JPH1056573A - Block coding method - Google Patents

Abstract

(57) [Summary] [Problem] To provide an encoding method in which image information is divided into blocks and encoded, in which image degradation due to block noise is reduced and a compression ratio is improved. In a block encoding method for dividing image data into blocks and quantizing each pixel included in each block between a maximum value and a minimum value in each block, when the pixels are quantized, Is distributed to pixels around the pixel. Also,
When the difference between the maximum value and the minimum value in the block is equal to or larger than a predetermined threshold, the pixel is quantized by the first quantization bit number, and when the difference is smaller than the threshold, the pixel is quantized. The pixel is quantized with a second quantization bit number smaller than the first quantization bit number.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an encoding method used for a digital copying apparatus, a facsimile apparatus, a scanning apparatus, an image filing apparatus, a digital camera, and the like. More specifically, the present invention relates to a block coding method for dividing image data into blocks and quantizing each pixel included in each block between a maximum value and a minimum value in each block. It is.

[0002]

2. Description of the Related Art Conventionally, in an apparatus for handling image data, such as a facsimile machine and a digital copying machine, the purpose is to reduce the memory capacity for storing the image data and to shorten the data transfer time. In some cases, an encoding method for encoding image data to compress the data amount has been used. One of the encoding methods is
Block coding method (BTC: Block Trunc)
Coding). In this block encoding method, original image data is divided into blocks of a fixed size (hereinafter, the divided blocks are referred to as “B
It is called "TC block". ), An encoding method for quantizing each pixel between a maximum value and a minimum value in a BTC block with a predetermined number of quantization bits. When encoding is performed by this method, the original image data is a code including a maximum value, a minimum value, and a quantized value of each pixel belonging to each BTC block. As described above, since the maximum value and the minimum value are generated for each block, as the size of the BTC block increases, the number of blocks included in the image information decreases. And the ratio of the minimum value to the encoded code is reduced. Therefore, in order to increase the compression ratio of image data, the larger the block size, the better.
Another approach to improving the compression ratio is to reduce the number of quantization bits in each pixel itself. From this point of view, what is ideal is to set the number of quantization bits to 1 bit.

[0003]

However, in the conventional block coding method, if the block size is simply increased or the number of quantization bits is simply set to 1 bit, the compression rate is improved, but the code is decoded. When an image is formed by the above method, a gradation difference occurs at a boundary portion between blocks, and the gradation difference causes so-called block noise that the boundary portion becomes noticeable.

The present invention has been made in view of the above problems, and an object of the present invention is to provide a coding method for coding by dividing image information into blocks while reducing image degradation due to block noise. , An encoding method for improving a compression ratio.

[0005]

According to a first aspect of the present invention, there is provided a block encoding method, comprising dividing image data into blocks, and determining a maximum value and a minimum value in each block. In a block encoding method for quantizing each pixel included in the block, a quantization error generated when the pixel is quantized is distributed to pixels surrounding the pixel.

According to the block encoding method of the present invention, a quantization error generated when a pixel is quantized is distributed to pixels existing around the quantized pixel. According to this, the image density corresponding to the quantization error generated in each pixel is sequentially stored by peripheral pixels, but not all of them, and as a set of pixels,
An image density close to the density of the original image composed of the image data can be reproduced, and the gradation of the entire block can be improved. As a result, each of the adjacent blocks approaches the density of the original image. Therefore, even if the size of the block is enlarged more than before, the occurrence of block noise can be suppressed.

Here, in the block coding method according to the first aspect, each pixel is quantized between the maximum value and the minimum value. However, even if the difference between the maximum value and the minimum value is large or small, uniform quantization is performed. The quantization is performed by the number of quantization bits. However, when the difference between the maximum value and the minimum value is small, the quantization error does not increase even if the number of quantization bits is reduced.

Therefore, in the block encoding method according to the present invention, when the difference between the maximum value and the minimum value in the block is equal to or larger than a predetermined threshold value, the pixel is encoded with the first quantization bit number. , And if the value is smaller than the threshold value, the pixel is quantized with a second quantization bit number smaller than the first quantization bit number.

According to the block encoding method of the second aspect, when the difference between the maximum value and the minimum value of the block is smaller than a predetermined threshold, the quantization is performed without increasing the quantization error. The number of bits can be reduced.

Here, when the quantization errors are sequentially distributed to the peripheral pixels, the quantization errors of the pixels in the vicinity of the extension in the block are also distributed to the pixels of the adjacent block beyond the block. . In this case, a positive quantization error is added to the pixels of the adjacent block, which exceeds the original maximum value in the original adjacent block, or a negative quantization error is added, In some cases, the value falls below the original minimum value. Then, the maximum value or the minimum value used as a reference when quantizing the pixels in the adjacent block is changed from the maximum value or the minimum value of the original image data. As a result, appropriate encoding is performed. Block noise may occur.

According to a third aspect of the present invention, there is provided a block encoding method, wherein the allocation of the quantization error is performed over a block, and as a result of the allocation of the quantization error, a block adjacent to the block is determined. When the maximum value or the minimum value is changed, quantization is performed in the adjacent block using the maximum value and the minimum value before the change.

Further, in the block encoding method according to the fourth aspect, the quantization error is distributed to only the same block.

Thus, in the block coding method according to claim 3 or 4, the original maximum value and minimum value of the adjacent block are not changed.

[0014]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, a first embodiment in which the present invention is applied to a digital copying machine (hereinafter, referred to as "copier") which is an image forming apparatus will be described. First, copy machine 1
The configuration and operation of the copying machine according to the present embodiment will be described with reference to FIG. The copying machine 1 according to the present embodiment mainly includes a scanner 2 as an image reading unit and a printer 3 as an image output unit. The scanner 2 is for optically reading an original image, and includes a contact glass 4 as an original mounting table, an exposure lamp 5, a reflection mirror 8, an imaging lens 7, a CCD image sensor 6, and the like. Exposure lamp 5
Generally, a halogen lamp is used. Reading of a document image by the scanner 2 is performed as follows.

A document placed on the contact glass 4 is irradiated with light by an exposure lamp 5 and light reflected from the document is guided to an imaging lens 7 by a reflection mirror 8 or the like. The reflected light is imaged on the CCD image sensor 6 by the imaging lens 7. The CCD image sensor 6 converts the reflected light into a digital electric signal corresponding to a document image. The CCD image sensor 6 is a full-color image sensor, and converts a given optical signal into, for example, R (red), G
(Green) and B (blue) are separated into colors, and a digital electric signal of 8 bits for each color and 256 gradations is output. The CCD image sensors 6 are arranged in a row in a direction perpendicular to the drawing (this direction is also referred to as a main scanning direction). The digital electric signal output from the CCD image sensor 6 is subjected to image processing such as color conversion processing in an image processing section described later, and is subjected to cyan (Cyan: hereinafter).
C), magenta (hereinafter referred to as M), yellow (hereinafter referred to as Y), and black (hereinafter referred to as K) color image data. The color image data is converted by the printer 3 described below into
Visualization is performed using C, M, Y, and K toners, and the obtained toner images are superimposed to form a full-color image.

At substantially the center of the printer 3, a photosensitive member 21 as an image carrier is arranged. The photoconductor 21
Is an organic photoreceptor (OPC) drum, the outer diameter of which is
It is about 120 mm. A charging device 11 for uniformly charging the surface of the photoconductor, a K developing unit 1 are provided around the photoconductor.
2, a C developing unit 13, an M developing unit 14, a Y developing unit 16, an intermediate transfer belt 19, a cleaning device 10, and the like. A laser that generates a light beam based on the above-described color image data and scans the surface of the uniformly charged photoconductor 21 above the photoconductor 21 and below the scanner 2. An optical system 9 is provided. The laser optical system 9 includes a laser diode that generates a light beam, a polygon mirror that deflects the light beam, and the like.

An image forming operation in the printer 2 performed by the above configuration will be described below by taking as an example a case based on K image data. The photosensitive member 2 is irradiated with a light beam based on the K image data from the laser optical system 9.
The latent image formed on one surface is developed by the corresponding K developing unit 12 to become a K toner image. This toner image is transferred to the intermediate transfer belt 19. Hereinafter, the transfer of the toner image from the photoconductor 21 to the intermediate transfer belt 19 is referred to as belt transfer. A series of operations such as formation of a latent image, development, and belt transfer as described above are performed by CMY
K for the four colors K.
A color superimposed toner image is formed. The four-color superimposed toner image is collectively transferred by a transfer bias roller 23 onto a recording medium fed from the paper feeding unit 22, for example, a recording paper.

The recording medium on which the four-color superimposed toner image is formed is conveyed to a fixing device 17 by a conveying belt 16. The fixing device 17 melts the four-color superimposed toner image by heating and pressing, and fixes the toner image on a recording medium. The recording medium on which the fixing is completed is discharged onto the paper discharge tray 24. On the other hand, the toner remaining on the surface of the photoconductor 21 is collected by the cleaning device 10, and the surface of the photoconductor 21 is cleaned. Photoconductor 21 after cleaning
The surface is neutralized by a neutralization device. Further, after transferring the four-color superimposed image from the intermediate transfer belt 19 onto the recording medium, the toner remaining on the intermediate transfer belt 19 is collected by the belt cleaning device 20, and the surface of the intermediate transfer belt 19 is cleaned. .

Next, the image processing section 30 of the copying machine 1 will be described with reference to FIG. FIG. 11 is a block diagram illustrating a schematic configuration of the image processing unit 30. The digital signal color-separated into three colors of R, G and B from the scanner 2 is
The shading correction circuit 1 via the line memory 31
02 is input. The line memory is provided mainly for encoding image data described later. The shading correction circuit 32 is for correcting the effects of variations in the characteristics of each element of the CCD image sensor 6, uneven illuminance of the exposure lamp 5, and the like. The output from the shedding correction circuit 32 is input to the RGBγ correction circuit 33. The RGBγ correction circuit 33 converts the digital signal from the scanner 2 from reflectance data to brightness data.

The output of the RGB γ correction circuit 33 is supplied to an image separation circuit 34 and an MTF (Modulation Tran).
sfer Function) correction circuit 35. The image separation circuit 34 determines a character portion and a picture portion in the document image and determines a chromatic color and an achromatic color in the document image, and outputs the result. The output of the image separation circuit 34 is supplied to an MTF correction circuit 35, a color conversion-UCR processing circuit 37, a scaling circuit 38,
9, input to the MTF filter 40, the γ correction circuit 41, the encoding device 42, the image memory 43, and the decoding device 44, and used for processing in each circuit and the like.

The MTF correction circuit 35 corrects the frequency characteristics of the input system of the scanner 2 and the like, and particularly corrects the deterioration of the frequency characteristics in a high frequency region. MT above
The signal after the F correction is input to the hue determination circuit 36 and the color conversion / UCR processing circuit 37. The hue determination circuit 36 determines that the output signal from the MTF correction circuit 35 is R,
It is determined which of the hues G, B, C, M, and Y the signal is, and the color conversion-UCR processing circuit 37 selects a color conversion coefficient used for color conversion. The color conversion-UCR processing circuit 37 includes a color conversion unit and a UCR (Und
er ColorRemoval) processing unit.
The color conversion unit corrects the difference between the color separation characteristics of an input system such as a scanner and the spectral characteristics of a colorant (color toner) used in a printer, and converts each of Y, M, and C necessary for faithful color reproduction. Is a part for calculating the amount of the color material.

The UCR processing section performs a UCR process for replacing a portion where three colors of Y, M and C overlap (a common density portion) with K. This is because, if three colors of C, M, and Y are superimposed, the color should theoretically be black, but it is not completely black, but the gray balance is slightly degraded and reproduced. .

The output of the color conversion / UCR processing circuit 37 is input to a scaling circuit 38. This scaling circuit 38
This is a circuit that performs vertical and horizontal scaling of an image. The output of the scaling circuit 38 is input to an image processing circuit 39. The image processing circuit 39 is a circuit that performs a specific image processing such as an image repeat process. The output from the image processing circuit 39 is MT
Input to the F filter 40. The MTF filter 40
According to the user's preference, processing to change the frequency characteristics of the input signal, such as edge enhancement processing for prioritizing the resolution of the image and smoothing processing for prioritizing the gradation of the image, is performed. Do.

The output signal of the MTF filter 40 is input to the gamma correction circuit 41 as an input image signal. The gamma correction circuit 41 is a circuit that performs gamma correction (also referred to as gamma conversion). Specifically, it is a circuit for converting the input image signal into image data to be used for image formation in the printer 2 by using an image signal conversion table according to the characteristics of the printer 3.

As described above, the digital signal from the scanner 2 becomes image data by a series of image processing from the shading correction to the γ correction circuit 110. Normally, the image data is sent to the printer 2 as it is, but in the present embodiment, the image data can be compressed by the encoding device 42 and stored in the image memory. The decoded code can be combined by the decoding device 44, and an image can be formed in the printer 3 based on the combined image data. This will be described in detail later.

In order to control each circuit constituting the image processing unit 30 as described above, a ROM 47, a RAM 46 and a C
The PU 45 is connected by a bus line 49.
The CPU 45 is connected to a main control unit 48 of the copying machine 1 via a serial interface.
Thus, the user's instruction input from the operation unit of the copying machine 1 and received by the main control unit 48 can be transmitted to the CPU 45 as a command.

The image data sent to the printer 3 is input to a laser modulation circuit 50 for controlling light emission of a laser diode in the laser optical system 9 described above. The laser modulation circuit 50 will be described with reference to FIG. FIG. 12 is a circuit block diagram of the laser modulation circuit 50. Input image data is composed of 8 bits per pixel. This image data is subjected to gamma conversion using a look-up table 51 and then input to a pulse width modulation circuit 52. The pulse width modulation circuit 52 determines the pulse width based on the upper 3 bits of the 8-bit image data. This pulse width is selected from eight states (eight values) that can be represented by a 3-bit signal. When the pulse width is determined, power modulation, that is, modulation of light emission intensity (light emission amount) is performed by a power modulation circuit 53 provided at a subsequent stage of the pulse width modulation circuit 52. The power modulation by the power modulation circuit 53 is performed in the lower 5 bits of the image data.
This is performed based on a bit signal, and one state is selected from 32 states (32 values) that can be expressed by a 5-bit signal.

Thus, based on the image data after the pulse width modulation and the power modulation, the laser diode 54
Emits light. The emission intensity of this laser diode 54 is
The light emission intensity is corrected for each pixel by monitoring with the photo detector 55 and feeding back the output of the photo detector 55 to the power modulation circuit 53.

Next, the configuration and operation of the encoding device according to the first embodiment will be described with reference to FIGS. FIG. 1 is a configuration block diagram of an encoding device 42 according to the first embodiment, and FIG. 2 is a flowchart illustrating an encoding processing operation performed by the encoding device 42 according to the first embodiment. (Hereinafter, margin)

As shown in FIG. 1, a coding apparatus 42 according to the present embodiment includes a block converter 101 for dividing image data into BTC blocks, and an n × m for temporarily storing the image data of the BTC blocks. A buffer 102, a maximum / minimum value detector 103 for detecting the maximum and minimum values of the pixels present in the BTC block, a maximum / minimum value entropy encoder 104 for encoding the maximum and minimum values, a maximum value And minimum / maximum storage 10 for storing the minimum and maximum values
5. Quantizer 106 that performs quantization for each pixel in accordance with the maximum value and minimum value, quantized value entropy encoder 107 that encodes the quantized value, and stores the quantization error generated at the time of quantization. A quantization error storage unit 110, a diffusion matrix storage unit 111 for storing a diffusion matrix used for diffusing the quantization error, and an error to be diffused (allocated) to each pixel based on the quantization error and the diffusion matrix. It comprises a quantization error diffusion unit 109 for performing an operation, and an adder 108 for adding a quantization error to each pixel based on the operation result of the quantization error diffusion unit 109. Among them, the point that the quantization error storage unit 110, the diffusion matrix storage unit 111, the quantization error diffusion unit 109, and the adder 108 are provided is a characteristic part of the encoding device 42 according to the present embodiment. Is similar to that of the conventional block coding apparatus.

In the encoding apparatus according to the first embodiment having such a configuration, the processing operation as shown in FIG.
Encodes image data. Hereinafter, description will be made with reference to FIG. First, the scanner 2 reads image information consisting of 256 gradations with 8 bits per pixel into the line memory 31 (step 201). As described above, this image information is subjected to each processing from shading correction to γ correction, and is input to the encoding device 42 as image data. The image data is converted into the block
In step 1, the image is divided into BTC blocks of nxm pixels (step 202).

Here, as shown in FIG. 9, the BTC block has a line extending in the main scanning direction of the scanner as a row and a line extending in the sub-scanning direction orthogonal to the main scanning direction as a column. It is a block composed of n rows × m columns (n and m are positive integers of 1 or more). Therefore, it is configured as a block of n pixels × m pixels consisting of m pixels continuous in the main scanning direction and n pixels continuous in the sub-scanning direction. In order to perform such division into BTC blocks, the line memory 31 needs to be able to store at least n rows of image data. The configuration of the BTC block itself is not different from the conventional block encoding method. Returning to FIG. 2, the description will be continued.

The image data divided into BTC blocks in step 202 is stored in the n × m buffer 102 in block units. The stored image data is read by the maximum / minimum value detector 103, and the maximum value (maximum value) and the minimum value (minimum value) of the density of the pixels existing in the BTC block are detected. (Step 20
3) The detected maximum value and minimum value are stored in the maximum value / minimum value storage unit 105, and are encoded by the maximum value / minimum value entropy encoder 104. On the other hand, the quantizer 1
At 06, based on the maximum value and the minimum value stored in the maximum value / minimum value storage unit 105, each pixel existing in the BTC block stored in the n × m buffer 102 is
The quantization is performed sequentially from the upper left pixel of the BTC block illustrated in FIG. 6 (step 204).

This quantization is performed between the maximum value and the minimum value with a predetermined number of quantization bits. For example, when the number of quantization bits is 2 bits, the minimum value is 00, the maximum value is 11, and the image data of each pixel in the BTC block is rounded to 00, 01, 10, and 11 as if it were. To one of the four stages. The quantized value output from the quantizer 106 is sent to the quantized value entropy encoder 107 and encoded. Further, the quantizer 106 sends the quantization error generated by the quantization of each pixel to the quantization error storage unit 110 and stores it at the same time as the quantization.

The quantization error diffusion section 109 uses the diffusion matrix stored in the diffusion matrix storage section 111 to convert the quantization error stored in the quantization error storage section 110 into any one of the above. The degree of distribution to pixels is calculated by calculation, and the result is sent to the adder 108. Then, the quantization pixel is distributed to pixels around the pixel of interest in which the quantization error has occurred by this adder (step 205).

Here, the diffusion matrix is, for example, a matrix as illustrated in FIG. In the example shown in the figure, the quantization error caused by quantizing the target pixel (*) is distributed to the neighboring pixels to the right, below, and diagonally right below the target pixel as viewed in the figure. Is shown. "0.4" in the figure,
The numbers “0.3” and “0.3” are matrix coefficients, and their sum is 1.0. As a result, the right pixel has a quantization error × 0.4 error, the lower pixel has a quantization error × 0.3 error, and the lower right pixel has a quantization error × 0.4. An error of 0.3 is allocated to each. Note that the target pixel to which the quantization error is distributed need not necessarily be the three pixels described above. When the distribution target pixel is changed, the diffusion matrix also changes accordingly.

The above steps 204 and 205
Is repeated for all the pixels in the BTC block (step 206). Step 20
The processing from step 3 to step 206 is performed on all the image data stored in the line memory (step 20).
7). Then, steps 201 to 207 are performed.
The above processing is performed for all lines of image data (step 208), and the encoding is completed.

Next, the configuration and processing operation of the encoding apparatus according to the second embodiment will be described with reference to FIGS. FIG. 3 shows an encoding device 4 according to the second embodiment.
FIG. 4 is a flowchart showing an encoding processing operation performed by the encoding device 42 according to the second embodiment. As shown in FIG. 3, the encoding device 42 according to the second embodiment has substantially the same configuration as that of the encoding device 42 according to the above-described first embodiment. 112 is provided, and accordingly, the quantization process performed by the quantizer 106 is also different from that of the first embodiment.

In the encoding device according to the second embodiment,
The image data is encoded by the processing operation shown in FIG. Hereinafter, a description will be given with reference to FIG. 4 focusing on differences from the first embodiment. First, Scanner 2 to 1
Image information consisting of 256 gradations with 8 bits per pixel is read into the line memory 31 (step 401). As described above, this image information is subjected to each processing from shading correction to γ correction, and is converted into image data by the encoding device 4.
2 is input. This image data is divided into n × m BTC blocks by the blocker 101 of the encoder 42 (step 402).

The image data divided into BTC blocks in step 402 is stored in the n × m buffer 102 in block units. The stored image data is read by the maximum / minimum value detector 103, and the maximum value (maximum value) and the minimum value (minimum value) of the density of the pixels existing in the BTC block are detected. (Step 40
3) The detected maximum value and minimum value are stored in the maximum value / minimum value storage unit 105, and are encoded by the maximum value / minimum value entropy encoder 104. On the other hand, the quantizer 1
At 06, based on the maximum value and the minimum value stored in the maximum value / minimum value storage unit 105, each pixel existing in the BTC block stored in the n × m buffer 102 is
Quantize sequentially.

The most characteristic feature of the second embodiment compared to the first embodiment is the quantization processing operation. The quantizer 106 according to the second embodiment calculates the difference between the maximum value and the minimum value stored in the maximum value / minimum value storage unit 105 upon quantization, and the calculated difference is determined in advance. It is determined whether the value is larger or smaller than the threshold value. As a result of this determination, when the “difference” is equal to or more than the threshold, quantization is performed with the number of quantization bits being s bits, and when the “difference” is less than the threshold, the s T bits are selected as the number of quantization bits having a value smaller than the bit, and quantization is performed using the t bits (steps 404 and 405).

The "s" and "t" are preset in the quantization bit allocator 112 in relation to the threshold, and the quantizer 106 determines the "difference" and the threshold. According to the result, by reading “s” or “t” set in the quantization bit allocator 112,
It is determined which quantization bit performs the above-described quantization. Here, the threshold can be appropriately selected according to the characteristics of the image forming apparatus.

The quantized value output from the quantizer 106 is sent to the quantized value entropy encoder 107 and encoded. Further, the quantizer 106 sends the quantization error generated by the quantization of each pixel to the quantization error storage unit 110 and stores it at the same time as the quantization. The quantization error diffusion unit 109 uses the diffusion matrix stored in the diffusion matrix storage unit 111 to store the quantization error stored in the quantization error storage unit 110 in any pixel. The degree of distribution is determined by calculation, and the result is sent to the adder 108. Then, the adder performs the distribution of the quantized pixels to the pixels around the target pixel having the quantization error (step 40).
6).

The above steps 405 and 406
Is repeated for all the pixels in the BTC block (step 407). Step 40
The processing from step 3 to step 407 is performed on all the image data stored in the line memory (step 40).
8). Then, steps 401 to 408 are performed.
The above process is performed for all lines of image data (step 209), and the encoding in the second embodiment is completed.

As described above, in the second embodiment,
Although the number of quantization bits is changed according to the difference between the maximum value and the minimum value of each block, the "difference" between the maximum value and the minimum value results in a slow change in the image density in the block. It is different depending on whether it is steep or steep. Therefore, according to the second embodiment, the compression efficiency of encoding can be improved by reducing the number of quantization bits for a portion where the density change of an image is slow (for example, a halftone portion).

Next, the configuration and processing operation of the encoding device according to the third embodiment will be described with reference to FIGS. FIG. 5 shows an encoding device 4 according to the third embodiment.
FIG. 6 is a configuration block diagram of FIG. 2, and FIG. 6 is a flowchart showing a characteristic portion in the encoding operation performed by the encoding device 42 according to the third embodiment. The third embodiment will also be described in comparison with the first embodiment for easy understanding of the contents.

As shown in FIG. 5, the encoding device 42 according to the third embodiment has substantially the same configuration as the encoding device 42 according to the above-described first embodiment, but is newly added to the peripheral block. The difference is that a buffer 113 and a peripheral block maximum / minimum value detector 114 are provided. And
On the premise of such a difference in the configuration, the processing operation in the third embodiment is significantly different from the processing operation in the first embodiment in the manner of diffusing the quantization error. Hereinafter, a characteristic processing operation in the third embodiment will be described with reference to FIG.
This will be described with reference to the flowchart of FIG.

The image information is read into the line memory 31, divided into BTC blocks of n × m pixels, stored in the n × m buffer 102, and stored in the n × m buffer 10.
2 until the detection of the maximum value and the minimum value of the BTC block stored in step 2 in FIG.
02 and step 203. In the flowchart of FIG. 6, steps from “detection of maximum value and minimum value in target block” (step 601) corresponding to step 203 described above are described. Here, the “target block” means a BTC block that is a current encoding target.

In the third embodiment, in parallel with the detection of the maximum value and the minimum value in the block of interest stored in the n × m buffer 102 (step 601), the quantization error The maximum value and the minimum value are detected from among the densities of the pixels in the peripheral block in which the pixel to which is to be assigned exists (step 602). For this,
The image data of the peripheral block from the block generator 101 is stored in the peripheral block buffer. The detection of the maximum value and the minimum value of the peripheral block is detected by the maximum value / minimum value detector 114 for the peripheral block. The maximum value and the minimum value of the target block and the maximum value and the minimum value of the peripheral block are all stored in the maximum value / minimum value storage unit 105. Step 60
The maximum and minimum values of the block of interest detected at 1 are encoded by the m maximum / minimum entropy encoder 104 in parallel with the above storage.

On the other hand, based on the maximum value and the minimum value stored in the maximum value / minimum value storage unit 105 of the block of interest in the quantizer 106, the inside of the block of interest stored in the n × m buffer 102 is Are sequentially quantized (step 603). In this quantization, the quantizer 106 calculates a quantization error (step 604),
The result is stored in the quantization error storage unit 604.

The quantization errors stored in the quantization error storage unit 604 are distributed to the peripheral pixels to be quantized, as in the first embodiment. At this time, the quantization error diffusion unit 109 Determines whether the target pixel to which the quantization error is to be distributed is in the same block of interest (step 605). If it is within the block of interest, the error is distributed as it is, as in the first embodiment (step 606).

If it is determined in step 605 that the quantization error is to be allocated to a pixel of a peripheral block, the density (P) of the pixel after adding the error to be allocated is added.
Is compared with the maximum value (max0X) and the minimum value (min0X) of the peripheral block stored in the maximum / minimum value storage unit 105 (step 607). As a result, when the density P is larger than the minimum value (min0X) and smaller than the maximum value (max0X), the error is directly allocated (step 609). On the other hand, when the density P is equal to or more than the maximum value (max0X), the density P is replaced with the maximum value (max0X). When the density P is equal to or less than the minimum value (min0X),
The density P is replaced with a minimum value (min0X) (step 608). With the above processing, encoding for one pixel is completed (step 610). The processing for one pixel is repeated for all the pixels in the block of interest (step 611), and the encoding of one block is completed. The point that the above processing is performed for all the image data from the scanner 2 is the same as in the first embodiment.

As described above, in the third embodiment,
The greatest feature is that the change of the maximum value and the minimum value of each BTC block due to the distribution of the quantization error can be prevented. Another modification for preventing the change of the maximum value and the minimum value of each BTC block, focusing on this point, will be described with reference to FIG. FIG. 7 is a flowchart illustrating the processing operation of the above-described modification.

This modification is characterized in that the distribution of the quantization error is limited to the above-mentioned block of interest. The maximum and minimum values of
Do not cause a change. More specifically, the quantization error diffusion unit 109 determines whether or not the error is to be distributed to a pixel in the same block as the target pixel currently being quantized (step 704). If there is, the quantization error is distributed as it is (step 7).
05), if not in the same block, the quantization error is not allocated (step 706). This gives 1
The encoding for the pixels is completed (step 707). This one
The processing for pixels is repeated for all the pixels in the block of interest (step 708), and encoding of one block is completed. The point that the above processing is performed for all the image data from the scanner 2 is the same as in the above example described with reference to FIG.

In the above modification, since it is not necessary to detect the maximum value and the minimum value of the peripheral block, the peripheral block buffer 113 and the peripheral block maximum / minimum value detector 114 are inevitably unnecessary. Become. Therefore, the encoding device 42 for executing the processing operation of the above-described modified example includes:
The configuration is the same as that of the encoding device 42 according to the first embodiment described with reference to FIG.

The encoding device 42 according to various embodiments has been described above. Copier 1 according to the present case
In the above, since the image data encoded by the encoding device 42 as described above is stored in the image memory 43, the capacity of the image memory 43 is significantly reduced as compared with the case where the image data is stored without encoding. And the cost of the entire copying machine 1 can be reduced.

As described above, the encoded image data stored in the image memory 43 is then combined by the decoding device 44 and sent to the printer 3. Here, the decoding device 43 first decodes the maximum value and the minimum value for each BTC block, and then decodes the quantized value of each pixel that has been quantized. As a result, the image data is composed of gradations before encoding.
It should be noted that the decoding apparatus described above has no difference from the conventional decoding apparatus used for combining image data encoded by block encoding, and a detailed description thereof will be omitted.

While the embodiment in which the present invention is applied to a digital copying machine has been described above, the present invention is not limited to this and can be widely applied to facsimile apparatuses, image filing apparatuses, and the like. In particular, when the present invention is applied to a facsimile apparatus, an effect other than a reduction in the storage capacity of an image memory or the like is brought about that the transfer time can be reduced by improving the compression ratio.

[0059]

According to the block encoding method of the first aspect, even if the size of the block is enlarged more than before, it is possible to suppress the occurrence of block noise. It has an excellent effect that the rate can be improved.

In particular, according to the block encoding method of the second aspect, the number of quantization bits can be reduced without increasing the quantization error. Compared with the block coding method, there is an excellent effect that the compression rate can be further improved.

In particular, according to the block coding method of claim 3 or 4, since the maximum value and the minimum value serving as the quantization reference are not changed, the change of the maximum value or the minimum value is not changed. This has an excellent effect that it is possible to prevent block noise caused by the above.

It is known that a plurality of encoders having different characteristics are appropriately switched in accordance with the density gradient of input image information to perform block encoding (Japanese Patent Laid-Open No. 5-5 / 95).
No. 6282). Also, a block in which a 1-bit code is assigned to a block in which a quantization error does not exist for each pixel of an original image is known (Japanese Patent Laid-Open No. 2-2 / 1990).
No. 14263). However, they do not disclose at all about distributing the quantization error to peripheral pixels. In addition, in each case, the encoding algorithm is complicated, and is expected to impose a considerable burden on software and hardware.

[Brief description of the drawings]

FIG. 1 is a configuration block diagram of an encoding device according to a first embodiment.

FIG. 2 is a flowchart showing an encoding processing operation performed by the encoding device according to the first embodiment;

FIG. 3 is a configuration block diagram of an encoding device according to a second embodiment.

FIG. 4 is a flowchart showing an encoding processing operation performed by the encoding device according to the second embodiment;

FIG. 5 is a configuration block diagram of an encoding device according to a third embodiment.

FIG. 6 is a flowchart showing a characteristic part in an encoding operation performed by the encoding device according to the third embodiment;

FIG. 7 is a flowchart showing a processing operation of a modification for preventing a change in the maximum value and the minimum value of each BTC block.

FIG. 8 is an explanatory diagram of a diffusion matrix.

FIG. 9 is an explanatory diagram of a BTC block.

FIG. 10 is a schematic configuration diagram of a copying machine.

FIG. 11 is a block diagram illustrating a schematic configuration of an image processing unit.

FIG. 12 is a circuit block diagram of a laser modulation circuit.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Copier 2 Scanner 3 Printer 4 Contact glass 5 Exposure lamp 6 CCD image sensor 7 Imaging lens 8 Reflection mirror 9 Laser optical system 10 Cleaning device 11 Charging device 12 K developing unit 13 C developing unit 14 M developing unit 15 Y developing unit DESCRIPTION OF SYMBOLS 16 Conveying belt 17 Fixing device 19 Intermediate transfer belt 20 Belt cleaning device 21 Photoreceptor 22 Paper feed unit 23 Transfer bias roller 24 Discharge tray 30 Image processing unit 31 Line memory 32 Shading correction circuit 33 RGB gamma correction circuit 34 Image separation 35 MTF Correction circuit 36 Hue determination unit 37 Color conversion-UCR processing circuit 38 Magnification circuit 39 Image processing circuit 40 MTF filter 41 Gamma correction circuit 42 Encoding device 43 Image memory 44 Decoding device 45 C U 46 RAM 47 ROM 48 Main control unit 49 Bus line 50 Laser modulation circuit 51 Look-up table 52 Pulse width modulation circuit 53 Power modulation circuit 54 Laser diode 55 Photodetector 101 Block generator 102 n × m buffer 103 Maximum / minimum value detection Unit 104 maximum / minimum value entropy encoder 105 maximum / minimum value storage unit 106 quantizer 107 quantized value entropy encoder 108 adder 109 quantization error diffusion unit 110 quantization error storage unit 111 diffusion matrix storage Unit 112 Quantized bit allocator 113 Buffer for peripheral block 114 Maximum / minimum value detector for peripheral block

Claims (4)

[Claims] 1. A block coding method for dividing image data into blocks and quantizing each pixel included in each block between a maximum value and a minimum value in each block, wherein the pixels are quantized. A block encoding method characterized in that a quantization error that occurs sometimes is assigned to pixels surrounding the pixel. 2. The block encoding method according to claim 1, wherein a difference between a maximum value and a minimum value in the block is equal to or larger than a predetermined threshold value, and is equal to a first quantization bit number. When the difference between the maximum value and the minimum value is less than the threshold value, the pixel is quantized with a second quantization bit number smaller than the first quantization bit number. A block encoding method characterized in that: 3. The encoding method according to claim 1, wherein the allocation destination of the quantization error is generated over a block, and as a result of the allocation of the quantization error, a block adjacent to the block is obtained. Wherein the maximum value or the minimum value is changed, the quantization is performed in the adjacent block using the maximum value and the minimum value before the change. 4. The block encoding method according to claim 1, wherein the quantization error is allocated only to the same block.