JP2003309848A - Image encoding apparatus and control method thereof - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image encoding apparatus and a control method thereof capable of reducing a hardware scale by adopting the configuration of the apparatus wherein parts of units for performing entropy encoding processing are shared. <P>SOLUTION: A coefficient distribution section 101 divides a DCT coefficient matrix of received image data into areas and distributes AC components to run length decoders 102, 104 and DC components to a predictive value arithmetic section 103. A variable length encoding section 107 or 110 outputs an encoded bit stream by one block on the basis of symbol data comprising combinations of zero length and non-zero coefficient of received data and a prediction error distributed by a distribution section 105. Further, a data connection section 112 connects encoded bit streams by each block to produce an encoded bit stream with respect to the original image data. <P>COPYRIGHT: (C)2004,JPO

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 for performing variable length coding of image data and a control method thereof, and more particularly, to JPEG (Joint Photo) of image data.
multi-gradation image coding method such as a to digital Experts Group) and MPEG (Moti).
on Pictures Experts Group)
The present invention relates to an image encoding apparatus suitable for use in a moving image encoding method such as an encoding method and a control method thereof.

[0002]

2. Description of the Related Art A variable length coding system is one of the effective techniques for compressing and coding image data. The variable-length coding method, for example, when coding an information sequence consisting of alphabets, according to the appearance frequency of each alphabet of the information sequence, a code with a short code length is used for a high-frequency alphabet, and a code with a low frequency is used. This is a method of shortening the average code length by assigning a code with a long code length to the alphabet.

In such a variable length coding method of image data, a two-dimensional Huffman coding method is generally used, and it is also used in a JPEG coding method, an MPEG coding method and the like.

FIG. 6 is a block diagram showing the arrangement of an image coding apparatus which performs variable length coding processing. In the image coding apparatus, the image data is divided into pixel blocks of 8 pixels × 8 pixels, and the coding process is executed for each pixel block. That is, each pixel block is input to the discrete cosine transform (DCT) unit 601, where the two-dimensional discrete cosine transform (DCT) is performed, and an 8 × 8 transform coefficient matrix is obtained.

Next, these transform coefficient matrices are quantized by a quantizer 602 and further entropy coder 60.
In 3, entropy coding is performed by the two-dimensional Huffman coding method and output to the outside of the device.

Here, the entropy encoder 603 based on the two-dimensional Huffman encoding method will be described in detail.
In the entropy encoder 603, the coefficient distribution unit distributes each element of the 8 × 8 DCT transform coefficient matrix to the encoder or the prediction value calculation unit.

8 × quantized by quantizer 602
Each element of the 8 two-dimensional DCT coefficient matrix is scanned in a so-called zigzag scan, and 64 1's are scanned.
Converted to a dimensional array. Here, the first (first) element of the one-dimensional array is the direct current (DC) of the DCT transform coefficient matrix.
Represents an ingredient. In the zigzag scan, the DCT transform coefficient matrix is scanned diagonally in order from the low band to the high band.

The DC component of the distributed DCT transform coefficient matrix (the first component of the one-dimensional array) is input to the prediction value calculation section, where the prediction error value is calculated. Then, the variable length coding unit refers to the variable length coding table and assigns the variable length code to the calculated prediction error value.

The coefficient distributor distributes 63 AC components other than the DC component of the DCT transform coefficient matrix to the zero run length / non-zero coefficient detector. Then, the non-zero coefficient amplitude (M) and the preceding zero run length (the length of the zero coefficient) (L) preceding the non-zero coefficient amplitude (M) are sequentially combined, and these combinations (L, M) are output. Then, the variable length coding unit refers to the two-dimensional variable length coding table,
A variable length code is assigned to the output (L, M) combination.

Conventionally, as a method for realizing the speedup of the variable length coding processing as described above, there has been a method in which a part of the processing series is parallelized and the data processing is performed in parallel.

[0011]

However, in order to perform the variable length coding processing in parallel, it is necessary that the prediction value calculation unit, the variable length coding unit, etc. in the entropy encoder are parallel. Therefore, there is a problem that the hardware scale becomes large.

The present invention has been made in view of the above circumstances, and an image code whose hardware scale can be reduced by sharing a part of an apparatus for performing entropy coding processing. It is an object of the present invention to provide an apparatus and a control method thereof.

[0013]

In order to solve the above-mentioned problems, the present invention is an image coding apparatus for coding a series of transform coefficients obtained by orthogonally transforming image data, wherein the series of transforms. Coefficient distribution means for dividing the coefficient into n parts and distributing it to a predetermined encoding means, and at least n or more encoding means for encoding the series of divided conversion coefficients to generate encoded data. , And block-encoded bit string concatenation means for concatenating the generated n pieces of encoded data to generate a bit string.

Further, in the image coding apparatus according to the present invention, the coefficient distribution means divides the series of transform coefficients into one DC component and (n-1) AC components, and the coding means. Is provided with one DC encoding means for encoding the DC component of the series of transform coefficients, and at least (n-1) AC encoding means for encoding the AC component of the series of transform coefficients. Is characterized by.

Furthermore, the present invention is an image coding apparatus for coding a series of transform coefficients obtained by orthogonally transforming image data, and a DC coding means for coding the DC component of the transform coefficients. A plurality of alternating current encoding means for encoding the alternating current component of the conversion coefficient, a coefficient distributing means for distributing the alternating current component of the conversion coefficient to a predetermined alternating current encoding means, and a direct current component and an alternating current component of the conversion coefficient. Block coded bit string concatenation means for concatenating encoded data to generate a bit string.

Furthermore, the present invention is an image coding apparatus for coding a series of transform coefficients obtained by orthogonally transforming image data divided into predetermined blocks, wherein the series of transform coefficients of each block are encoded. DC coding means for coding the DC component of, a plurality of AC coding means for coding the AC component of the series of conversion coefficients in block units, and a predetermined AC component of the series of conversion coefficients in block units Coefficient distributing means for distributing to the AC encoding means, a plurality of block coded bit string connecting means for connecting the AC component and DC component encoded data of the transform coefficient of each block to generate a bit string, and the series of the same block A distribution unit that distributes the encoded data of the AC component and the DC component of the transform coefficient to a predetermined block coded bit string concatenation unit and the generated bit string for each block are combined. Characterized in that it comprises an image encoded bit string connecting means for generating bit streams of image data.

Furthermore, the image coding apparatus according to the present invention is characterized in that the AC coding means is provided with at least the number of blocks into which the image data is divided. Furthermore, the image coding apparatus according to the present invention is characterized in that the DC coding means predictively codes the DC component of the input series of transform coefficients.

With the above configuration, in the JPEG encoding system, the prediction error calculation for the DC component is performed only once in the run length encoding of one block. Therefore, one DC component is used by a plurality of run length encoders. It becomes possible to reduce the scale of hardware by sharing the predicted value calculation processing for.

[0019]

BEST MODE FOR CARRYING OUT THE INVENTION An image coding apparatus and a control method thereof according to an embodiment of the present invention will be described below with reference to the drawings. In the following embodiments, the input image signal is
For each pixel, R (Red), G (Green), B
It is assumed that the data format is composed of 3 elements of (Blue) and has 8 bits each. Further, one block is 8 pixels × 8 pixels obtained by dividing the input image data.

FIG. 1 is a block diagram showing the arrangement of an image coding apparatus according to an embodiment of the present invention. As shown in FIG. 1, the image coding apparatus according to the present embodiment has a coefficient distribution unit 101, run length encoders 102 and 104, a predicted value calculation unit 103, a distribution unit 105, and a variable length coding unit 10.
7, 110, DC encoding tables 108, 109, AC
Encoding tables 106 and 111 and data connection unit 11
It consists of two.

FIG. 2 is a schematic diagram for explaining the outline of the coefficient distribution operation using the image coding apparatus shown in FIG.
Further, FIG. 3 is a flowchart for explaining an operation procedure of the coefficient distribution unit 101 of the image coding device according to the present invention.

First, the transform coefficient matrix composed of 8 rows × 8 columns elements is input to the coefficient distribution unit 101 in block units (step S31). The coefficient distribution unit 101 divides the input transformation matrix into a DC component and an AC component (step S32). Further, the coefficient distribution unit 101 determines whether it is a DC component or an AC component (step S33), and the DC component is distributed to the predicted value calculation unit 103 (step S33).
34), the AC component is distributed to either one of the run length encoders 102 or 104 (step S35).

FIG. 4 is a diagram for explaining an example of a signal waveform distributed by the coefficient distribution unit 101 in the image coding apparatus according to the present invention. In the present embodiment, for example, in cycle t4 of FIG. 4, the direct current component (DC) of the transform coefficient matrix of block 1 shown in FIG. The AC component (AC1) is distributed. Further, in cycle t5 of FIG. 4, the direct current component (DC) of the transform coefficient matrix of block 2 following the transform coefficient matrix of block 1 shown in FIG.
Is distributed to the prediction value calculation unit 103, and the AC component (AC1) of the same matrix is distributed to the run length encoder 104.

As described above, one prediction value calculation unit 103 is provided for each of the two run length encoders 102 and 104.
By sharing, the use efficiency of the predicted value calculation unit 103 is increased and the hardware scale can be reduced, as compared with the case where the predicted value calculation is performed in each run length encoder.

Further, it is possible to easily realize the configuration in which the direct current component of the block processed immediately before, which is necessary for the calculation of the prediction error value in the prediction value calculation unit 103, is referred to.

Next, FIG. 5 is a schematic diagram for explaining the coefficient distribution operation in the image coding apparatus according to the present invention. As shown in FIG. 2, the distribution unit 105 includes a prediction value calculation unit 1
The output (prediction error value) from 03 is distributed to one of the two variable length coding units 107 and 110. Hereinafter, the coefficient distribution operation will be described in detail with reference to FIG.

Run length encoders 102, 104
Calculates and generates symbol data (L, M) in which the zero run length L and the coefficient value (non-zero coefficient amplitude) M are combined from the AC component distributed by the coefficient distribution unit 101. Next, the variable length coding units 107 and 110 are configured as shown in FIG.
As shown in FIG. 5, variable length encoded data for one block is calculated and generated from the prediction error value output from the prediction value calculation unit 103 and the symbol data generated by the coefficient distribution unit 101.

That is, the variable-length coding unit includes a DC coding means for coding the DC component of the transform coefficient, a plurality of AC coding means for coding the AC component of the transform coefficient, and a DC component of the transform coefficient. It functions as a block-encoded bit string concatenation unit that concatenates the encoded data of the AC components to generate a bit string for one block.

For example, the run length encoder 102
A bit string for one block is generated from the symbol data (symbols 1 to 4) output from the above and the prediction error value (symbol 0) output from the prediction value calculation unit 103.
The prediction error value and the symbol data input to the variable length coding units 107 and 110 must correspond to the same block.

More specifically, the coefficient distribution unit 1 will be described.
01, the DC component of the transform coefficient matrix is input to the prediction value calculation unit 103, and the AC component of the conversion coefficient matrix is input to the run length encoder 10.
2 or 104. Therefore, the distribution unit 10
5 distributes the prediction error value so that the prediction error value and the symbol data corresponding to one block are input to the same variable length coding unit. For example, the cycle t4 shown in FIG.
In the above, from the transform coefficient matrix of the block 1 input to the run length encoder 102, the run length encoder 102 outputs the symbol data AC1, AC2,
... is obtained. Then, the symbol data is input to the variable length coding unit 107.

Since the prediction error value DC at this time is calculated and generated from the transform coefficient matrix of block 1, the distribution unit 105 calculates and generates the prediction error value from the transform coefficient matrix of block 1. Symbol data AC1,
AC2, ... Are distributed to the variable length coding unit 107 to which the input is made.

Similarly, in cycle t5 shown in FIG. 4, the prediction error position DC and the symbol data AC1, AC2, ... Obtained from the transform coefficient matrix of the block 2 input to the run-length encoder 104 have variable lengths. It is input to the encoding unit 110.

As a method of controlling the distribution operation in the distribution unit 105, the coefficient distribution unit 101 transmits to the distribution unit 105 information indicating the variable length coding unit to which the DC component should be distributed together with the DC component. Various methods are possible.

In the variable length coding unit 107, the input 1
Encoding is performed using the DC encoding table 108 based on the prediction error value for the block, and encoding is performed using the AC encoding table 106 based on the input symbol data of the same block. . Similarly, in the variable length coding unit 110, coding is performed using the DC coding table 109 based on the input prediction error value for one block, and the input symbol block of the same block is input.
Encoding is performed using the AC encoding table 111 based on the data. That is, the image coding apparatus according to the present invention is characterized in that the run-length encoder and the variable-length coding unit are provided in a number equal to or more than the number of blocks into which the target image data is divided.

Then, the data concatenation unit 112 generates the coded bit string in block units output from the variable length coding units 107 and 110 as one coded bit string according to the block order input to the coefficient distribution unit 101. Concatenate all data to.

As described above, in order to increase the processing speed in the run-length encoder, the image coding apparatus according to the present invention has a hardware scale larger than the conventional one even when the processing units are parallelized. Can be reduced.
The reason is that, as described above, the predicted value calculation unit (DC amplitude predicted value calculator) 103 that processes only the DC component and the at least two run length encoders 102 and 104 that processes only the AC component are used. This is because the prediction value calculation unit 103 can be shared.

In the above-described embodiment, each RGB has an information amount of 8 bits, but in the image encoding process according to the present invention, image data having other ratios or information amounts are processed in the same manner. It is possible. Further, in the above-described embodiment, the number of pixels forming one block is 8 pixels × 8 pixels, but the block applicable to the present invention is not limited to this.

Further, in the above-mentioned embodiment, there are two run length encoders related to the AC component.
Even in an image coding apparatus including more run length encoders, the coding processing procedure can be performed in the same manner. Further, in the above-described embodiment,
Although one run length encoder is assigned to each block, the AC component in one block is divided into a plurality of parts and the run length
You may make it allocate an encoder and a variable length coding part.

That is, the present invention is an image coding apparatus for coding a transform coefficient matrix obtained by orthogonally transforming image data, wherein the transform coefficient matrix is divided into n (n: natural number) parts. And a coefficient distributing means for distributing to a predetermined coding means, at least n or more coding means for coding the divided transform coefficients to generate coded data, and the generated n coded data are connected. Block coding bit string concatenation means for generating a bit string.

Further, in the image coding apparatus according to the present invention, the coefficient distributing means converts the transform coefficient matrix into one DC component (n-
1) One AC coding component for coding the DC component of the transform coefficient, and at least (n-1) N coding for the AC component of the transform coefficient. AC encoding means is provided.

The image coding apparatus according to the present invention can be applied to a system including a plurality of devices (for example, a host computer, an interface device, a reader, a printer, etc.), but a device including one device ( For example, it may be applied to a copying machine, a facsimile machine, etc.).

Further, the control of the image coding apparatus having the above-described configuration supplies a recording medium (or a storage medium) recording a program code of software to a system or apparatus, and the computer of the system or apparatus. It is needless to say that (or CPU or MPU) can also be achieved by reading and executing the program code stored in the recording medium.

Furthermore, after the program code read from the recording medium is written in the memory provided in the function expansion card inserted in the computer or the function expansion unit connected to the computer, based on the instruction of the program code. , The CPU provided in the function expansion card or the function expansion unit performs some or all of the actual processing,
It goes without saying that the processing includes the case where the functions of the above-described embodiments are realized.

When the present invention is applied to the recording medium, the recording medium stores the program code corresponding to the above-mentioned flowchart.

[0045]

As described above, according to the present invention,
By sharing a part of the device for performing the entropy coding process, the hardware scale can be reduced.

[Brief description of drawings]

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

FIG. 2 is a schematic diagram for explaining an outline of a coefficient distribution operation using the image encoding device shown in FIG.

FIG. 3 is a coefficient distribution unit 10 of the image encoding device according to the present invention.
3 is a flowchart for explaining the operation procedure of No. 1;

FIG. 4 is a diagram for explaining an example of a signal waveform distributed by a coefficient distribution unit 101 in the image encoding device according to the present invention.

FIG. 5 is a schematic diagram for explaining a coefficient distribution operation in the image encoding device according to the present invention.

FIG. 6 is a block diagram showing a configuration of an image coding apparatus that performs variable length coding processing.

[Explanation of symbols]

101 coefficient distribution unit 102, 104 Run length encoder 103 Prediction value calculator 105 distributor 107, 110 Variable length coding unit 108, 109 DC encoding table 106, 111 AC encoding table 112 Data connection part

   ─────────────────────────────────────────────────── ─── Continued front page    F term (reference) 5C059 KK07 KK13 MA00 MA23 MC11                       MC33 MC34 MC38 ME08 PP01                       PP04 PP15 TA57 TB08 TC04                       TC06 TD07 TD15 UA02 UA05                 5C078 AA04 BA57 CA26 DA01                 5J064 AA03 BA08 BA09 BA16 BB03                       BB13 BC02 BC03 BC23 BD01

Claims (12)

[Claims] 1. An image coding apparatus for coding a series of transform coefficients obtained by orthogonally transforming image data, wherein the series of transform coefficients is divided into n parts and a predetermined coding means is provided. A coefficient distribution means for distributing the divided transform coefficients, at least n or more coding means for coding the series of divided transform coefficients to generate coded data, and a bit string connecting the generated n coded data And a block coded bit string concatenating means for generating 2. The coefficient distributing means divides the series of transform coefficients into one DC component and (n-1) AC components, and the encoding means comprises the DC component of the series of transform coefficients. 2. The image according to claim 1, further comprising: one DC encoding unit that encodes an AC component, and at least (n-1) AC encoding units that encode an AC component of the series of transform coefficients. Encoding device. 3. An image coding apparatus for coding a series of transform coefficients obtained by orthogonally transforming image data, comprising: a DC coding means for coding a DC component of the series of transform coefficients; A plurality of alternating-current encoding means for encoding alternating-current components of a series of conversion coefficients, coefficient distribution means for distributing alternating-current components of the series of conversion coefficients to predetermined alternating-current encoding means, and direct-current components of the series of conversion coefficients And a block coded bit string concatenation means for concatenating the coded data of the AC component to generate a bit string. 4. An image coding apparatus for coding a series of transform coefficients obtained by orthogonally transforming image data divided into predetermined blocks, wherein a DC component of a series of transform coefficients of each block is coded. A direct current encoding means for encoding, a plurality of alternating current encoding means for encoding the alternating current component of the series of transform coefficients in block units, and a predetermined alternating current encoding means for the alternating current component of the series of transform coefficients in block units. Coefficient distributing means for distributing, a plurality of block coded bit string concatenation means for concatenating encoded data of AC component and DC component of a series of transform coefficients of each block to generate a bit string, and the series of transform coefficients of the same block Distribution means for distributing the encoded data of the AC component and the DC component to a predetermined block encoded bit string connecting means, and the generated image data by connecting the generated bit string for each block And an image coding bit string concatenating means for generating a bit string of. 5. The image coding apparatus according to claim 4, wherein the AC coding means is provided in a number equal to or larger than the number of blocks into which the image data is divided. 6. The image coding apparatus according to claim 2, wherein the DC coding means predictively codes the DC component of the input series of transform coefficients. 7. A control method for an image coding apparatus, comprising: at least n or more coding means for coding a series of transform coefficients obtained by orthogonally transforming image data, wherein the series of transform coefficients are A coefficient distribution step of dividing into n parts and distributing to a predetermined encoding means, and a code for causing each encoding means to encode each of the divided series of transform coefficients to generate encoded data A method for controlling an image coding apparatus, comprising: a coding step; and a block coding bit string concatenating step of connecting the generated n pieces of coded data to generate a bit string. 8. The encoding means encodes a DC component of the series of transform coefficients, and at least (n) encodes an AC component of the series of transform coefficients.
-1) A method of controlling an image coding apparatus configured by AC coding means, wherein the coefficient distribution step includes one DC component and (n-1) AC components for the series of transform coefficients. And the DC component is distributed to the DC encoding means, and the (n-1) AC components are distributed to the (n-1) AC encoding means. Item 7. A method of controlling the image coding device according to Item 7. 9. A DC encoding means for encoding a DC component of a series of transform coefficients obtained by orthogonally transforming image data,
A method of controlling an image encoding device comprising a plurality of alternating-current encoding means for encoding alternating-current components of the series of conversion coefficients, wherein a direct-current component of the series of conversion coefficients is distributed to the direct-current encoding means, A coefficient distributing step of distributing an alternating current component to a predetermined alternating current encoding means; a direct current encoding step of causing the direct current encoding means to encode a direct current component of the series of transform coefficients to generate encoded data; For each AC coding means, an AC coding step of coding the AC component of the series of transform coefficients to generate coded data, and connecting the generated DC component and the coded data of the AC component to each other. And a block coding bit string concatenation step of generating a bit string, the method of controlling an image coding device. 10. A DC encoding means for encoding a DC component of a series of transform coefficients obtained by orthogonally transforming image data divided into predetermined blocks, and an AC component of the series of transform coefficients in block units. Image coding including a plurality of AC coding means for coding, and block coding bit string connecting means for connecting the coded data of the AC component and DC component of the series of transform coefficients in the same block to generate a bit string A method of controlling an apparatus, wherein the DC encoding means encodes a DC component of a series of conversion coefficients of each block to generate encoded data, and an AC of the series of conversion coefficients. A coefficient distributing step of distributing the components to predetermined AC encoding means in block units, and encoding the AC components of the series of transform coefficients in block units for each AC encoding means. An AC encoding step of generating encoded data by means of the above, and a distributing step of distributing the encoded data of the AC component and the DC component of the series of transform coefficients of the same block to a predetermined block encoded bit string concatenation means. The block coded bit string concatenation step of concatenating the AC component and the DC component encoded data of the series of conversion coefficients of each block to generate a bit string, and the generated bit string of each block are combined to obtain An image coding bit string concatenation step of generating a bit string. 11. The alternating current encoding means is provided in a number equal to or larger than the number of blocks into which the image data is divided, and the alternating current encoding step includes the series of conversions in block units for a predetermined alternating current encoding means. The control method of the image coding apparatus according to claim 10, wherein the AC component of the coefficient is coded to generate coded data. 12. The image coding apparatus according to claim 8, wherein the DC coding step predictively codes a DC component of a series of input transform coefficients. Control method.