JPH09149260A - Information processor and information processing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To suppress the deterioration of the quality of encoding data and to encode data with a target compression rate by controlling encoding object data whether it is reversibly encoded or non-reversibly encoded in accordance with the quantity of generated codes and encoding object data based on a controlled result. SOLUTION: Data length on encoding information which is encoded by an encoding part 15 is inputted to a generated code quantity monitor part 14. The generated code quantity monitor part 14 outputs a flag for controlling a gradation conversion part 13 so that the compression rate becomes the target one which is previously set based on inputted data length. The flag is stored in a flag storage memory part 26 and is inputted to a gradation switch indication part 20 on a decoding side. The encoding method of reverse encoding and non-reverse encoding is switched by controlling the gradation of a picture being the encoding object. A gradation conversion part 21 gradation-converts respective pieces of picture information inputted based on an instruction from the gradation switch indication part 20 and outputs converted information to an interface part 22.

Description

Detailed Description of the Invention

[0001]

TECHNICAL FIELD The present invention relates to an information processing apparatus and method for processing information.

[0002]

2. Description of the Related Art As a well-known digital black-and-white binary image coding method, there is a method of coding an image in raster units. For example, M used in FAX
The H, MR, MMR method and the like are variable-length coding methods that perform coding by paying attention to the length (run length) of continuous black and white pixels.

In addition to the variable length coding system, a fixed length coding system is known. This method includes vector quantization in which an image is divided into blocks and a fixed-length code is assigned to each block, and encoding can be performed at the same compression rate. However, there is a difference in the quality of the encoded data information for each block, and there is a difference in the quality (image quality) of the decoded data.

The coding system used in image communication devices such as facsimiles is a reversible coding system in which the original image can be completely restored on the decoding side. Lossy coding schemes that can increase the are also used.

[0005]

As described above, in the variable-length coding system, the generated code amount differs depending on the image, so that it is possible to control the compression ratio and perform the coding at the target compression ratio (constant memory capacity). There was a problem that it was difficult.

On the other hand, in the fixed length coding method, the compression rate can be suppressed to the target value, but there is a problem that the deterioration of the information of the coded data is conspicuous locally.

In order to solve the above problems, the present invention is
It is an object to suppress deterioration of the quality of encoded data and to encode the data at a target compression rate.

Further, it is preferable to suppress deterioration of the image quality due to the encoding in encoding the image data and to encode the data at a target compression rate.

Another object is to suppress deterioration of a visual image when encoding color image data and to encode the data at a target compression rate.

[0010]

In order to solve the above-mentioned problems, according to an information processing apparatus according to claim 1 of the present invention, in an information processing apparatus for sequentially encoding data in a predetermined unit, encoding is performed. Monitoring means for monitoring the amount of generated code when the already-encoded data is encoded, and control means for controlling whether the data to be encoded is lossless-encoded or lossy-encoded according to the amount of generated code And encoding means for encoding the encoding target data based on the result controlled by the control means.

[0011]

BEST MODE FOR CARRYING OUT THE INVENTION

(First Embodiment) FIG. 1 is a block diagram showing the configuration of the entire information processing apparatus for explaining the first embodiment of the present invention.

Reference numeral 10 is a dither image memory, 11 is a coding side prediction unit, 12 is a coding side prediction judgment unit, 13 is a gradation conversion unit, 14 is a generated code amount monitoring unit, 15 is a coding unit, 16
Is compression memory.

Reference numeral 17 is a decoding unit, 18 is a decoding side prediction unit, 19 is a decoding side prediction determination unit, 20 is a gradation switching instruction unit, 21 is a gradation conversion unit, 22 is an interface unit, 2
6 is a flag storage memory unit.

The dither image memory 10 stores image information converted into 2 bits / pixel by a multi-valued dither circuit as shown in FIG.

FIG. 2 is a block diagram of a four-value dither circuit. First, the image information is output from the image memory 23 in which 8-bit image information is stored to the 4-value dither circuit 24,
The 4-value dither circuit 24 performs 4-value dither processing using a threshold value matrix (see FIG. 3) described later, and then stores the dither image memory 10. The four-valued dither circuit is a simple circuit composed of a comparator or the like that compares a threshold value with image information, and therefore detailed description thereof is omitted.

In the gradation conversion unit 13, the dither image information of 2 bits / pixel is sequentially converted into 1 bit / pixel of dither image information for each pixel of the image information, and the 1-bit information 21.
Output as 0. The simplest method can be implemented by taking the two most significant bits (MSB). Details of the conversion process of this dither image information will be described later.

In the prediction unit 11, whether the 1-bit image information of the pixel to be encoded is 0 based on the 1-bit information of the surrounding pixels encoded before the pixel to be encoded in the input information 210. 1 is performed, and the predicted image information (predicted symbol) 203 and the index 2 to be described later are executed.
12 is output. Also, update information 213, which will be described later, is input from the encoding unit 15.

In the prediction judgment unit 12, the prediction symbol 203
And the value (0/1) of the image information of the pixel to be encoded input as the information 210 are determined, and the information 202 indicating the determination result is encoded by the encoding unit 15.

The compression memory 16 stores the coded information coded by the encoder 15. The decoding operation starts after the image information encoding process described above is completely completed.

In the decoding unit 17, the above-mentioned prediction judgment unit 12 is used.
Information of the judgment result of the prediction judged in (corresponding to information 202)
Is decoded and output as information 207. Prediction unit 18
Has the same configuration as the prediction unit 11, and is a prediction symbol 206 that predicts the value (0/1) of the image information of the pixel to be encoded based on the decoded 1-bit image information of the surrounding pixels.
(Prediction symbol 203 and prediction symbol 206 are equal)
Is output to the prediction determination unit 19.

The prediction determination unit 19 outputs the value (0/1) of the image information of the actual pixel to be coded by referring to the information 207 and the prediction symbol 206. Specifically, if the predicted symbol 206 is 1, the information 2
When 07 is the information indicating "not matching the prediction symbol", the value of the image information of the pixel to be encoded is 0, and when it is the information indicating "matching the prediction symbol". Outputs 1. When the predicted symbol is 0, 1 or 0 is output by the same judgment.

The generated code amount monitoring unit 14 includes an encoding unit 15
The data length of the coded information coded in is input. The generated code amount monitoring unit 14 controls the gradation conversion unit 13 based on the input data length so as to achieve a preset target compression ratio (2 bits / pixel or 1 bit / pixel). Output the flag to do. This flag is also output and stored in the flag storage memory unit 26. The information stored in the flag storage memory 26 is input to the gradation switching instruction unit 20 on the decoding side.

The gradation switching instructing section 20 determines whether the value of the image information input to the gradation converting section 21 is gradation converted by the gradation converting section 13 based on the input information. When gradation conversion is performed, the gradation conversion unit 21 outputs a command for gradation conversion.

The gradation conversion unit 21 converts the gradation of each of the image information input based on the command from the gradation switching instruction unit 20, and outputs the converted image information to the interface unit 22. This image information is output from the interface unit 22 to an external device such as a printer.

FIG. 3 shows an example of a threshold value matrix (matrix size is 2 × 2 pixels) for performing four-value dither processing. The multi-level dither circuit 24 performs 4-level dithering on the 8-bit input image information (0 to 255) based on this threshold value.
Specifically, the following processing is performed. First, the pixel value of the image information of 2 × 2 pixels is compared with the first threshold value,
Pixels below the first threshold are determined as level 0.

Next, the above-mentioned image information is compared with the second threshold value, and the pixels below the second threshold value (only the pixels whose level has not been determined) are determined to be level 1.

Next, the above-mentioned image information is compared with the third threshold value, and the pixels below the third threshold value (only the pixels whose level has not been determined) are determined to be level 2.

Next, the pixels whose image information is equal to or larger than the third threshold value (only the pixels whose level has not been determined) are determined as level 3.

By the above processing, one pixel has 2 bits (4
Value).

In the present invention, the image is divided into bands, and the process of switching the coding method of the next band is performed according to the amount of codes generated up to the preceding band.

The switching of the encoding method is realized by controlling the gradation of the image to be encoded. That is, assuming that the initial image information is 4-valued (2 bits / pixel), lossless encoding is performed when the 4-valued dither image is encoded as it is, and the 4-valued dither image is gradation-converted into a 2-valued dither image. If it is encoded later, it will be lossy.

FIG. 4 is a diagram for explaining the above-described image band division method. Here, an example of dividing an image into six bands is shown. In the flag area of each band, whether the 4-valued dither image is encoded (encoding flag 0) or whether the 4-valued dither image is converted into the binary dither image and then encoded (encoding flag 1) is indicated. The flag (corresponding to the flag stored in the flag storage memory unit 26) is stored.

At the time of decoding on the decoding side, this flag is used to determine the type of dither image in each band (corresponding to the gradation switching instruction section 20 described above).

FIG. 5 is a diagram for explaining the target value of the code generation amount for each band and the timing of flag switching.

In the figure, the diagonal line extending from the origin to the final target value (the code amount calculated from the required compression rate) is the target value of the total code amount at the time when the coding of each band is completed,
The range of the solid line shown perpendicular to the band number axis is the allowable range of the total code amount at the time when the coding of each band is completed. The broken line indicated by the broken line is data indicating the generated code amount.

The above-mentioned allowable range is set, for example, within the range of -10% to + 10% of the generated code amount of each band.

Next, a method for controlling the generated code amount when the image information for one screen is sequentially encoded for each band will be described.

First, since the code amount at the time of encoding band 0 as 2 bits / pixel is within the allowable range, the next band is encoded with the flag set to the initial value (0).

Next, band 1 is encoded (encoded as 2 bits / pixel) based on the flag value (0). At this time, the total code amount of the encoded band 0 to band 1 is calculated, and it is determined whether or not the total code amount is within the allowable range (corresponding to the generated code amount monitoring unit 14). In the case of FIG.
Since it is over the allowable range, the flag is changed from 0 to 1. As a result, 1 bit / pixel is coded in band 2 and thereafter, so that the generated code amount can be reduced.

Next, when the band 2 is coded, the total code amount of the band 0 to the band 3 is calculated. As a result, since it is within the allowable range, the band 3 is coded while the flag remains 1.

Next, as a result of calculating the total code amount from band 0 to band 3 at the time of encoding band 3, the total code amount is below the allowable range, so the flag is changed from 1 to 0. . As a result, since 2 bits / pixel are coded in band 4 and thereafter, the generated code amount can be increased.

Next, as a result of encoding the band 4, since the total code amount of the bands 0 to 4 exceeds the allowable range, the flag is changed from 1 to 0 and the encoding of the band 5 is performed. .

By performing the above processing, it is possible to control the code amount so that the total code amount of the image information of one screen approaches the final target value.

Further, in the present embodiment, the flag may remain 0 even if the total code amount remains below the allowable range while continuing the lossless encoding (the flag is 0), and this state continues. In this case, the code amount can be suppressed to be smaller than the target code amount.

FIG. 6 is a diagram showing how the image information used for encoding is switched by switching the flag.

(A) is an area to be coded as a four-valued dither image, in which each pixel has 2-bit information (most significant bit MS).
B and the least significant bit LSB) are encoded in order from MSB to LSB.

(B) is an area to be coded as a binary dither image. When the coded pixel changes from the area (A) to the area (B) (the flag changes from 0 to 1), the 4-valued dither image of the area (B) is changed to the binary dither image (M in the present embodiment).
By taking 1 bit of SB, it is converted to binary dither image information) and encoded. When the coded pixel changes to the area (C) (the flag changes from 1 to 0),
The area (C) is encoded in the same manner as the area (A).

FIG. 7 is a gradation conversion unit 13 which selectively outputs a four-value dither image and a two-value dither image in order to perform the processing of FIG.
2 is an example of a block diagram of FIG.

In the figure, reference numeral 80 denotes a parallel / serial conversion unit which converts information of a 4-valued dither image (2-bit parallel signal) into a 1-bit serial signal.
It is output to the 4-value / binary conversion unit 81 and the selector 82 as 4-valued image information.

Reference numeral 81 denotes a four-value / binary conversion unit for converting four-value image information into binary image information, which is output to the selector 82 as binary image information. In this embodiment, a method of taking only the MSB of 4-valued (2-bit) information is used.

The selector 82 includes the 4-value image information input from the parallel / serial converter and the 4-value / binary converter 8.
The binary image information input from 1 is selectively switched according to the flag (1/0) output from the generated code amount monitoring unit 14 and output.

FIG. 8 is a block diagram of the prediction unit 11.

The coded image information 210 is sequentially input to the template generator 50 and stored for a predetermined number of lines. Also, based on the stored image information, the data 211 indicating the state of the image information of the surrounding pixels necessary for predicting the image information of the pixel to be encoded is output to the prediction state memory 51.

The prediction state memory 51 stores a probability estimation table used for arithmetic coding, which is a known technique. In addition, index information of the probability estimation table is stored for each state of image information of pixels around the pixel to be encoded.

This predictive state memory 51 receives the data 2 when the data 211 is input from the template generator 50.
The index information corresponding to 11 is taken out, a prediction symbol 203 corresponding to this index information is generated from the probability estimation table, and this is output to the prediction determination unit 12.
At the same time, the index information 212 used at this time is output to the encoding unit 15.

By inputting the index information 213 which is updated after being encoded using this index information, it is overwritten and stored as the updated index information. This is based on the learning of the predictive coding method performed every time the coding is performed by the coding unit 15, and the predicted symbol predicted when the state of the image information of the same surrounding pixel is next is the actual coding. There is a process of optimizing so as to increase the probability of becoming the same as the target pixel, and it is overwritten and stored for each state of image information of surrounding pixels.

FIG. 9 is a diagram for explaining in detail the process of using the predicted reference pixel (peripheral pixel) inside the template generation unit 50. Image information of surrounding pixels c, d, e, f around the pixel x to be encoded and pixels a, b, g, h, i corresponding to the dither period (2 × 2 in the present embodiment), The data is combined by the combiner 90 and output to the prediction state memory 51 as the above-mentioned data 211.

FIG. 10 is a block diagram of the encoding unit 15, in which the index information 212 is input to the arithmetic encoding unit 102. In addition, the arithmetic coding unit 102 includes the arithmetic coding unit 10
The index information 213 updated after being encoded by 2
Is output to the prediction unit 11.

FIG. 11 is a block diagram of the decoding unit 17, in which the index information 208 is input to the arithmetic decoding unit 104. In the arithmetic decoding unit 104, the arithmetic decoding unit 1
Index information 20 updated after being decrypted in 04
9 is output to the prediction unit 18.

As the arithmetic coding, it is conceivable to use QM-Coder or the like used in the ISO international standard coding method.

According to this embodiment, it is possible to suppress the deterioration of the image quality due to the encoding and to encode the data at the target compression rate.

(Second Embodiment) In the second embodiment, in the DPCM coding (differential pulse coding) of a multivalued image, the generated code amount is controlled by switching between lossless coding and lossy coding. Is the way to do it.

In order to encode with the reversible DPCM, the difference data is directly Huffman-encoded. Further, in order to encode with the irreversible DPCM, the differential data is quantized and then Huffman-encoded.

FIG. 12 is a block diagram of the entire system.

In FIG. 12, 30 is an 8-bit image memory, 31 is a subtractor, 32 is a quantizer, 33 is a 1-pixel delay, 34 is an inverse quantizer, 35 is a variable length encoder, 36.
Is a compression memory, 37 is a variable-length code decoder, 38 is an inverse quantizer, 39 is an adder, and 40 is a one-pixel delay.

The image information of the pixel to be coded input from the image memory 30 is subtracted from the image information output of the previous pixel by the subtractor 31, and the difference data is output. This difference data is quantized by the quantizer 32. The quantized difference data and the information on the quantization switching control, which will be described later, are input to the dequantizer 34 and used for local decoding.

The quantized difference data is the variable length encoder 3
5, coding such as Huffman coding is performed and stored in the compression memory 36. At the same time, the quantized difference data is dequantized and locally decoded by the local decoding unit 34. The locally-decoded image information is delayed by one pixel in the delay circuit 33 and output to the subtractor 31 to obtain the difference from the next pixel.

The code amount of the encoded data encoded by the variable length encoder 35 is input to the generated code amount monitoring unit 14.
The variable length encoder 35 controls the quantization method by the quantizer according to the difference between the input code amount and the expected code amount.

Switching between lossless encoding and lossy encoding is
This is done by switching the parameters given to the quantizer.

That is, if the quantized coefficient is set to 1, information is completely left. On the other hand, if the number of quantized values exceeds 1, irreversible quantization is performed.

By performing the above quantization, the histogram of the difference data is concentrated near 0, so that the amount of generated codes can be suppressed by performing Huffman coding based on this histogram.

The switching flag indicating the timing for switching the above quantization is stored in the flag memory 26 and used at the time of decoding.

The output of the compression memory 36 is the variable length decoder 3
7, and the inverse quantizer 38 performs inverse quantization processing.
As a result, difference data is output, and the adder 39 adds the difference data and the decoded image information of the previous pixel to reproduce the image information. This image information is I / F2
It is output through 2. Also, the reproduced image information is
It is also input to the delay 40 and used for addition with the next difference data.

In the above-described inverse quantization of the inverse quantizer 38, the flag storage memory 26 outputs a switching signal for switching the quantization parameter at the time of inverse quantization.

As in the first embodiment, the control method of the generated code amount is performed according to FIG. 5, and when the code amount exceeds the allowable range, switching is performed so as to perform lossy encoding, If the amount falls below the permissible range, switching is performed so that lossless encoding is performed. By the above processing, it is possible to perform control so that the generated code amount of the final target value is close to the generated code amount.

According to this embodiment, lossless encoding and lossy encoding are switched based on the amount of generated code before the data to be encoded, thereby suppressing deterioration of image quality due to encoding and achieving a target compression rate. The data can be encoded with.

In the present embodiment, reversible and irreversible coding is switched by switching the quantization parameter of DPCM coding. However, the present invention is not limited to this, and MH, MR coding or the like is used for reversible coding. , JPEG for lossy encoding
Encoding or the like may be used. That is, the lossless encoding may be any method capable of completely reproducing the data, and the lossy encoding may be any method capable of efficiently encoding the data.

(Third Embodiment) FIG. 13 is a block diagram of the entire apparatus showing a third embodiment.

Reference numeral 23 is an image memory (8 bits) for storing 8-bit image data for one pixel, 24 is a 4-value dither processing unit, 25 is a 4-value dither image memory, 26 is a binary dither processing unit, and 27 is a Binary dither image memory, 28 is a selector for selecting either 4-value dither image data or binary dither image data, 29 is a compression unit, 16 is compression memory 15, flag information storage unit, 14 is a decoding unit, 13 Is 4
It is a mixed memory for value and binary dither image data (2-bit image data).

The 8-bit image data stored in the image memory 23 is input to the 4-valued dither section 24 and is converted into 4-valued data, and then stored in the 4-valued dither memory 25. Similarly 8
The bit image data is input to the binary dither unit 26, binarized, and then stored in the binary dither memory 27.
(The four-valued dither threshold of FIG. 3 and the two-valued dither threshold of FIG. 14 are used.)

In the selector 28, 4-valued dither image data and 2
The value dither image data is switched and output to the compression unit 29.

The output data is encoded by the compression unit 29 and stored in the compression memory 16.

The switching flag information is output to the selector 28 and also to the storage unit 31 for storage. This flag is referred to in the decoding performed by the decoding unit 32.

The decoding unit 32 reads the encoded data in the compression memory and stores whether each bit of the decoded data is the MSB, the LSB, or the value of the binary dither image data of the 4-level dither image data. The determination is made based on the switching flag information from the device 31, and the result is stored in the mixed dither image memory 33 having a depth of 2 bits.

From the mixed dither memory 33, the decoded image data is output to an external device such as a printer through the I / F circuit 22.

FIG. 15 is a block diagram showing the internal structure of the compression unit 29.

Reference numeral 11 is a predictor, 12 is a prediction decision unit, 15 is an encoder, and 14 is a generated code amount monitoring unit.

The dither image data is input to the predictor 11 and the prediction determiner 12. In the predictor 11, for each bit plane, 0/1 of the bit of the pixel to be encoded is calculated from the surrounding pixels.
Is predicted.

The prediction decision unit 12 decides whether or not the predicted value and the actual bit value (0/1) match, and the encoder 15 encodes the result.

The compression memory 16 stores the encoded data encoded by the encoder 15.

The code data length is input to the generated code amount monitoring unit 14, and the dither switching flag is controlled so that the target compression rate is achieved.

FIG. 16 is a block diagram of the decoding unit 32.

Reference numeral 17 is a decoder, 18 is a predictor, 19 is a prediction decision device, and 21 is a gradation corrector. After the encoding of all the dither image data is completed, the following decoding operation starts.

The decoder 17 decodes the prediction match / mismatch result. The predictor 18 has the same configuration as the predictor 11, predicts the 0/1 value of the pixel to be encoded from the surrounding pixels, and the prediction determiner 19 restores the value to the actual 0/1 value.

The flag decoder 20 uses a flag from the flag memory to determine whether gradation conversion has been performed, and issues a switching instruction for 4-value dither processing or 2-value dither processing. The tone corrector 21 performs tone correction on the dither image data corresponding to 2 bits. Here, the predictor 18 has the same configuration as the predictor 11.

The configuration of the four-valued dither processing section is to use the configuration of FIG. 2, and the above-mentioned processing is performed.

FIG. 17 is an example of a 13-level dither pattern when it is quaternarized. Each pixel is represented by four levels of white, light gray, dark gray, and black, and 13 levels of halftone expression are possible by the area gradation of dither.

According to this embodiment, it is possible to suppress the deterioration of the image quality due to the coding and code the data at the target compression rate.

(Fourth Embodiment) A drawback of band processing is that the band processing breaks out when the processing frequently changes for each hand.

Color image (YMC image, Y: yellow,
When performing band processing on (M: magenta, C: cyan), it is effective to preferentially convert four values into a binary value for a color (for example, yellow) whose processing difference is not noticeable for each band.

FIG. 18 is an overall configuration diagram for explaining this.
Shown in The encoding process is performed in the order of Y band, M band, C band, Y band, and M band in FIG.
Similarly to the above, encoding is performed by switching between 4-value and 2-value dither processing. RG different from FIG. 13 including the above switching processing
The B switching control and the like are performed by the RGB switching control unit 34, and the RGB switching control unit 34 controls the R of the selector 28.
It also controls the switching for each GB and controls the switching process that accompanies the accumulation of the generated code amount of the RGB signals in band units.

FIG. 19 shows an example in which a switching flag is set for each color of each band. There are four types of combinations from mode 0 to mode 3.

Mode 0 shows the case where each band of YMC performs 4-level dither compression.

Mode 1 shows a case where Y performs binary dither compression and M and C perform quaternary dither compression.

In mode 2, YC is binary dither compression and M is 4
The case where value dither compression is performed is shown.

Mode 3 shows a case where each YMC band performs binary dither compression.

These are switched based on the comparison result of the target code amount and the generated code amount.

For example, when the generated code amount is considerably smaller than the target code amount, mode 0 compression is performed, and as it becomes larger, mode 1, mode 2 and mode 3
Control to use.

An example of specific mode switching will be shown. FIG. 20 is a mode switching table. There are four cases in which the difference between the target code amount and the generated code amount is within the permissible range or outside the permissible range, and the modes are switched according to the corresponding tables.

When the generated code amount is smaller than the allowable range, 0 is set if the current coding mode is 0, 0 if the mode is 1, 0 if the mode is 2, and 3 if the mode is 3. In this case, the compression rate is significantly reduced by switching to 1.

When the generated code amount is smaller than the target code amount and is within the allowable range, 0 when the current coding mode is 0, 0 when the mode is 1 and 2 when the mode is 2. When the mode is 1 and the mode is 3, the compression rate is lowered to some extent by switching to the mode 2.

When the generated code amount is larger than the target code amount and is within the allowable range, it is 1 when the current coding mode is 0, 2 when the mode is 1, and 2 when the mode is 2. 3. If the mode is 3, the compression rate is increased to some extent by switching to 3.

When the generated code amount is larger than the allowable range, the current coding mode is 0, the mode is 2, the mode is 3, the mode is 2, the mode is 3, and the mode is 3. In this case, the compression rate is greatly increased by switching to 3.

By the mode switching processing as described above, the code amount when the color image is compressed can be controlled.

Further, by encoding a color component signal of low visual importance with a high compression rate, the target code amount can be finally suppressed.

In the above embodiments, the color component signal may be Lab or YIQ, in which case the compression rate of the chromaticity signal, which is a color component signal of low visual importance, is controlled. , It is possible to control the code amount while suppressing the deterioration of the image visually.

(Fifth Embodiment) By applying the above embodiments, a four-value image (four-value dither image) and a two-value image (two-value image)
By changing the value dither image) on a pixel-by-pixel basis, switching noise can be made inconspicuous.

The four-valued dither threshold value shown in FIG. 21 is a combination different from the combination of the above-mentioned threshold values, and the image pattern resulting therefrom is as shown in FIG.

FIG. 23 shows an example in which four 2 × 2 dither images are collected to form a 4 × 4 pixel group, and the setting of the pixel position treated as a binary image is changed to create modes 0 to 3. Pixels with circles in the figure show pixel positions in which the binary image is replaced in the 4-value dither image. These are, according to the planned code amount and the generated code amount as described above,
Switch the mode in band units.

Further, in the second to fifth embodiments, the generated code amount may be equal to or less than the target value as in the first embodiment, and the flag at that time is not changed. As a result, the final code amount can be suppressed below the target value.

In the present embodiment, since the code amount is controlled by changing the image quality only for the minimum area in the band, the generated code amount can be easily controlled.

As described above, in the first to fifth embodiments, the encoding method of the data to be encoded is controlled based on the generated code amount of the encoded data, so that the data can be encoded in real time.

[0123]

As described above, according to the present invention, it is possible to suppress the deterioration of the quality of encoded data and encode the data at a target compression rate.

Further, in the encoding of image data, it is possible to suppress the deterioration of the image quality due to the encoding and to encode the data at the target compression rate.

Further, when encoding color image data, it is possible to suppress deterioration of a visual image and to encode the data at a target compression rate.

[Brief description of the drawings]

FIG. 1 is a block diagram of a first embodiment.

FIG. 2 is a block diagram of a four-value dither circuit.

FIG. 3 is a four-value dither threshold matrix

FIG. 4 is an explanatory diagram of band division.

FIG. 5 is a diagram illustrating a process of switching an encoding method according to a generated code amount.

FIG. 6 is an explanatory diagram of encoding processing when a 4-valued dither image and a 2-valued dither image are mixed.

FIG. 7 is a block diagram of a gradation conversion circuit.

FIG. 8 is a block diagram of a prediction unit 11.

FIG. 9 is an explanatory diagram of reference pixel positions of a dither image.

FIG. 10 is a block diagram of an encoding unit 15.

FIG. 11 is a block diagram of a decoding unit 17.

FIG. 12 is an overall block diagram for explaining a second embodiment.

FIG. 13 is an overall block diagram for explaining a third embodiment.

FIG. 14 is a diagram showing a binary dither threshold used in the third embodiment.

FIG. 15 is a block diagram showing an internal configuration of a compression unit 29.

FIG. 16 is a block diagram of a decoding unit 32.

FIG. 17 is a diagram showing an example of a 13-level dither pattern when quaternarized.

FIG. 18 is an overall configuration diagram for explaining a fourth embodiment.

FIG. 19 is a diagram showing an example in which a switching flag is added for each color of each band.

FIG. 20 is a diagram showing a coding mode switching table.

FIG. 21 is a four-valued dither threshold matrix 2

FIG. 22 is a diagram showing an example of a 13-level dither pattern when quaternarized.

FIG. 23 is a diagram illustrating a method of setting a pixel position treated as binary image data.

[Description of Codes] 10 dither image memory 11 prediction unit 12 prediction determination unit 13 gradation conversion unit 14 generated code amount monitoring unit 15 encoding unit 16 compression memory 17 decoding unit 18 decoding side prediction unit 19 prediction determination unit 20 Gradation switching instruction section 21 Gradation conversion section 22 Interface section 26 Flag storage memory section

Claims (18)

[Claims] 1. An information processing apparatus for sequentially encoding data in units of a predetermined unit, monitoring means for monitoring the generated code amount of encoded data, and reversible encoding target data according to the generated code amount. Information processing characterized by having control means for controlling encoding or lossy encoding, and encoding means for encoding the encoding target data based on the result controlled by the control means. apparatus. 2. The information processing according to claim 1, wherein the monitoring of the generated code amount by the monitoring unit is performed for each band generated by dividing the encoded image data into a plurality of pieces. apparatus. 3. The control by the control means is switched to the lossy encoding means when the generated code amount is above the allowable range and switched to the lossless encoding means when the generated code amount is below the allowable range. The information processing device according to claim 1. 4. The information processing apparatus according to claim 1, wherein the lossless encoding and the lossy encoding are DPCM encoding of a multi-valued image. 5. In the encoding by the encoding means,
The information processing apparatus according to claim 4, wherein the quantization parameter is switched and used based on a result controlled by the control means. 6. The lossless encoding is encoding of N-valued image data, and the lossy encoding is encoding of M-valued image data obtained by converting the N-valued image data. The information processing device according to item 1. 7. An information processing method for sequentially encoding data in predetermined units, a monitoring step of monitoring a generated code amount of encoded data, and reversible encoding target data according to the generated code amount. Information processing characterized by having a control step of controlling whether to perform encoding or lossy encoding, and an encoding step of encoding the data to be encoded based on a result controlled by the control step. Method. 8. A halftone processing means for performing halftone processing on image data, a monitoring means for monitoring a generated code amount when the image data halftone processed by the halftone processing means is coded, and the generated code. A control means for controlling the halftone processing method by the halftone processing means according to the amount; and an encoding means for encoding the halftone processed image data under the control of the control means. Information processing device. 9. The information processing apparatus according to claim 8, wherein the halftone processing means converts the M-value image data into N-value image data with M> N. 10. The information processing apparatus according to claim 8, wherein the generated code amount is a code amount when the encoded data is encoded. 11. The halftone processing means has a processing first processing means for generating M-value image data and a processing second processing means for generating N-value image data different from the M-value image data. The information processing apparatus according to claim 8, wherein 12. A halftone processing step of performing halftone processing of image data, a monitoring step of monitoring a generated code amount when the image data subjected to the halftone processing in the halftone processing step is monitored, and the generated code. A control step for controlling the halftone processing method by the halftone processing means in accordance with the amount; and an encoding step for encoding the halftone-processed image data based on the control in the control step. A characteristic information processing method. 13. An information processing apparatus for encoding color image data for each predetermined unit, monitoring means for monitoring the generated code amount of coded color image data, and an encoding target for encoding in accordance with the generated code amount. An information processing apparatus comprising: a control unit that controls a compression ratio of color image data; and an encoding unit that encodes the encoding target data based on a result controlled by the control unit. 14. The control of the compression rate by the control means comprises:
The information processing apparatus according to claim 13, wherein a ratio of compression ratios assigned to each color component data of the color image data is changed. 15. An information processing method for encoding color image data for each predetermined unit, the monitoring step of monitoring the generated code amount of encoded color image data, and the encoding target of the encoded object according to the generated code amount. An information processing method comprising: a control step of controlling a compression rate of color image data; and an encoding step of encoding the encoding target data based on a result controlled by the control step. 16. A dividing unit for dividing a screen into a plurality of blocks, a monitoring unit for monitoring a generated code amount, and a control unit for controlling a compression method of each of the plurality of blocks according to the generated code amount. The information processing apparatus, characterized in that the control means controls the compression method by mixing N-valued images and M-valued images (N ≠ M) in the block. 17. The control of the compression method by the control means is such that N-value images and M-value images (N
The information processing apparatus according to claim 16, wherein a ratio of ≠ M) is controlled. 18. A dividing step of dividing a screen into a plurality of blocks, a monitoring step of monitoring a generated code amount, and a control step of controlling a compression method of each of the plurality of blocks according to the generated code amount. The information processing method, wherein the control of the compression method in the control step includes mixing an N-value image and an M-value image (N ≠ M) in the block.