encoding processing step of step S5 is changed to a step of processing based on the orthogonal transformation image data D11.
The other contents are the same as the operation of the first embodiment.

As described above, the image coding apparatus according to the fourth embodiment determines the AT pixel position based on the AT pixel position at the end of the coding process of the image data related to the multivalued image data to be coded. Then, since the multivalued image data is converted into the pseudo image data having a high compression rate and the encoding process is started, it is possible to relatively improve the prediction accuracy of the pixel of interest from the beginning of the encoding. The coding efficiency can be increased from the beginning of conversion.

<Others> Instead of the binary data input unit 1 of the first to third embodiments, a multivalued data input unit 10 may be provided to take in the multivalued image data D10. In this case, it is desirable that the image data stored in the frame memory 8 and the data table 9 is also multivalued image data.

[0079]

As described above, the surrounding element determining means in the image coding apparatus according to the first aspect of the present invention relates to the image data to be coded and has already been coded for the reference image data. The first surrounding element is determined based on the information of the second surrounding element for predictive conversion.

The second peripheral element has been used for predictive conversion of already encoded reference image data and has sufficiently high prediction accuracy. Therefore, the relation between the image data to be encoded and the reference image data is high. If so, by determining the first surrounding element based on the information of the second surrounding element, it is possible to predict the pixel of interest of the image data to be encoded with relatively high prediction accuracy from the beginning of encoding.

As a result, it is possible to generate coded image data with high coding efficiency from the beginning of conversion from the coded data generating means.

Further, when the resolutions of the image data to be encoded and the reference image data are different from each other, the surrounding element determining means of the image encoding apparatus according to the second aspect is adapted to match the resolution of the image data to be encoded. Since the first surrounding element is determined, if there is some correlation between the image data to be encoded and the reference image data even if the resolutions are different, the image to be encoded with relatively high prediction accuracy from the beginning of encoding. The pixel of interest in the data can be predicted.

As a result, even if the image data to be encoded and the reference image data have different resolutions, the encoded data generating means can generate encoded image data with high encoding efficiency from the beginning of conversion. .

The image coding apparatus according to claim 3 is
Since the pixel selectively extracted from the pixel group defined by the image data to be encoded is used as the first peripheral element with the pixel of interest on the image data to be encoded as the base point, the image data to be encoded The encoding can be achieved by using only.

The image coding apparatus according to claim 4 is
A pixel selectively extracted from a pixel group defined by the reference image data is used as a first surrounding element, with a pixel on the reference image data corresponding to a target pixel on the image data to be encoded as a base point. Therefore, if the reference image data and the image data to be encoded are highly related, the pixel of interest of the image data to be encoded can be predicted from the beginning of conversion with high prediction accuracy.

According to a fifth aspect of the present invention, in the image encoding method according to the fifth aspect, in the step (a), the second image encoding method for predictive conversion of the previously encoded reference image data which is related to the image data to be encoded. The first surrounding element is determined based on the information on the surrounding elements.

The second peripheral element is used for predictive conversion of already encoded reference image data and has sufficiently high prediction accuracy. Therefore, the relation between the image data to be encoded and the reference image data is high. If so, by determining the first surrounding element based on the information of the second surrounding element, it is possible to predict the pixel of interest of the image data to be encoded with relatively high prediction accuracy from the beginning of encoding.

As a result, in step (b), it is possible to output coded image data with high coding efficiency from the beginning of conversion.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of an image coding apparatus that is a first embodiment of the present invention.

FIG. 2 is a flowchart showing an operation of the image coding apparatus according to the first embodiment.

FIG. 3 is an explanatory diagram showing AT pixel candidate positions.

FIG. 4 is an explanatory diagram showing details of image data.

FIG. 5 is a flowchart showing another operation of the image coding apparatus according to the first embodiment.

FIG. 6 is a block diagram showing a configuration of an image coding apparatus which is a second embodiment of the present invention.

FIG. 7 is an explanatory diagram for explaining the operation of an encoding processing data generation unit in the image encoding device according to the second embodiment.

FIG. 8 is a block diagram showing a configuration of an image coding apparatus which is a third embodiment of the present invention.

FIG. 9 is a block diagram showing a configuration of an image coding apparatus which is a fourth embodiment of the present invention.

FIG. 10 is an explanatory diagram showing the concept of arithmetic codes.

FIG. 11 is an explanatory diagram showing reference pixel positions.

[Explanation of symbols]

1 binary data input unit, 2 encoding data generation unit, 3 arithmetic encoder, 4, 12 memory, 5 optimal surrounding element condition checking unit, 6 surrounding element condition computing unit, 7 data output unit, 8 frame memory, 9 data table, 10
Multi-value data input unit, 11 Orthogonal transformation unit.

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI technical display location H04N 1/41 B 7/32

Claims (5)

[Claims] 1. An image encoding apparatus for predictively converting and encoding the value of each pixel defined by image data to be encoded based on a first surrounding element, the image encoding device being associated with the image data to be encoded. The surrounding element determining means for determining the first surrounding element based on the information of the second surrounding element for predictive conversion of the already encoded reference image data, and receiving the image data to be encoded. An image coding apparatus, comprising: a coded data generation unit that predictively converts the value of each pixel defined by the image data based on the first surrounding element to generate coded image data. 2. The surrounding element determining means, when the image data to be encoded and the reference image data have different resolutions, the first surrounding element is adapted to match the resolution of the image data to be encoded. Determine the elements,
The image coding apparatus according to claim 1. 3. The first surrounding element is a pixel selectively extracted from a pixel group defined by the image data to be encoded, with a pixel of interest on the image data to be encoded as a base point. The image coding apparatus according to claim 1 or 2. 4. A reference image data information storage unit for storing reference image data in a state before being encoded, which is closely related to the image data to be encoded and further comprises: The surrounding element is a pixel selectively extracted from a pixel group defined by the reference image data, with a pixel on the reference image data corresponding to a pixel of interest on the image data to be encoded as a base point. The image coding apparatus according to claim 1 or 2. 5. An image encoding method for predictively converting and encoding the value of each pixel defined by image data to be encoded based on a first surrounding element, comprising: (a) the image to be encoded Determining the first surrounding element based on the information of the second surrounding element for predictive conversion of the reference image data that is related to the data and has already been encoded; and (b) the encoding target Predicting and converting the value of each pixel of interest of the image data based on the first surrounding element to generate encoded image data.