JPH06165152A - Image compressing and coding device - Google Patents

Abstract

PURPOSE:To accurately control the quantity of coding in respect of the control of coding quantity for a VTR or the like at the time of compressing and coding digitized image data for record by controlling and recording the image data so that the coding quantity of compressed/coded data is equal to or less than a prescribed coding quantity. CONSTITUTION:When prescanning is executed by the 2nd quantizer 9 and a block coding quantity calculating circuit 10, an initial scale factor calculating circuit 11 calculates an initial scale factor. In addition, the coding quantity at the time of quantizing data based upon the initial scale factor in each group unit divided by the order of coded outputs is forecasted and the feedback control of the scale factor is executed based upon an error from the coding quantity of data practically coded by a Huffman encoder 6.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention controls the recording so that the code amount after compression encoding becomes less than a predetermined code amount when the digitized image data is compression encoded and recorded.
The present invention relates to an image compression encoding device that controls a code amount such as TR.

[0002]

2. Description of the Related Art As a highly efficient compression coding technique for natural images, a system combining orthogonal transformation and variable length coding is effective, and this system is also adopted as an international standard for color natural image coding systems. It has been decided (Journal of the Television Society: VOL.46 NO.8 PP1021 to 1024).

In this method, the amount of code changes for each image because of variable-length coding. However, in a system such as a VTR, since it is necessary to record at a constant rate, it is necessary to control the code amount for each image, and a code amount control method has been proposed (for example, 1992 Television Engineering Society of Japan). Next Meeting 16-15 "Code amount control in JPEG-DCT encoding").

FIG. 7 is a block diagram of a conventional image compression coding apparatus, FIG. 2 is a diagram showing a zigzag scan order of DCT, and FIG. 3 is a diagram showing a quantization step of each coefficient of DCT.
FIG. 5 is a graph for calculating a target scale factor from the scale factor and the code amount, and FIG. 8 is a diagram showing the relationship between the block and the scale factor.

In FIG. 7, 1 is an input terminal for inputting image data, 2 is a blocking circuit for dividing the image data into 8 × 8 blocks, and 3 is a discrete cosine transform (DCT) for each 8 × 8 block. DCT conversion circuit to perform, 4 DCT
Field memory for delaying the coefficient by one field time, 5
Is a first quantizer for quantizing DCT coefficients, 6 is a Huffman coding circuit for two-dimensional Huffman coding the output of the first quantizer 5, and 7 is a predetermined rate for the coded data. Buffer memory for buffering as described above, 8 as an output terminal, 9 as a second quantizer for quantizing DCT coefficients,
Reference numeral 10 is a block code amount calculation circuit for calculating the code amount per block obtained by two-dimensionally Huffman coding the output of the second quantizer 9, and 15 is the code amount by determining the scale factor of the first quantizer 5. It is a code amount control circuit for controlling.

The operation of the image compression coding apparatus configured as described above will be described. First, the data inputted from the input terminal 1 is converted into 8 × 8 D data by the blocking circuit 2.
It is divided into CT blocks. Data is DCT conversion circuit 3
Is converted into 64 DCT coefficients for each block and output in the zigzag scan order shown in FIG. 2 and delayed by one field in the field memory 4. The delayed DCT coefficient is linearized by the first quantizer 5 by a quantization value obtained by multiplying the quantization step different for each coefficient shown in FIG. 3 and the scale factor αi determined by the code amount control circuit 15. Variable-length data that is quantized and two-dimensionally Huffman coded by the Huffman coding circuit is output. The encoded variable-length data is converted into a predetermined rate in the buffer memory 7 and then output together with the scale factor αt.

The calculation of the scale factor αt is performed by the following prescan during the delay time.

For all blocks of one field, as shown in FIG. 8, there are M different scale factors α1 for each block.
.. .alpha.M (.alpha.1 <.alpha.2 <... <.alpha.M) and a quantization value obtained by multiplying the quantization step shown in FIG. 3 to perform linear quantization and variable when the block code amount calculation circuit 10 performs two-dimensional Huffman coding. The code amount of long data per block is calculated. Here, the code amount control circuit 15 calculates integrated values N1 to NM of the M code amounts for each scale factor. The relationship between the scale factors α1 to αM and the code amounts N1 to NM becomes a graph as shown in FIG. Here, if the target code amount per field is Nt, the optimum scale factor αt can be predicted as shown in FIG.

[0009]

However, in the above-described conventional configuration, the scale factor αt is predicted once per field, and if the prediction error is large, a predetermined code amount may overflow. In that case, for example, the VTR has a problem that the last image data cannot be recorded.

It is also possible to reduce the overflow probability by setting the scale factor αt to a value slightly larger than the calculated value at the time of prediction, but in this case, the code amount becomes smaller than necessary, so that the image quality deterioration due to compression is large. There is also a possible problem.

An object of the present invention is to solve the above-mentioned conventional problems and to provide an image compression coding apparatus which improves the accuracy of code amount control.

[0012]

In order to achieve this object, an image compression coding apparatus of the present invention delays the output data of the orthogonal transformer and the orthogonal transformer that orthogonally transforms the input image data block by block. Memory, a first quantizer for quantizing the output data of the memory, an encoder for variable-length encoding the output of the first quantizer, and the output data of the encoder After that, a buffer memory that outputs at a constant rate, a code amount integrator that integrates the code amount of the output data of the encoder, and output data of the orthogonal transformer of the encoder according to a predetermined rule. A second quantizer that classifies K groups corresponding to the output order and allocates one of M quantized coefficients and quantizes with the quantized coefficient, and an output of the second quantized variable When encoded A code amount calculator that calculates the code amount per lock, and predicts M code amounts when one screen is quantized by the M quantization coefficients from the block code amount output from the code amount calculator Then, an initial quantization coefficient detector that determines a quantization coefficient (initial quantization coefficient) for making one screen have a predetermined code amount, and for each group from the relationship between the M number of quantization coefficients and the code amount. A group code amount calculator that predicts a target code amount when quantized by the initial quantization coefficient, and an actual number accumulated by the code amount integrator each time data of each group is output from the encoder. The configuration includes a code amount controller that determines the quantization coefficient of the first quantizer by correcting the initial quantization coefficient with a prediction error between the code amount and the target code amount.

[0013]

With this configuration, the code amount prediction can be performed for each group divided in the order of data output, and the feedback control with the actual code amount can be performed. Therefore, the precision of the code amount control is improved, and the data overflow, Problems such as image quality deterioration can be improved.

[0014]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram of an image compression encoding apparatus according to an embodiment of the present invention, FIG. 4 is a diagram showing blocks on the screen and scale factors, and FIG. 5 is a graph showing scale factors and code amounts in one field. 6 is a graph showing the integrated value of the output code amount and the target code amount, and FIG. 9 is a graph showing the scale factor and the code amount in each group.

In FIG. 1, 1 is an input terminal for inputting image data, 2 is a blocking circuit for dividing the image data into 8 × 8 blocks, and 3 is an example of orthogonal transformation for each 8 × 8 block. DCT for cosine transform (DCT)
A transform circuit, 4 is a field memory for delaying the DCT coefficient by one field time, 5 is a first quantizer for quantizing the DCT coefficient, and 6 is a Huffman for two-dimensional Huffman coding the output of the first quantizer 5. An encoding circuit, 7 is a buffer memory for buffering encoded data at a predetermined rate, and 8 is an output terminal.

Reference numeral 9 is a second quantizer for quantizing the DCT coefficient, and 10 is a block code amount calculating circuit for calculating the code amount per block obtained by two-dimensionally Huffman coding the output of the second quantizer 9. Reference numeral 11 is an initial scale factor calculation circuit for calculating an initial scale factor from the block code amount, 12
Is a target code amount calculation circuit that predicts the target code amount for each group, 13 is a code amount control circuit that determines the scale factor of the first quantizer 5 and controls the code amount, and 14 is Huffman-coded data Is a code amount integrating circuit for integrating the code amount of.

The operation of the image compression coding apparatus configured as described above will be described below. First, the data input from the input terminal 1 is converted into 8 by the blocking circuit 2.
It is divided into × 8 DCT blocks. The data is converted into 64 DCT coefficients for each block by the DCT conversion circuit 3 and output in the zigzag scan order shown in FIG. 2, and is delayed by one field in the field memory 4. Delayed D
The CT coefficient is linearly quantized by the first quantizer 5 by a quantized value obtained by multiplying a different quantization step for each coefficient shown in FIG. 3 and a scale factor αt determined by the code amount control circuit 13. The Huffman encoding circuit 6 outputs two-dimensional Huffman-encoded variable-length data. The encoded variable-length data is converted into a predetermined rate in the buffer memory 7 and then output together with the scale factor αt.

Next, the calculation of the scale factor αt is shown below. First, the DCT coefficient for each block output from the DCT conversion circuit 3 is input to the second quantizer 9, and each block of one field is encoded data according to a predetermined rule as shown in FIG. Of K groups corresponding to the output order of M and scale factors α1 of M ways
~ ΑM (α1 <α2 <... <αM) is assigned, and the DCT coefficient of each block is linearly quantized by the assigned scale factor and a quantization value obtained by multiplying the quantization step shown in FIG. Be converted. The block code amount calculation circuit 10 calculates the code amount per block of the variable length data when the quantized DCT coefficient is two-dimensionally Huffman-coded using the M scale factors. The code amount is calculated by outputting the code amount from a combination of 0 run and non-zero coefficient using a ROM or the like and integrating the value for each block.

The initial scale factor calculation circuit 11 includes
The code amount for each block is sequentially input, and the integrated value of the code amount for M different scale factors is calculated for each group classification assigned in advance. Next, the relationship between the scale factor and the code amount for one field is derived from the relationship between the scale factor and the code amount for each group. The sum of the code amounts of the scale factor αi in the group j is N (i,
j), the code amount Ni when the data of one field is quantized by the scale factor αi is the sum of the code amounts quantized by the scale factor αi for each group, and the number M corresponding to the type of scale factor. It is predicted as the following (Equation 1).

[0020]

[Equation 1]

The relationship between the scale factor αi and the code amount Ni becomes a graph as shown in FIG. 5 in which Ni decreases as αi increases. Then, an initial scale factor α init for encoding the code amount of one field into a predetermined target code amount is calculated. Assuming that the target code amount of one field is Nt and the target code amount Nt is between the predicted code amounts N2 and N3 as shown in FIG. 5, a straight line (α2, N
2)-(α3, N3) can be interpolated to calculate the initial scale factor αinit for setting the code amount of one field to Nt as shown in (Equation 2).

[0022]

[Equation 2]

The group target code amount calculation circuit 12 is supplied with the sum N (i, j) of the code amounts of the scale factors αi in each group and the initial scale factor αinit, and is initialized for each of the K groups. The integrated value of the code amount when quantization is performed with the scale factor αinit is calculated. In the group j, the sum of the code amounts when the scale factor αi is predicted is M · N (i, j). Therefore, the relationship between the scale factor αi and the sum of the code amounts in the group j is shown in FIG. It is derived in the graph as shown. As shown in FIG. 9, the predicted code amount Nj in the group j when the scale factor is α init is predicted by the linear interpolation similar to the calculation of α init as shown in (Equation 3).

[0024]

[Equation 3]

The predictive code amount Nj for each group calculated by the group target code amount calculating circuit 12 is input to the code amount control circuit 13, and the target code amount integrated value for each group obtained by integrating the predictive code amount Nj is input. After being calculated, the data amount actually encoded by the Huffman encoding circuit 6 is integrated by the code amount integrating circuit 14 and input as a code amount integrated value. FIG. 6 shows the relationship between the target code amount integrated value and the code amount integrated value.
Is shown in the graph. The difference between the target code amount integrated value and the code amount integrated value is used as the prediction error Δ, and each time the data output of each group is completed, the scale factor αt of the next group is controlled by the prediction error Δ. For example, group (j-1)
Is completed, and the prediction error between the predicted code amount integrated value and the actual code amount integrated value in the group (j-1) is Δ (j-
1), the target code amount in group j is Nj−Δ
By outputting the scale factor .alpha.t for setting (j-1) to the first quantizer 5 and encoding by the Huffman encoding circuit 6, the prediction error .DELTA. (J-1) is canceled and the next group code is obtained. The code amount integrated value at the end of conversion can be corrected to a value near the target code amount integrated value.

From the relationship between the scale factor and the code amount for each block (FIG. 9), αt is calculated by calculating the scale factor αt (numerical expression) for making the sum of the code amounts of the groups Nj-Δ (j-1). It is calculated by linear interpolation shown in 4).

[0027]

[Equation 4]

When the prediction error Δ is 0, or in the first group 1 of the field, αt = α init is used for quantization.

As described above, according to the present embodiment, the DCT blocks of one field are classified into K groups, and the initial scale factor calculation circuit 10 is calculated from the block code amount quantized with M different scale factors. The initial scale factor α init for quantizing to the target code amount of 1 field by
Since the group target code amount calculating circuit 12 calculates the code amount for each group in the coding output order when quantized by init from the relationship between the block code amount and the scale factor, the initial scale factor and the group The accuracy of code amount prediction for each is improved.

Further, the code amount integrating circuit 14 is provided for each group.
Since the code amount control circuit 13 controls the code amount on the basis of the difference between the actual code amount and the target code amount accumulated in step S1, it is possible to perform fine code amount control M times during one field period.

In the present embodiment, the compression is performed in field units, but it is clear that the same effect can be obtained in compression in frame units or in macroblock units composed of a plurality of blocks.

Further, in this embodiment, the prediction of the code amount by the prescan is performed in all the blocks, but the same effect can be obtained when the sampling is performed in a part of the blocks.

Further, although the orthogonal transform by the DCT transform is shown in this embodiment, it is obvious that the same effect can be obtained in any orthogonal transform such as LOT and Hadamard transform.

[0034]

As described above, according to the present invention, 1
Since the initial quantized coefficient and the integrated value of the code amount for each group obtained by dividing the data of one screen into K groups can be accurately predicted only by performing the pre-scan once, By compensating the initial scale factor by comparing the above, the accuracy of the code amount control is improved, and the image compression encoding device that prevents the data overflow and improves the image quality of the compressed image without increasing the time required for the prediction calculation is provided. realizable.

[Brief description of drawings]

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

FIG. 2 is a diagram showing a zigzag scan of DCT coefficients.

FIG. 3 is a diagram showing quantization steps of DCT coefficients.

FIG. 4 is a diagram showing blocks on a screen according to an embodiment of the present invention.

FIG. 5 is a diagram showing a relationship between a scale factor and a code amount in one field.

FIG. 6 is a graph showing a time history of code amount integrated values.

FIG. 7 is a block diagram showing a configuration of a conventional image compression encoding device.

FIG. 8 is a diagram showing blocks on a screen of a conventional example.

FIG. 9 is a diagram showing a relationship between a scale factor and a code amount in one group.

[Explanation of symbols]

 2 block forming circuit 3 DCT converting circuit 4 field memory 5 first quantizer 6 Huffman encoder 7 buffer memory 9 second quantizer 10 block code amount calculating circuit 11 initial scale factor calculating circuit 12 group target code amount Calculation circuit 13 Code amount control circuit 14 Code amount integration circuit

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁵ Identification code Office reference number FI technical display location H04N 5/92 H 4227-5C

Claims (2)

[Claims] 1. Input image data is divided into a plurality of blocks of 1 block p.times.p pixels, and each divided block is compressed to a predetermined code amount by a combination of orthogonal transformation and variable length coding. An apparatus, which is an orthogonal transformer for orthogonally transforming the input image data block by block, a memory for delaying output data of the orthogonal transformer, and a first quantizer for quantizing output data of the memory. An encoder for variable-length encoding the output of the first quantizer, a buffer memory for outputting the output data of the encoder at a constant rate after writing the output data of the encoder, and an output data of the encoder A code amount accumulator for accumulating the code amount, classification of K groups corresponding to the output order of the encoder according to a predetermined rule for the output data of the orthogonal transformer, and M number of quantizers Second quantizer for quantizing one of the two and quantizing with the quantized coefficient, and a code amount calculator for calculating the code amount per block when the output of the second quantizer is variable-length coded And to predict M code amounts when one screen is quantized by the M quantization coefficients from the block code amount output from the code amount calculator, and set one screen to a predetermined code amount. Quantized coefficient (initial quantized coefficient)
And a group code amount calculator that predicts a target code amount when quantized by the initial quantization coefficient for each group from the relationship between the M kinds of quantization coefficients and the code amount. And by correcting the initial quantization coefficient with a prediction error between the actual code amount integrated by the code amount integrator and the target code amount each time data of each group is output from the encoder. An image compression coding apparatus comprising a code amount controller for determining a quantization coefficient of the first quantizer. 2. The code amount controller subtracts the prediction error between the actual code amount and the target code amount accumulated by the code amount integrator for each of the K groups from the target code amount of the next group. The image compression coding apparatus according to claim 1, wherein the quantization coefficient of the next group is determined by using as a target code amount of the next group.