JPH10200757A - Image processing unit, its method, and image processing system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To attain the compression by which decoding with a higher compression rate, at a higher speed, at a lower cost and with high quality is attained in the case of compressing multi-value image data. SOLUTION: A difference value calculation section 200 assigns a code in response to a frequency of incidence to a difference value between each picture element of multi-value image data and its preceding picture element so as to code the data and the result is stored in a page memory 100. Then a density value calculation section 400 decodes sequentially code data in the page memory 100 and integrates obtained differences to decode the multi-value image data.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus, an image processing method, and an image processing system, for example, an image processing apparatus for storing, transferring, or transferring multi-valued image information, and a method and an image processing system.

[0002]

2. Description of the Related Art Generally, an image processing apparatus such as a laser printer has one computer built therein. The image processing apparatus communicates with an external device such as a host computer, receives commands and print data from the host computer, performs drawing development on a bitmap memory inside the image processing apparatus, and performs print output.

In recent image processing apparatuses, improvements in output resolution and advanced functions such as colorization have been implemented in accordance with the demand for higher print quality.

[0004] Further, formation and output of an image with gradation other than characters have been required. Along with this, as a storage form of output image information in the image processing device,
Conventionally, a method of storing the information of each pixel as a binary value of printing / non-printing was sufficient. However, it has become necessary to provide gradation for each pixel and store it as multi-valued gradation information. . For this reason, the storage capacity required for one pixel increases dramatically,
As a whole, a storage element for an enormous capacity is required.

Therefore, an image processing apparatus that performs multi-tone printing requires an enormous amount of storage elements in the image forming section, and is therefore expensive.

In order to realize an image processing apparatus that performs multi-tone printing at low cost, it is conceivable to reduce the required memory amount by reducing the number of tones and resolution.
This degrades the image quality of the output image.

[0007] Therefore, in order to reduce the price of the image processing apparatus while maintaining the image quality, a memory saving technique for efficiently using a limited memory is required.

As one of the memory-saving techniques, a technique for compressing image data is known. By using a compression technique, for example, by storing image data that is not used immediately in a compressed format, the occupation amount of the memory by the image data can be reduced. Then, the image data compressed and stored in this way is returned to the original format by being expanded as necessary, and used.

[0009] The spread of image processing apparatuses having a color image processing function has been remarkable. In an image processing apparatus that performs such color image processing, various color image processing methods are employed. Many image processing methods individually store a plurality of color pigments, and express an arbitrary color by changing the overlapping ratio of the plurality of color pigments when forming an image on a recording medium. .

Accordingly, in the case of monochrome image processing, it is sufficient to manage only the presence or absence of one dot when storing image data in the apparatus. On the other hand, in the case of performing color image processing, It is necessary to store image images of a plurality of colors and gradation information for each color. Therefore, in an image processing apparatus that performs color image processing, for example, when one pixel is represented by 8 bits for each color, a memory capacity required for storing image data needs to be about 30 times that of a monochrome binary image.

As described above, an image processing apparatus that expresses an arbitrary color by superimposing a plurality of primary color dyes is classified in terms of configuration according to a procedure of placing a dye on a recording medium. When roughly classified, one pixel or a small area, for each pixel color pigment is sequentially stacked, and first, a specific pigment pigment is placed on the entire recording medium,
Each of the color pigments is superimposed in a color sequential manner, and is classified into a plane sequential method. Specific examples of the image processing apparatus employing the sequential method include an ink jet printer, a dot impact printer, an ink ribbon type sublimation printer, and the like.
As a specific example of the image processing apparatus adopting the frame sequential method,
Typical printing machines, electrophotographic printers, film sheet type sublimation printers, and the like.

In general, the cost of an image processing apparatus of the frame sequential method is higher than that of an image processing apparatus of the sequential processing method. This is because the frame sequential method requires a much larger amount of image data to be stored for each color. Therefore, in particular, a multivalued color image processing apparatus of the frame sequential type requires a huge memory capacity for storing image data.

Therefore, in a multi-valued color image processing apparatus which performs frame sequential processing, the cost is reduced by providing various memory saving mechanisms. As a typical memory saving mechanism, a compression mechanism for compressing image data is also known. Storing the compressed data by the compression mechanism reduces the amount of memory used. At the same time, a decompression mechanism for decompression is provided to refer to the compressed data at the time of image formation.

In an image processing apparatus which receives digital image data from an external device such as a host computer to form a color multi-valued image and outputs it on a recording medium, the image data to be processed is In many cases, image information such as a bitmap is transferred rather than a character code or a graphics drawing output command.
Therefore, since the amount of image data transferred from the host computer is enormous, the transfer time is longer than in the case of monochrome, and the operability is deteriorated.

Incidentally, in the conventional color multi-value image processing apparatus, even if the same image forming mechanism is provided, the delicate color development differs depending on the apparatus. Therefore, when the subtle color adjustment processing is performed on each device side, the data transmission algorithm on the host computer side can be shared.

For example, while the original image data on the host computer side is multi-valued, the printing unit of the image processing apparatus has two values.
In the case of a configuration in which value recording is performed, the multivalued image data is binarized on the host computer side, and the binary
The transfer time is shorter when the value data is transferred. However, it is difficult for the printer driver on the host computer side to appropriately adjust and binarize according to the color reproducibility of the image processing apparatus of the transfer destination. Therefore, in practice, in order to perform optimal color balance adjustment, multivalued data is transferred from the host computer to the image processing apparatus, and the image processing apparatus often makes adjustments according to its own color reproduction characteristics.

As described above, when the original image data is multi-valued, the transfer data also becomes multi-valued, so that the transfer time is inevitably long. Therefore, it is effective to compress the image data in the host computer.
By compressing the image data to be transferred and reducing the data amount, the transfer time can be reduced.

As described above, in the conventional image processing apparatus, the technique of compressing image data is indispensable.

[0019]

However, when applying the image data compression technique to the above-mentioned conventional image processing apparatus, information which is frequently rewritten, such as image information held in a page memory, is compressed. If the data is saved, the calculation time required for compression and decompression becomes long, and the image generation time until print output becomes very long.

Further, since it is necessary to separately prepare an area for storing the compressed data and an area for generating and expanding the image on the memory, the memory saving effect is not significantly improved.

As a compression technique, there is a method of stably obtaining a higher compression effect by truncating a part of the original image information. According to this method, a further memory saving effect can be obtained, but deterioration that affects the output image quality is undesirable.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problem, and an image processing apparatus and a method thereof capable of performing high-speed and low-cost compression with high memory saving effect on multi-valued image data are provided. The purpose is to provide.

It is another object of the present invention to provide an image processing system and method capable of reducing the transfer time of image data by reducing the transfer amount of color image data.

[0024]

As one means for achieving the above object, the image processing apparatus of the present invention has the following arrangement.

That is, input means for inputting multi-valued image data, coding means for coding multi-valued image data input by the input means, and a difference value between the predicted value and the multi-valued image data, and the coding means Holding means for holding the code data by the decoding means, decoding means for decoding the code data held by the holding means to restore multi-valued image data, and forming the multi-valued image data restored by the Output means for outputting to the means.

Further, in order to achieve the above object, the image processing system of the present invention has the following arrangement.

That is, in an image processing system in which an external device and an image processing device are connected,
Difference value calculation means for calculating difference value information of multi-valued image data, encoding means for encoding the calculated difference value information, and transmission means for transmitting the encoded image data, On the image processing device side, receiving means for receiving the encoded image data, decoding means for decoding the received image data to obtain difference value information, and image data based on the decoded difference value information. The image processing apparatus further includes a restoration unit for restoring, and an image forming unit for forming an image based on the restored image data.

[0028]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment according to the present invention will be described below in detail with reference to the drawings.

<First Embodiment> FIG. 1 shows a laser beam printer (LB) as an image processing apparatus to which this embodiment is applied.
P) is a cross-sectional view showing the internal structure of 1000,
Reference numeral 000 is configured to allow registration of a character pattern and registration of a fixed format (form data) from a data source (not shown) (a host computer or the like).

In FIG. 1, reference numeral 1000 denotes an LBP main body, including image data supplied from an externally connected host computer or the like via an input I / F 1013, character information (character code), form information, or a macro. A command or the like is input and stored, and a corresponding character pattern, form pattern, or the like is created according to the information, and an image is formed on a recording sheet as a recording medium. Hereinafter, information to be subjected to image formation in the LBP 1000 is generally referred to as image information.

Reference numeral 1012 denotes an operation panel on which various switches and LED indicators for operation are arranged. Reference numeral 1001 denotes control of the entire LBP 1000 and analysis of character information supplied from the host computer. Is a printer control unit that executes compression processing of image data. This printer control unit 1001
For example, character information is converted into an image signal of a corresponding character pattern and output to the laser driver 1002.

The laser driver 1002 is the semiconductor laser 1
003 is a circuit for driving the semiconductor laser on / off according to an input video signal.
The laser beam 1004 is swung right and left by the rotary polygon mirror 1005 to scan on the electrostatic drum 1006. This allows
An electrostatic latent image of a character pattern is formed on the electrostatic drum 1006. This latent image is developed by a developing unit 1007 around the electrostatic drum 1006 and then transferred to a recording sheet. A cut sheet is used as the recording paper, and the cassette recording paper is stored in a paper cassette mounted on the LBP1000, and is supplied with a paper feed roller 1009 and transport rollers 1010 and 1010.
11 and the electrostatic drum 100
6.

In this embodiment, image data is compressed and held in the page memory in the printer control unit 1001. Note that the LBP 100 of the present embodiment
At 0, 8-bit gradation expression is performed.

FIG. 2 is a block diagram showing the basic configuration of the printer control unit 1001 for implementing the compression processing in this embodiment.

In FIG. 2, image data input via the input I / F 1013 is first calculated by a difference value calculation unit 200 for each pixel in the form of difference information of the density value from the previous pixel. The calculated difference information is stored in the page memory 100.
At this time, the address control unit 3
00, the address on the page memory 100 is determined.

Reference numeral 1014 denotes a printer unit, which includes a configuration for performing image forming processing after the laser driver 1002 in the configuration shown in FIG. When the difference value information stored in the page memory 100 is read and an image is formed by the printer unit 1014, the density value calculation unit 400 reproduces the original density value from the difference value information stored in the page memory 100. Then, it is sent to the printer unit 1014. The reading of the difference value information from the page memory 100 is controlled by the address control unit 300.

For example, when image data is input from the input I / F 1013 to rewrite a part of the image data already stored in the page memory 100, the difference information of the difference information already stored in the page memory 100 is input. Only specific areas need to be updated. In this case, the density value calculation unit 400 reads the difference information from the page memory 100 based on the address information, calculates the density value, and transfers the density value to the difference value calculation unit 200. Thereby, the difference information to be newly stored in the difference value calculation unit 200 can be obtained, and the difference information of the corresponding address in the page memory 100 can be updated.

Reference numeral 500 denotes a CPU, which controls the components in the printer control unit 1001 according to a control program stored in the ROM 600. 700
Is a RAM, which is used as a work area of the CPU 500.

As described above, in the present embodiment, as a method of compressing and storing image data in the page memory 100 for expanding a generated image, individual pixel information does not directly represent a density value, It is characterized by being expressed as difference information.

In this embodiment, since one pixel is represented by 8 bits, possible density values are "0" to "255".
It is. Therefore, the difference information corresponding thereto is “± 2
55 ". Therefore, in order to express the difference, 9 bits including the sign are required. However, in the present embodiment, the code range can be treated as the same 8 bits as the density information by ignoring the overflow at the time of calculation. That is, in the present embodiment, the difference information is 2
Is stored in the page memory 100 as the complement of the above. In the density calculating section 400, a hardware subtraction mechanism and negative number processing of data information when reproducing a density value from a difference value are not required.

Hereinafter, the process of restoring the density value in the density value calculating section 400 will be described in detail. For the sake of simplicity, the configuration is described in three stages (FIGS. 3, 4, and 7). I do. The detailed configuration of the difference value calculation unit 200 is the same as the configuration of the density value calculation unit 400 and will not be described.

First, FIG. 3 shows a density value calculating section 400 for restoring and outputting density information from difference information when outputting image data to the printer section 1014, and an address control section 300.
The detailed configuration of is shown.

In FIG. 3, the density value calculation unit 400 includes an adder 401 and a register 402 for holding the density value of the previous pixel. Further, the address control unit 300 controls the page memory 1
A counter 301 for determining a read address of 00 and a delay unit 302 for controlling the timing of addition by the density value calculation unit 400.

In FIG. 3, the control signals 501, 502 and 503 input to the address control unit 300 are
This is input from 500.

The page memory 103 already stores a density difference value for each pixel. Hereinafter, the operation of reading the difference value, restoring the density value of each pixel, and printing out, that is,
The density reading operation will be described.

First, the timing of data transmission to the printer unit 1014 is determined by a timing arbitration unit (not shown) for adjusting the timing between the printer control unit 1001 and the printer unit 1014. Then, the counter 301 of the address control unit 300 is instructed by the control signal 503 from the CPU 500 to set the head address of the transmission pixel row. At the same time, the control signal 503 for setting the start address is also transmitted to the register 40 of the density value calculation unit 400.
Clear 2 to "0". That is, in the present embodiment, it is assumed that the density of the pixel immediately before the end pixel that does not actually exist is “0”. The head address of the pixel column is determined by the CPU 5 according to a timing signal from the timing arbitration unit.
00, the CPU bus 5
02 is set to the counter 301 via the counter 02.

Here, if the code length (data length for each pixel) of the image data stored in the page memory 100 is constant (8 bits in this case) and the lengths of the pixel columns are all equal, each pixel By arranging the columns continuously in the memory space and setting the addresses of the first column, the addresses of all the pixels can be obtained. That is, in response to the output start signal from the printer unit 1014 for one image output, the CP
Only the first address of the first column is set via the U bus 502.

It should be noted that such a configuration for performing address setting can be realized by hardware. For example, a circuit configuration is provided in which one register for storing the start address of the page memory 100 is prepared, and the stored value of the register is loaded into the counter 301 in response to an output start signal from the printer unit 1014. And it is sufficient.

When the time T1 elapses as a delay corresponding to the access time to the page memory 103, the page memory 100
Image data (difference value) is read from the
The adder 401 of 00 adds the density value of the previous pixel stored in the register 402. The overflow that occurs at this time is ignored. After the delay time T2 for addition has elapsed, the reproduced density information is output to the printer unit 1014.

Delay unit 30 in address control unit 300
Reference numeral 2 denotes a simple delay mechanism or a timing generation mechanism that generates an appropriately delayed level signal or latch pulse based on a control signal 501 input from the CPU 500 and outputs the control signals 311 and 312. I do. The control signal 312 notifies the printer unit 1014 of the validity of the data at a timing that allows a delay time of T1 + T2 from the determination of the new address. Further, the printer unit 10
After the elapse of the time T3 required for the data to be read by the memory 14, the control signal 311 is output to the register 402 as a timing pulse for latching the density information of one pixel output from the adder 401. Here, the adder 4
01, the latch time of the register 402 is sufficiently shorter than the delay time T2.

Here, if the delay time T2 in the adder 401 is larger than the latch time of the register 402 and a stable value cannot be fed back, the density information output from the adder 401 is replaced with the density information as necessary. It is only necessary to add another stage to the output of the adder 401 of a latch for feeding back to the register 402.

Reference numeral 501 denotes an address update signal controlled by the CPU 500. Here, the page memory 100
The code length of the difference value stored in the page unit 100 is constant, and the amount of data read from the page memory 100 and the printer unit 1
014 is equal to the output data amount. Therefore, the page memory 100 is simply set at the same timing as the pixel clock.
Is generated.

As described above, the difference value stored in the page memory 100 is read to reproduce the density information, and the printer unit 101
4 has been described. Next, an operation for partially updating the difference value stored in the page memory 100, that is, an update process will be described.

To partially update the image data stored in the page memory 100, it is necessary to read the difference value of the target pixel already stored in the page memory 100. Here, if the process of scanning the pixel array up to the pixel to be updated in the page memory 100 and obtaining the density information of the pixel is performed by software, it takes an extremely long calculation time. In this embodiment, the process is performed by hardware. Realize.

FIG. 4 shows a configuration for realizing the updating process of the page memory 100 in addition to the configuration shown in FIG. FIG. 4 shows that the address control unit 300 is turned on / off of the address control unit in the configuration of FIG.
This figure shows a configuration in which a flip-flop (F / F) 303 for controlling off, a register 304 for holding position information of a writing pixel, and an AND gate 305 for controlling an address update signal 501 are added. The updating process of the page memory 100 is also controlled according to a control program executed by the CPU 500.

When partially updating the page memory 100, the CPU 500 first determines the start address of the pixel row to be calculated from the CPU bus 502 through the counter 301.
Is set, and the position information of the pixel of the pixel column in which the density is to be obtained, that is, the position information of the pixel to be accessed is indicated by the control signal 5.
05, the register 304 is set, and the control signal 504 is activated. Thereby, the access preparation of the set pixel is completed, and the address control unit 300 is driven.

Since the address update signal 501 controlled by the CPU 500 is gated by the AND gate 305, the address update operation is controlled by the F / F 303. A new address of the page memory 100 is supplied via the counter 301 by the address update signal 501, and the value at the address of the page memory 100 is input to the adder 401. Then, the adder 401 adds the value stored in the register 402 to reproduce the density value, and outputs the density value as density information 410. In delay section 302, address update signal 50
1, a timing signal 311 for feeding back the density information 410 to the register 402 is generated.

The counter 301 determines the read address of the page memory 100, and stores the head address of the target pixel column. The register 304 is a register for counting the number of the pixel in the target pixel column to be updated. Therefore, register 3
By sequentially counting down the held value of “04” and performing an adding process from the head address of the target pixel column in the density value calculating unit 400 until the held value becomes “0”,
The density information 410 of the pixel to be updated can be obtained.

The reproduced density information 410 of the pixel to be updated
Is input to the difference value calculation unit 200 by the CPU 500. As a result, the difference I / F
A new difference value to be stored in the page memory 100 can be obtained from the new density value of the pixel to be updated input from 1013. Furthermore, in order to maintain the consistency of the difference information in the page memory 100, the difference value of the next pixel of the pixel row to be updated is also adjusted based on the density information 410.

The difference value calculating section 200 calculates the difference between the density information 410 of the pixel to be updated calculated by the density value calculating section 400 and the density value newly set for the pixel to be updated. By adding to the stored difference value information, it is stored as new difference value information of the update target pixel. Then, a new density value of the last pixel to be updated and density information 4
The difference value from 10 is stored.

In the normal updating process for the page memory 100, it is rare that only one pixel is updated, and pixels in a continuous area are updated. Therefore, in this embodiment, the page memory 100 is divided into pixel rows of a predetermined length, and only the density value of the first pixel of the pixel row is calculated by the density value calculation unit 400. This becomes possible in the calculation unit 200. That is, it is possible to minimize the number of calculations of the density calculation in the density value calculation unit 400.

As shown in FIGS. 3 and 4 described above, the page memory 100 is accessed by the CPU 500. However, an access arbitration unit (not shown) avoids access conflicts with the hardware of this embodiment. .

FIG. 5 shows a flowchart of address control for updating the page memory 100, which will be briefly described below. As described above, the processing shown in the flowchart is executed by the CPU 500.

In the flowchart, the address of the page memory 100 is represented by X and Y coordinates, and a value having the same Y coordinate is treated as one pixel row. For each pixel, P is old density information, Q is new density information (updated density information), A is an address on the page memory 100, D is a difference value between the pixel and the previous pixel, that is, Indicates the value actually stored.

First, a pixel range to be updated is obtained as an object to be updated (S101). For example, if the pixel range to be updated is rectangular, and the address of the upper left pixel is (Xa, Ya) and the address of the lower right pixel is (Xb, Yb), the object to be updated is (Xa, Ya)-( X
b, Yb).

In this embodiment, the page memory 10
In order to perform the update process of 0 for each pixel column, the object to be updated is divided for each column, and the column whose Y coordinate is Ya (the Ya column)
The update process is performed sequentially from column to Yb column (S102, S1)
03, S118).

As a specific update process in the Yi column (a ≦ i ≦ b), first, an update range on the i column is obtained. That is, in the update range Xa to Xb on the i-th column, Xa is set as the start pixel Xi1 to be updated, and Xb is set as the end pixel Xik (S104).
Then, the start address Ai of the Yi column is obtained (S106),
Ai is output to the CPU bus 502, Xi1 is output to the control signal 505, and the control signal 504 is activated (S10
7). Thus, the access preparation for the update target top pixel Xi1 is completed, and the address control unit 300 is driven.

Next, the address Ai1 of Xi1 is obtained (S10
8) After the lapse of a predetermined time, the density value calculation section 400 obtains the density information 410 (old density Pi1) for Ai1 (S109, S110). That is, the density value of the first pixel to be updated is obtained in the Yi column.

Then, in the difference value calculating section 200, based on the density value Pi1 of the first pixel Xi1 to be updated, a new difference value (Qij
-Pij + Dij) are sequentially calculated and stored in corresponding addresses of the page memory 100 (S111 to S116).

When the storage of the difference value up to Xik is completed, the difference value of the pixel after Xik is further updated, whereby
The old and new densities are matched (S117). This allows Yi
The update process in the column ends.

With the above address control, a process of partially updating the page memory 100 in the present embodiment is realized.

In the configuration of FIGS. 3 and 4 described above,
As a basic concept of the present embodiment, holding image data as a difference value from a previous pixel (difference value page memory) has been described. However, even if the image data is held in the page memory 100 as a difference value from the previous pixel in this way, the difference value information has the same data length as the density value information (8 bits in this case). Is not obtained.

Normally, an image processed by the image processing apparatus is an image having a density correlation between neighboring pixels in most areas. This means that the image to be treated is not an image in which the density changes irregularly for each pixel, but has the property that its brightness gradually changes in a region much larger than one pixel. Is shown.

For this reason, when the image information is converted into the difference information of the density value with the previous pixel as described above, most of the difference values concentrate on a relatively small value due to the density correlation of the neighboring pixels. This tendency becomes more pronounced as the resolution increases.

Therefore, by replacing the difference information with a code having a shorter data length, the occupied memory amount can be effectively reduced.

Further, as described above, the tendency that the appearance frequency of the difference value is higher as the value (mutation value) is smaller does not depend on a specific image, but is a stable property in a normal image. Therefore, when replacing the difference value with a code, dynamic allocation of the code is not required, and the compression effect can be sufficiently improved by the fixed allocation. In addition, by fixing the correspondence between the difference value and the code in this way, the encoding and decoding processes and the like become relatively simple, and can be realized with a simpler configuration.

Therefore, in this embodiment, a fixed code is assigned to the difference value for each pixel, and the page memory 100
Is stored.

In the present embodiment, 8 is used as the gradation expression.
An example in which difference information is stored as a 4-bit code in an image processing apparatus that performs bit representation as a format that simplifies hardware circuits and ensures processing speed will be described below.

First, the encoding method of the difference value in the present embodiment will be described.

In the present embodiment, the difference value calculating section 20
At 0, an encoding unit is provided, and the difference value calculated by the encoding unit is converted into a predetermined code. In this embodiment,
The difference values of +7 to -7, which have relatively high appearance frequencies, are converted into codes represented by 4-bit two's complement expressions. The other difference values, that is, the density conversion area having a low appearance frequency, are represented by a 4-bit exception code + 8-bit difference information, that is, a 12-bit representation. Therefore, in the present embodiment, the code lengths of the high-frequency area and the low-frequency area are different, but in the density value calculation unit 400, the 4-bit information of the low-density area is sign-extended and added as 8-bit difference information. It is input to 401.

In the present embodiment, when calculating the difference value of the 8-bit density information in the two's complement representation, by ignoring the overflow, white and black, which appear frequently in a character (text) image, are particularly high. , That is, the density change (difference value) between the density value 0 and the density value 255 can be regarded as ± 1. Here, the difference value ± 1 is a difference value having a high appearance frequency in a natural image. Therefore, in the present embodiment, in the encoding process for the text image and the graphic image, the difference value having a high appearance frequency becomes a smaller value in each case. I can do it.

Further, by expressing the difference value in a two's complement expression, a large density displacement other than the density values 0 and 255 can be obtained.
Similarly, it is regarded as a small difference value. For example, the density shift +254 is treated in the same way as the density shift -2. In general, the frequency of occurrence of minute concentration changes is much higher than that of large concentration variations, and therefore, even if such an expression is used, the frequency of occurrence of statistical minute changes has no practical effect.

Since the digital arithmetic circuit of this embodiment ignores carry, the restored difference value is
No distinction is made whether to be restored as a large displacement or a small displacement. That is, for example, it does not distinguish whether to perform text image processing or graphic image processing.

FIG. 6 shows a code table defining code assignment in the present embodiment. According to FIG. 6, a 4-bit code is assigned to each of the difference values 0 to ± 7. For the difference value exceeding ± 7, “1000” is assigned as an exception code. By assigning a fixed-length code to each difference value in this way,
Encoding processing by hardware becomes easy. The code table is stored in the ROM 600 in the apparatus in advance, but may be stored in the RAM 700 and updated.

When a code based on the code table is stored in the page memory 100, 0 to ±
For the difference value of 7, the assigned 4 bits are stored, and for the difference value of low occurrence frequency exceeding ± 7, the 8-bit difference value itself is stored after the exception code 4 bits. That is, if it is an exception code, 12
A bit indicates one difference value.

By performing such encoding, for example, 90% of the density information of all the pixels is
Considering the case of processing an image in which only a level of density displacement appears, the occupied memory amount of the code in the page memory 100 is only 60% of the case where the above coding is not performed. Therefore, assuming such 90% 4-bit encoding, 60% of the page memory capacity for storing density values should be allocated as a memory for storing each pixel column when the device is mounted. . If a pixel row in which a sign having a difference value larger than expected appears occurs, the memory amount normally allocated is not enough, and the allocation of the page memory 100 is changed.
What is necessary is just to control so that the storage capacity for storing the code pixel sequence becomes larger. Therefore, in this embodiment,
In order to efficiently allocate the page memory 100,
A means for changing the size of the storage capacity allocated to the pixel column to be subjected to the difference processing is prepared.

FIG. 7 shows, in addition to the configuration shown in FIG. 4, the density for restoring and outputting density information when code data by the above-mentioned code allocation is stored in the page memory 100 as a difference value for each pixel. 2 shows a detailed configuration of a value calculation unit 400 and an address control unit 300. 7, the same components as those in FIG. 4 described above are denoted by the same reference numerals, and description thereof will be omitted.

In this embodiment, since the code length stored in the page memory 100 is in units of 4 bits, in FIG. 7, in order to simplify the hardware configuration, the page memory 100 is accessed with 4 bits per word. Is shown.

In this embodiment, when the exception code occurs, the code length changes from 4 bits to 12 bits.
The number of pixels read from the page memory 100 does not match. Therefore, at the time of restoring density information of the present embodiment, a clock three times as large as a pixel signal is required as an address update signal of the page memory 100. Thus, four bits can be read from the page memory 100 three times per pixel, that is, 12 bits can be read per pixel. Of course, the normal 4
At the time of performing bit reading, masking is performed twice during three readings of the page memory.

In FIG. 7, an AND gate 306 is provided in the address control unit 300 to perform the above-described memory access control. The AND gate 306 controls updating of the readout target pixel number held in the register 304 by a control signal from the sequencer 403 described later.

The time required to access the page memory 100 can be further reduced by using a configuration in which the memory access and the reading are completely asynchronous so that the code data is read from the page memory 100 in word units. Can be.

In the density value calculation section 400 shown in FIG. 7, the 4-bit code read from the page memory 100 is normally expanded to 8-bit difference value data, and the adder 401 is added.
Enter At the time of sign extension, when an exception code indicating a difference value of ± 7 or more occurs, 8-bit difference value data is read from the page memory 100 in addition to the 4 bits of the exception code. 403 are provided. In order to process 8-bit data in exception processing, latches 404 and 405 for holding upper and lower 4-bit data are provided. Note that the above-described sign extension processing is realized by the latch 404.

At the time of sign extension, the output of the extended upper 4 bits is output from the control signal 41 from the sequencer 403.
2 is controlled. The control signal 412 is disabled when the difference value read from the page memory 100 is not 4 bits.

Hereinafter, the sign extension processing in the present embodiment will be described.
The control of exception processing will be described in detail.

The control signal 411 output from the sequencer 403 causes the 4-bit code data read from the page memory 100 to be sign-extended as upper 4 bits of an 8-bit differential value output, which is obtained by sign extension in the latch 404. It is determined whether to use the upper 4-bit output or the 4-bit output of the latch 405. or,
The latch of the data on the data bus read from the page memory 100 into the latches 404 and 405 is the control signal 4
12, 413.

The sequencer 403 includes the page memory 100
Has a function of detecting the pattern of the 4-bit code read from the. For example, as a 4-bit code, "100
When the exception code “0” is recognized, the latch 404
Is output to the control signal 411
To disable. And furthermore, Gate 3
The next two address update clock inputs to register 304 are masked by control signal 312 to 06,
The correct number of pixels is counted. In the exception processing, the update of the density value of the register 402 is masked by the gate 406 and the control signal 414 until the upper and lower 4 bits are read from the page memory 100.

The above-described sign extension processing is also performed by the CPU 50.
0 is controlled according to the control program executed by the control program.

The details of the sign extension control in the sequencer 403 described above are shown in the flowchart of FIG.

In FIG. 8, first, the control signal 312 is enabled (S501). When a clock signal is input (S502), the sequencer 403 sends the clock signal from the page memory 100 after delaying the clock signal by the page memory 100 access time. The 4-bit code data on the data bus is read (S503). Then, if the bit pattern of the 4-bit data is not "1000", normal sign extension processing is performed. That is, the sequencer 403 outputs the control signal 312
Is disabled, and the subsequent pixel count by the pixel clock is masked. Further, the control signal 414 is enabled, and the updating of the register 402 is permitted. Also, the control signal 411 is enabled, and an instruction is given to perform sign extension by the latch 404 (S505). Then, the control signal 412 is enabled (S506), and the latch 4
The 8-bit difference value data obtained as a result of the sign extension by the code number 04 is output to the adder 401. After waiting for two clocks to elapse (S507, S508), the process returns to step S501 to start processing the next pixel.

On the other hand, if the read 4-bit code data is an exception code of "1000", exception processing is performed. That is, the sequencer 403 disables the control signal 414 and prohibits updating of the register 402. In addition, the control signal 411 is disabled, and an instruction is given to use the latch 405 as the extension code output (S5).
09). After a lapse of one clock (S510), the sequencer 403 enables the control signal 412 after delaying the clock signal by the page memory 100 access time, and the four bits held in the latch 404 become the lower four bits of the 8-bit difference value. It is output as a bit (S511). Then, after a lapse of one more clock (S512), the sequencer 403 enables the control signal 413 after delaying the clock signal by the page memory 100 access time, and the four bits held in the latch 405 become the upper four bits of the 8-bit difference value. It is output as a bit (S513). Then, the control signal 414 is enabled to allow the update of the register 402 (S514), and the process returns to step S501 to start the processing of the next pixel.

As described above, the sign extension processing and the exception processing of the present embodiment are controlled by the sequencer 403 controlling each control signal.

In this embodiment, the page memory 1
Since the code length stored in 00 is not fixed, it is necessary to scan the stored pixel column in order to determine the storage position of the pixel to be read in the page memory 100. This operation is realized by the configuration shown in FIG.
In order for the CPU 500 to obtain the address of the writing position,
The value of the counter 301 is read. Therefore, the CPU bus 502 is bidirectional in FIG.

In FIG. 7, reference numeral 101 denotes a storage capacity changing unit. As described above, when the multivalued image data is encoded by the difference value calculating unit 200 and stored in the page memory 100, the page memory 100 is changed. In order to efficiently perform the allocation, the change in the size of the storage capacity allocated to the pixel column on which the difference processing is performed is controlled based on an instruction from the CPU 500. In the storage capacity changing unit 101, an additional area is simply allocated, or areas of several storage capacities having different sizes are prepared in advance, and reallocated according to the used amount.

The simplest reallocation method in the storage capacity changing unit 101 is that a minimum storage area is prepared in advance for all pixel columns as processing units, and a storage area allocated in the page memory 100 is prepared. Is a method of allocating a larger storage area when the remainder of the data becomes short of a predetermined value. Alternatively, if the encoded difference code string is short, a small storage area is allocated, and if it is long, a larger storage area is allocated, so that the reallocation can be performed more efficiently. The determination as to whether or not the remaining storage capacity is insufficient may be made, for example, by comparing (remaining storage capacity size / longest code length) with the number of unprocessed pixels. In such a reallocation method, the processing mechanism is simple because the addresses are continuous on hardware, but it takes time to transfer the existing data to the new area after the reallocation.

On the other hand, the additional allocation method is a method of simply accessing another spare area when the area exceeds the specified area. Therefore, it is not necessary to move the existing data after the additional assignment, but it is necessary to switch the address in the middle of the image data sequence. Therefore, when the difference value is integrated by software, the read address change processing is sufficient. However, when the difference value integration processing is realized by hardware as in the present embodiment, the address storage is performed. An additional configuration such as a register is required, resulting in a more complicated circuit configuration.

Therefore, the storage capacity changing unit 101 of the present embodiment
Adopts a reallocation method.

Incidentally, the storage capacity changing section 101 also controls the control signal 5
08 is controlled by the CPU 500.

In the configuration shown in FIG. 7, when the code data held in the page memory 100 is updated, the storage location of the pixel column data in the page memory 100 due to the occurrence or disappearance of the exception code "1000" It is necessary to consider the possibility of movement. Accordingly, the flowchart of the address control for accessing the page memory 100 shown in FIG. 5 described above changes as shown in FIGS. Hereinafter, the page memory 100 in the configuration of FIG.
Update processing will be described.

9 and 10, the same processes as those in FIG. 5 described above are denoted by the same step numbers, and description thereof is omitted. In the flowchart, the page memory 1
The address of 00 is represented by X and Y coordinates, and a value having the same Y coordinate is treated as one pixel column. For each pixel, P is old density information, Q is new density information (updated density information), A
Indicates an address on the page memory 100, and D indicates a difference value between the pixel and the previous pixel, that is, a value actually stored in the page memory 100.

In addition, if processing is performed on a pixel-by-pixel basis because the code length dynamically changes, the pixel column information on the page memory 100 is dynamically moved each time an exception code occurs or disappears. May be necessary. In order to perform this pixel column processing more efficiently than in pixel units, 4-bit line buffer groups LB0 to LBn are prepared.

In the updating process shown in FIGS. 9 and 10,
First, similarly to FIG. 5 described above, the update target object (X
a, Ya)-(Xb, Yb), and the update range Xi1 to Xi on column i
After obtaining k, the access preparation of the update target top pixel Xi1 is completed, and the address control unit 300 is driven (S101 to S101).
107).

The updated density information Qi1 to Qik is converted to a difference value to obtain difference values D'i2 to D'ik-1 other than the first and last to be newly stored. Store in buffer. That is, since the difference value of the first pixel cannot be calculated only from the updated density information Q, a 12-bit space (LB0 to LB2) is prepared at the head of the line buffer. Then, the next address where the last data is stored in the line buffer group is set as LBPe (S60).
1). Note that this processing is performed while the drive preparation processing of the address control unit 300 is performed by hardware.
It is efficient if performed by the CPU 500.

Then, the CPU 500 stores the page memory 10
The old density Pi1 of the first pixel Xi1 to be updated is obtained from the density information 410 read and restored from 0, and the CPU bus 502
, The address Ai1 of the page memory 100 is obtained from the counter 301 via the counter 301 (S602).

Subsequently, the value of the address Ai1-8 two clocks before, that is, eight bits before, is read from the page memory 100. If the data is the exception code "1000", the address Ai1-8 is updated to the start address of the update target. Ai1 (S60
3, S604).

Then, the original difference value Di1 is obtained from the address Ai1 on the page memory 100 (S605), Xik + 1 is output to the control signal 505, and the control signal 504 is activated (S606). As a result, the access preparation for the next pixel Xik + 1 of the update target terminal pixel is completed, and the address control unit 30
0 is driven.

In the meantime, Di1 + Qi1-Pi1 is calculated and encoded as a difference value to be newly stored for the first pixel. If the encoding result is an exception code, that is, if the data to be stored has all 12 bits, the head address of LB0 is set as the head data storage address LBPs of the line buffer, and the normal address is a normal 4-bit code that is not an exception code. For example, the head address of LB2 is set to LBPs (S
607-S609). Then, the obtained encoding result of the first pixel is stored in the line buffer from the LBPs (S610). This indicates that the update data length on the line buffer can be obtained by LBPs-LBPe.

Next, the CPU 500 sets the page memory 100
The old density Pik + 1 of the terminal next pixel Xik + 1 to be updated is obtained from the density information 410 read out and restored from the CPU bus 5.
The corresponding address Aik + 1 of the page memory 100 is obtained from the counter 301 via the counter 02 (S612).

Then, in order to obtain an updated value of the terminal pixel to be updated, Qik-Pik + 1 is obtained and encoded, and then stored in the address LBPe of the line buffer to update LBPe (S613).

Then, it is determined whether or not to shift the subsequent existing pixel row in accordance with the update of the data on the page memory 100. That is, the update data length on the line buffer is compared with the data length actually read from the page memory 100, and if they do not match, it is determined that a shift in the pixel column due to the occurrence or disappearance of the exception code has occurred, and Perform a move process. Specifically, assuming that the moving amount of the subsequent bit string is Mv, Mv can be expressed by the following equation.

Mv = (LBPe−LBPs) − (Aik + 1−Ai1) If the Mv is “0”, there is no shift. Transfer to address Ai1. On the other hand, M
If v is not “0”, it is determined that a deviation has occurred, and the bit string after address Aik + 2 of the page memory 100 is moved to address Aik + 2 + Mv, and then LB on the line buffer is read.
The data of Ps to LBPe is transferred to the address Ai1 on the page memory 100 (S614 to S617).

Thus, the updating process of the Yi column is completed, and the next Yi column is updated.
The process for the +1 column is started (S618).

As described above, according to the flowcharts of FIGS. 9 and 10, continuous rewriting on the pixel column to be updated in the page memory 100 can be performed collectively. As a result, even when it is necessary to move the existing pixel column information in the page memory 100, the moving process is performed only once for each pixel column serving as an update unit, which is far more than the pixel unit process. Only a small number of processings are required.

The operation in the configuration shown in FIG. 7, including the processing shown in the flowcharts of FIGS.
Control is performed according to a control program executed by CPU 500.

As described above, according to the present embodiment, when the multi-valued image data is encoded and stored in the page memory 100, the occurrence frequency deviation is made constant by treating it as difference information from the previous pixel. By statically assigning a code corresponding to the frequency of occurrence to the difference information, compression with a high memory saving effect can be performed.

Further, the restoration and updating of the image data held in the page memory 100 can be performed at high speed.

<Second Embodiment> Hereinafter, a second embodiment according to the present invention will be described.
An embodiment will be described.

In the above-described first embodiment, an example has been described in which a difference value having a high appearance frequency is assigned to a fixed-length code, and the other difference values are stored in the page memory as fixed-length exception codes. In the second embodiment, an example will be described in which a difference value is assigned to a variable-length code such that the appearance frequency of the difference value is inversely proportional to the code length, such as a Huffman code.

FIG. 11 shows a code table for defining code assignment in the second embodiment. FIG. 11 shows an example in which a Huffman code having a code length inversely proportional to the number of distributions is assigned to each 4-bit difference value having a distribution tendency in the center of the figure. In this example, the code length changes from 1 bit to 10 bits, and the code length assigned to the difference value “0” having the highest appearance frequency is as short as 1 bit. Since the difference values are expressed in two's complement, the difference values +1, +2,... Include -15, -14,. The code table is stored in the ROM 600 in the apparatus in advance, but may be stored in the RAM 700 and updated.

By performing encoding based on the code table on a predetermined image, the data size stored in the page memory 100 becomes 41% of the original image data. At this time, the average code length per pixel is 1.6.
4 bits. That is, the area occupied by the page memory 100 can be suppressed with high efficiency.

Of course, the appearance frequency of the difference value varies depending on the assumed format and image characteristics of the output image, and therefore the optimal code assigned in the code table differs.

The configuration of the image processing apparatus according to the second embodiment is the same as that shown in FIGS. 1 and 2 according to the first embodiment, and a description thereof will be omitted. FIG. 12 shows an address control unit 300 and a density value calculation unit 400 according to the second embodiment.
The detailed configuration of is shown. 12, the same components as those in FIG. 7 described above are denoted by the same reference numerals, and description thereof will be omitted.

In the second embodiment, since the code length stored in the page memory 100 differs for each difference value, the density calculator 400 converts the variable length code read from the page memory 100 into a difference value. Configuration is required. Further, since variable-length codes are handled, processing must be performed completely in bit units. Therefore, it is necessary to perform arbitration with the reading mechanism or the like of the page memory 100 based on the segmented word length such as 16 bits or 32 bits. That is, in order to recognize a bit position to be processed in the page memory 100, the bit position is stored in the page memory 1
It is necessary to provide a configuration for converting into two pieces of information, that is, address information in word units within 00 and bit position information in one word.

In the configuration shown in FIG.
0, before the adder 401, the page memory 100
, A reading mechanism for reading a bit string of a predetermined length from the bit string and extracting a Huffman symbol from the bit string, and a converting mechanism for converting the extracted Huffman code into difference value information are added. These components are the code conversion unit 407.

With the configuration shown in FIG. 12, the density value of the pixel to be written, the address position in word units, and the bit position in one word being processed can be obtained. When the CPU 500 executes the control program in the ROM 600, the code of the difference information at the write position of the page memory 100 is rewritten. Data bit shift processing is performed on 100.

Further, reading out the word unit by address designation in the page memory 100 and the printer unit 1014
Is completely asynchronous with the output of data.

In the configuration shown in FIG.
Reading from an address unit from 0 is managed by a code conversion unit 407 that extracts a code from a bit string and converts it into a difference value. In the code conversion unit 407, 701 is, for example, 1
A buffer having a word-length capacity, and a page memory 10
The code data of one word read from 0 is temporarily stored. Then, the register 702 stores a bit string of a predetermined length over a plurality of words. The CPU 500 determines whether the bit string stored in the register 702 is insufficient in a process to be described later, if the register 702 has an empty space of one or more words, if the output code value does not satisfy the number of pixels for one row, or if the page memory If the write target pixel in 100 has not been reached, the data held in the buffer 701 is read.

In the code conversion section 407, the code data read from the page memory 100 is stored in the register 702 as a bit string, and the pattern is detected by the pattern detection section 705 detecting the pattern of the bit string in the register 702. . The pattern detection unit 705 cuts out a portion corresponding to the detected code pattern from the bit string on the register 702 to create an empty space, and outputs a difference value corresponding to the removed code pattern to the adder 401.

The pattern detector 705 outputs a control signal 415 every time a pattern of one pixel is detected. The control signal 415 gives the register 402 permission to update the integrated value and, at the same time, the register 304 of the address control unit 300.
Is also entered. The register 304 counts the number of cut-out pixels and checks whether the target pixel has been reached.
Here, the target pixel is a pixel at which writing to the page memory 100 is to be started.

The address control unit 300 is driven in response to the writing to the register 304, and the code conversion unit 416
The control signal 313 for controlling start / stop of the operation is enabled to start the density value calculation processing. When the register 304 determines that the target pixel has been reached, the control signal 314 is disabled and the operation of the density value calculation unit 400 is stopped. By monitoring the interrupt or control signal 313, the CPU 500 can grasp the operation status of the code conversion unit 407.

The code conversion section 407 has a mechanism for confirming the position of the bit to be processed in one word in order to recognize the start position of the code to be processed in one word read from the page memory 100. Hereinafter, a configuration for confirming a bit position in one word will be described.

The data read from the page memory 100 is continuously stored in a free area generated on the bit string of the register 702. By holding the read data for a plurality of words in the buffer 701, the bit position can be adjusted, and the data can be stored on the register 702 without any gap.

The counter 704 has a pattern detection unit 70
When the code length 710 recognized by 5 is input, the bit shift amount in the register 702 is counted. The CPU 500 obtains the bit shift position information 506 by the counter 704, and thereby obtains the page memory 1
The correspondence between the pixel to be written on 00 and the code position on the register 702 can be obtained.

The code length 71 is also stored in the address generator 703.
When 0 is input, the effective data amount on the register 702 is monitored, and the address request signal 416 is generated. When the address request signal 416 is input to the counter 301 in the address control unit 300, the counter 3
01 counts up and requests reading of the next address information on the page memory 100, that is, the next word length. The address generation unit 703 counts how many times a new address has been generated with respect to the initially set address value. The counted value 507 is referred to by the CPU 500 together with the bit shift position information 506 in the counter 704, and Used to recognize the code position of a pixel.

As described above, the code for each pixel read from the page memory 100 is converted by the code conversion unit 407 into each difference value. Then, the difference value obtained for each pixel is added to the density value of the previous pixel held in the register 402 by the adder 401 as in the first embodiment, and the density information 410 is restored.

Hereinafter, a process of partially updating the page memory 100 according to the second embodiment will be described.

With the configuration shown in FIG. 12, the pixel position (address) and the density value of the writing target in the page memory 100 are determined, and the CPU 500 determines the difference value of the target pixel so as to obtain the density to be changed. At the same time, it is necessary to recalculate and determine the difference value of the next pixel so that the density of the next pixel of the target pixel does not change. And
The obtained difference value between the target pixel and the next pixel is sign-converted and rewritten on the page memory 100.

When a continuous pixel row is rewritten in the page memory 100, the respective components are operated as described below as necessary.

For example, after the update start position on the page memory 100 is determined, the update end position is further determined. If the total sum of the new code lengths due to the update processing has changed from the existing sum total of the code lengths of the difference codes of the pixel to be updated and the next pixel, the CPU 500 causes the address generation unit 703 and the counter 704 to store the new code length on the page memory 100. , And the bit position in the word being processed, and the subsequent pixel row is bit-shifted.

Hereinafter, the updating process of the page memory 100 according to the second embodiment will be described with reference to the flowcharts of FIGS.

In FIGS. 13 and 14, the same processes as those in FIGS. 9 and 10 of the first embodiment are denoted by the same step numbers, and description thereof will be omitted. In the flowchart, the address of the page memory 100 is represented by X and Y coordinates, and a value having the same Y coordinate is treated as one pixel column. For each pixel, P is the old density information, Q is the new density information (updated density information), A is the word address (word address) on the page memory 100, and H is the difference between the pixel and the previous pixel. A sign bit string indicating a value, that is, a value actually stored in the page memory 100 is shown. BP indicates bit position information within one word in the register 702. Also,
In order to efficiently perform the pixel column updating process, a bit string buffer BT having a size sufficient to hold a code for one pixel column is prepared.

In the update processing shown in FIGS. 13 and 14, first, the update target object (Xa, Ya)-(Xb, Yb) is obtained in the same manner as in FIG. 9 described above, and the update range Xi1 on the i-th column is obtained.
~ Xik, the access preparation for the update target top pixel Xi1 is completed, and the address control unit 300 is driven (S10).
1 to S106, S801).

Then, the update density information Qi1 to Qik is converted by a difference value and encoded to obtain codes H'i2 to H'ik-1 other than the head and end to be newly stored, and from the head on the bit string buffer BT, It is stored after the maximum code length of one pixel and after BTcmax. That is, the difference value of the first pixel is the updated density information Q
Since it cannot be calculated only from the above, a space for the maximum code length is prepared at the head of the bit string buffer BT. And BT
, The next address where the last data is stored is set to BTe (S802). This process is efficient if the CPU 500 performs the process while the drive preparation process of the address control unit 300 is performed by hardware.

Then, the CPU 500 stores the page memory 10
The old density Pi1 of the first pixel Xi1 to be updated is obtained from the density information 410 read and restored from 0, and the CPU bus 502
The word address Ai1 of the page memory 100 is obtained from the counter 301 via the counter 301, and the bit shift position information 50
6 to obtain bit position information BPi1 (S803). The current code value Hi1 can be obtained from this BPi1. At this time, the bit string stored in the register 702 is the code value Hi1.
If it is not enough to obtain, the bit position is adjusted by the buffer 701. (S804).

Then, the original difference value Di1 is obtained from the current code value Hi1 (S805), Xik + 1 is output to the control signal 505, and the control signal 313 is activated (S806). Thereby, the access preparation of the next pixel Xik + 1 of the update target terminal pixel is completed, and the address control unit 300 is driven.

In the meantime, Di1 + Qi1-Pi1 is calculated and encoded as a difference value to be newly stored for the first pixel,
The encoding result is stored at the head of the bit string buffer BT. Note that the head code is stored in the BT so as to be continuous with the code stored after BTcmax. That is, when the code of the first pixel is less than the maximum code length, a free area corresponding to the difference between the maximum code length and the code length of the first pixel is generated at the BT head. Then, the effective data start position in the BT, that is, the storage address of the leading code is BTs.
(S807-S809). This indicates that the update data length on the bit string buffer BT can be obtained by BTs-BTe.

Next, the CPU 500 sets the page memory 100
The old density Pik + 1 of the terminal next pixel Xik + 1 to be updated is obtained from the density information 410 read out and restored from the CPU bus 5.
02, the word address Aik + 1 of the page memory 100 is obtained from the counter 301 through the counter 301, and the bit position information BPik + 1 is obtained from the bit shift position information 506 (S81).
1).

Then, in order to obtain the updated value of the last pixel to be updated, Qik−Pik + 1 is obtained and coded, then stored in the address BTe of the bit string buffer BT, and BTe is updated (S812).

Then, it is determined whether or not to shift the subsequent existing pixel row in accordance with the update of the data on the page memory 100. That is, the update data length on the line buffer is compared with the data length actually read from the page memory 100, and if they do not match, it is determined that a pixel column shift has occurred due to the update, and the subsequent pixel column moving process is performed. Specifically, assuming that the moving amount of the subsequent bit string is BMv, BMv can be expressed by the following equation.

BMv = (BTe−BTs) − ((Aik + 1 +
1)-Ai1) x word length-(BPik + 1-BPi1) If the BMv is "0", no deviation occurs, and the data of BTs to BTe on the bit string buffer BT is directly stored on the page memory 100. The data is transferred to the bit position BPi1 of the word address Ai1, but if it is not "0", it is determined that a shift has occurred, and the bit string after the code string Hik + 2 of the page memory 100 is shifted by BMv, and the bit string buffer B
The data of BTs to BTe on T is transferred to bit position BPi1 of address Ai1 on page memory 100 (S813).
To S816).

Thus, the updating process of the Yi column is completed, and the next Yi column is updated.
The processing for the +1 column is started (S817).

As described above, according to the flowcharts of FIGS. 13 and 14, continuous rewriting on the pixel column to be updated in the page memory 100 can be performed collectively. As a result, even if it is necessary to move the existing bit string information in the page memory 100, the movement processing is performed only once for each pixel row as an update unit, which is far less than the processing in pixel units. Only the number of processes is required.

The operations in the configuration shown in FIG. 12, including the processes shown in the flowcharts in FIGS. 13 and 14, are controlled according to a control program executed by the CPU 500.

As described above, according to the present embodiment, when multi-valued image data is encoded and stored in the page memory 100, the smaller the number of bits, the higher the frequency of occurrence of the difference value from the previous pixel. By allocating a number of codes, it is possible to perform compression with higher memory saving effect.

Further, the restoration and updating of the image data held in the page memory 100 can be performed at high speed.

In the second embodiment, an example has been described in which a code is fixedly assigned according to the appearance frequency of a difference value for each pixel of image data. Normally, the appearance frequency of the difference value is slightly biased depending on the image. Therefore, by replacing the code length assigned according to the image, more efficient compression can be realized. However, since a code range having a high frequency of appearance differs for each image, it is necessary to dynamically assign a code to each image to be processed. In particular, when the page memory 100 is frequently rewritten, the appearance frequency of each density changes frequently, and the appearance frequency is not stable. Therefore, it is not practical to change the code assignment every time the page memory 100 is rewritten and store the code again.

<Third Embodiment> Hereinafter, a third embodiment according to the present invention will be described.
An embodiment will be described.

In the above-described second embodiment, since the code length is completely variable, it is necessary to perform code conversion and difference value accumulation processing from the end of the image when referring to the code stored in the page memory 100. there were. Therefore, even if there is support by hardware, processing takes longer than the method of storing a simple difference value as shown in the first embodiment.

Therefore, in the third embodiment, the page memory 1
00 is characterized in that the density information itself of the image data is discretely held in addition to the codes shown in the second embodiment.

That is, in the third embodiment, one image is conceptually divided to increase the number of image edges, and the number of pixels to be processed at one time in the pixel column direction (main scanning direction) is reduced. . FIG. 15 shows a conceptual diagram thereof. In FIG. 15, by adding three density points to each pixel column, the page memory 110 logically has dotted lines 91, 92,
An example in which the image is divided into four by 93 is shown. Hereinafter, the density point is referred to as a discrete density point, and the area logically divided is referred to as a divided area. That is, as the configuration of the page memory 100, a plurality of logical page memories (four in FIG. 15) in which the pixel column direction is shortened exist as divided areas.

In the third embodiment, by logically dividing the page memory 100 in this way, software can divide the processing target image for each divided region as pre-processing, or perform restoration when density information is restored. It is sequentially checked whether the density value matches the density value indicated by the discrete density point, and if not, processing for appropriately changing either the read density information or the stored density information is added according to the processing. There is a need.

For example, when an image to be processed is divided for each divided area by software as pre-processing, the hardware may have the same configuration as that shown in FIG. 12 of the second embodiment. However, since the storage position itself in the page memory 100 is also physically divided, when the restored density information 410 is output, it is necessary for the code conversion unit 407 to set an address a plurality of times for each pixel column.

On the other hand, in the case of performing the successive comparison at the time of restoring the density information, the pixel rows are continuous on the page memory 100. The start address only needs to be set once. However, when the page memory 100 is partially updated, information such as the density, address, coordinates, and bit position of intermediate pixels needs to be set in the code conversion unit 407. Therefore, in the configuration shown in FIG. It is necessary to make the information 506, the address count value 507, and the like bidirectional.

Hereinafter, the update processing of the page memory 100 in the third embodiment will be described with reference to the flowcharts of FIGS.

In FIGS. 16 and 17, the same processes as those in FIGS. 13 and 14 of the second embodiment are denoted by the same step numbers, and description thereof is omitted. In the flowchart, a method of simply dividing the object to be updated in advance is shown.

In the flowchart, the address of the page memory 100 is represented by X and Y coordinates,
A value having the same coordinates is treated as one pixel row. For each pixel, P is the old density information, Q is the new density information (updated density information), A is the word address (word address) on the page memory 100, and H is the difference between the pixel and the previous pixel. A sign bit string indicating a value, that is, a value actually stored in the page memory 100 is shown. BP indicates bit position information within one word in the register 702. In addition, in order to efficiently perform the pixel row updating process, a bit string buffer BT having a size sufficient to hold a code for one pixel row is prepared.

In the flowchart, the case of having three discrete density points as shown in FIG. 15 will be described, and the coordinates of the discrete density points are set to S0 to S3 including the head address of the pixel row. That is, S0 is the original image end. Also, let S4 be the end coordinate of the pixel row. Further, the start point and the end point in the updated pixel sequence in each divided region are set to Ws and We, respectively. In addition, the start and end positions of each divided area are indicated by subscripts s and e.

In the update processing shown in FIGS. 16 and 17, first, the update target object (Xa, Ya)-(Xb, Yb) is obtained in the same manner as in FIG.
1 to Xik are obtained (S101 to S104).

Then, the division area number j is initialized to "1" (S901), and it is determined whether or not even a part of the update range Xi1 to Xik exists in the division area j (S902). If the update range is not within the divided area, the processing is started for the next divided area (S927, S928).

If the update range is within the divided region, the start position Xi1 of the update range is compared with the start of the divided region Sj-1, and the update is actually performed on the page memory 100. Set to Ws. That is, if the start position Xi1 of the update range is before the division region head Sj-1, Sj-1 is the update pixel column head Ws.
s (S903 to S905).

Next, similarly, the end position Xik of the update range is compared with the end of the divided area Sj, and set to the updated pixel column end We in the divided area. That is, if the end position Xik of the update range is before the end of the divided region Sj, Xik is the end of the updated pixel column We, and if it is after, the end of Sj is We (S906 to S906).
S908).

With the above processing, the update range Ws-We in the divided area is determined. And subsequent steps S
In steps 909 to S926, FIG.
The processing of steps S106 to S816 shown in FIG. 14 is similarly performed for the update ranges Ws to We in the divided area.

Then, in step S927, the division area number j is incremented, and the above-described pixel row update processing in units of division areas is repeated until the update processing is completed for all the division areas (in this case, four areas) (S928). Then, when the update process for all the divided regions is completed, the process for the next pixel row is started (S929).

As described above, the divided area processing in the third embodiment is realized.

In order to realize the divided area processing, in the configuration shown in FIG. 12 of the second embodiment, the start position of each divided area on the page memory 100, specifically, the address information and the bit position Start information A new memory area is required to store the address information and the bit position start information.

In the above flowchart, the method of dividing the object to be updated in advance has been described. However, when performing the successive approximation processing at the time of restoring the density information, in the configuration shown in FIG. Further requires a memory area for storing the density information indicated by.

However, when the page memory 100 is physically divided, the bit position information and the discrete density information are
Omitted as a hardware-dependent initial value.

As described above, according to the third embodiment,
When a difference value of image data is stored in a page memory as a variable-length code corresponding to the frequency of appearance, a predetermined number of pieces of density information are stored discretely, thereby particularly improving the processing speed in a page memory update process. be able to.

In the first to third embodiments described above, an example has been described in which the primary density difference of the image data is encoded and stored in the page memory 100. However, the order in the density difference calculation of the present invention is not limited to the first order, and may be any order.

In general, as the order increases, the appearance frequency of the n-th difference value deviates, so that a higher memory saving effect can be obtained by code replacement. However, when the order in the difference calculation increases, the processing system becomes complicated. For example, in the update process of the page memory 100 in the case of the primary difference value as described above, the difference value between the pixel to be updated and the next pixel may be adjusted. This affects the next adjacent pixel. In an image having a poor S / N ratio,
Even if the order of the difference is increased, the effect of compression at the time of encoding cannot be obtained so much. Therefore, in each embodiment, only an example of the primary difference is shown.

In the first to third embodiments described above, an example has been described in which the multi-valued image data held in the page memory 100 is input from an external device via the input I / F 1013. A multi-valued image data may be input by optically reading a document image.

In the above description, the configuration having one page memory 100 has been described as an example. However, by providing the page memory 100 for each color, the present invention can be applied to color image processing.

<Fourth Embodiment> Hereinafter, a fourth embodiment according to the present invention will be described.
An embodiment will be described.

In the above-described first to third embodiments, an example has been described in which the difference between the input multi-valued image and the previous pixel is encoded and stored in the page memory 100 for each pixel. This example can be applied not only to a monochrome image processing apparatus but also to a color image processing apparatus having a plurality of page memories and performing image formation in a frame-sequential manner.

However, in the first to third embodiments, the input I / F from the external host computer is
Since multi-value image data is received via 1013,
In particular, when color image data is received, there is a problem that the transfer time becomes long. Therefore, in the fourth embodiment, when color image data is transferred from an external host computer, the amount of transfer data can be reduced by encoding the transfer data and the transfer time can be further reduced. Will be described.

First, the basic concept of encoding the transfer data in the fourth embodiment will be described.

In a normal natural object, an image obtained by taking it in, or an image recognizable by humans, the color change is relatively gradual even though the brightness may change drastically. That is, in the image data normally handled, the color axis fluctuation is more gradual than the density axis fluctuation between pixels. Generally, in data having a gradual change, the appearance rate of the difference value between adjacent data concentrates on a small difference value. Therefore, when a code having a shorter code length is assigned to each difference value having a higher appearance frequency as shown in the second embodiment, high compression efficiency can be obtained.

Therefore, in the fourth embodiment, a data transfer method with a higher compression effect is realized by utilizing the feature that image data normally handled by humans has a higher color correlation than a density correlation between neighboring pixels. It is characterized by the following.

However, an ordinary color image processing apparatus is a primary color separation system such as CMYK that uses a plurality of primary color dyes. Since such a primary color separation system has image data for each single color, the density information in a single color has low correlation between neighboring pixels as described above. That is, if the primary color separation system is used as it is, it is not possible to efficiently utilize the high color correlation with respect to the density correlation. In order to efficiently use the color correlation between adjacent pixels, it is necessary to convert image data of a primary color separation system into a logical color system such as color difference and saturation, which is directly connected to a human recognition system. . However, since the printing mechanism in the image processing apparatus is usually a primary color separation system, performing such conversion is inconvenient in terms of cost and processing speed.

Therefore, in the fourth embodiment, data with an increased color correlation is generated at high speed without converting full-color image data into color difference, saturation, and the like. Hereinafter, a specific description will be given.

That is, by taking the difference between each primary color in the primary color separation system, the density element included in each primary color information is canceled. As a result, it is possible to leave elements with gentle fluctuations in hue, saturation, and the like, and to perform conversion into information having a high compression effect, thereby achieving efficient data transfer.

Although the difference data between the primary colors does not indicate accurate hue and saturation information, the difference data can be generated in real time by a simple addition / subtraction mechanism, and the density information can be obtained. Can be accurately restored.

Hereinafter, encoding and decoding in the fourth embodiment will be described more specifically.

FIG. 18 shows a color L according to the fourth embodiment.
1 shows a configuration of a BP (laser beam printer). In FIG. 18, a transfer sheet P fed from a sheet feeding unit 2101 is held on the outer periphery of the transfer drum 2103 with a leading end held by a gripper 2103f via a conveyance path 2102. The latent images formed for each color on the photosensitive drum 2100 by the optical unit 2107 are stored in the respective color developing units Dy, Dc,
The image is developed by Db and Dm and is transferred to the paper 2102 on the outer periphery of the transfer drum 2103 a plurality of times to form a multicolor image. After that, the transfer paper P is separated from the separation claw 2113.
Then, the sheet is separated from the transfer drum 2103 by the fixing unit 2104, and is discharged to a discharge tray 2115.

The developing units Dy, Dm, Dc, Db of the respective colors are:
Each end has a rotating shaft, and each is held by the developing device selection mechanism 2108 so as to be rotatable about the shaft. Each color developing device Dy, Dm, Dc, Db is rotated for selecting a developing device while keeping its posture constant. And after the selected developing device moves to the developing position,
The developing device selecting mechanism 2108 is provided with the selecting mechanism holding frame 21.
09 and the fulcrum 210 by the solenoid 2109a.
The moving position is determined in the direction of the photosensitive drum 2100 around 9b.

Next, the operation of the color LBP having the above configuration will be specifically described.

Reference numeral 2105 denotes an operation panel on which various switches for operation and an LED display are arranged. Reference numeral 2114 denotes an input I / F for inputting image data from an externally connected host computer or the like. 2130 is LBP2
000 is a printer control unit for restoring the compressed image data transferred via the input I / F 2114 and control of the whole. The printer control unit 2130 generates laser drive signals on a plane-by-plane basis based on image data for each color, and outputs the laser drive signals to the laser driver 2106. Thereby, the laser driver 2106 emits the laser beam L according to the image signal.

On the other hand, the photosensitive drum 2100 is uniformly charged to a predetermined polarity by the charger 2111, and a latent image of, for example, Y (yellow) color is developed on the photosensitive drum 2100 by exposure with the laser beam L. The first toner image of Y color is formed on the photosensitive drum 2100 by development by the container Dy. On the other hand, the transfer paper P
Is fed, a transfer bias voltage having a polarity opposite to that of the toner is applied to the transfer drum 2103, the first toner image on the photosensitive drum 2100 is transferred to the transfer paper P, and the transfer paper P is transferred to the surface of the transfer drum 2103. Is electrostatically attracted. Thereafter, the remaining Y toner is removed from the photosensitive drum 2100 by the cleaner 112, and the photosensitive drum 2100 prepares for the next color latent image formation and development process.

Next, an M (magenta) second latent image is formed on the photosensitive drum 2100 by the laser beam L, and then the photosensitive drum 2
The second latent image on 100 is developed to form a second toner image of M color. The M-color second toner image is transferred onto the transfer paper P in accordance with the position of the Y-color first toner image previously transferred onto the transfer paper P. In the transfer of the second color toner image, a bias voltage higher than that during the transfer of the first toner image is applied to the transfer drum 2103 immediately before the transfer paper reaches the transfer portion. Similarly, third and fourth latent images of C (cyan) color and K (black) color are sequentially formed on the photosensitive drum 2100, and are respectively sequentially developed by the developing units Dc and Db. Are aligned with the previously transferred toner image,
The fourth toner images are sequentially transferred. As described above, four color toner images are formed on the transfer paper P in an overlapping state. In the transfer of the third and fourth color toner images, a bias voltage higher than that at the time of transfer of the second toner image is applied to the transfer drum 2103 immediately before the transfer paper reaches the transfer portion.

Thereafter, when the leading end of the transfer paper P on which the toner images of the four colors are superimposed and transferred approaches the separation position, the separation claw 211 is moved.
3, the leading edge thereof comes into contact with the surface of the transfer drum 2103, and the transfer paper P is separated from the transfer drum 2103.
The tip of the separation claw 2113 keeps contact with the surface of the transfer drum, and then separates from the transfer drum 2103 and returns to the original position.

Then, the separated transfer paper P is fed to the fixing device 21.
04, where the toner image on the transfer paper is fixed and discharged onto the discharge tray 2115. The above is the printing process in the color LBP to which the fourth embodiment is applied.

In the fourth embodiment, full-color image data is sequentially read out of the LBP 20 by the host computer.
00 is assumed. At this time, the image data of the first color is transferred as it is, and when transferring the image data of the second and subsequent colors, the density difference information for the image data of the previous color is calculated, encoded and transmitted.
Of course, the codes of the second and subsequent colors received on the LBP2000 side are restored by decoding and integrating.

In the image processing system configuration of the fourth embodiment, the above-described LBP2000 is connected to a host computer, and the host computer uses the LBP2000.
The function of the printer driver at the time of data transfer with the 2000 is the printer driver. FIG. 1 shows a block configuration of the image processing system.
9

In FIG. 19, reference numeral 3000 denotes a printer driver in the host computer, which controls the transfer of full-color multivalued image data generated in the host computer to the LBP2000. Note that the printer driver 3000 sends out data when a reception request is issued from the LBP 2000 side.

In the printer driver 3000, 60
A color separation unit 1 separates the input full-color multivalued image data into color components that can be processed by the LBP 1000. In the fourth embodiment, it is assumed that the light is separated into four colors of Y (yellow), M (magenta), C (cyan), and K (black). Then, a difference value calculator 602 calculates a difference value between the colors according to the order of transmission. After the encoding unit 603 assigns a code having an appropriate code length according to the frequency of appearance of the difference value to the difference value, the difference unit is subjected to transfer encoding used during normal data transfer. The encoded image data is temporarily stored in the image memory 604.

In the difference value calculating section 602 and the coding section 603, as described above, only the transfer coding of the original density information is performed without performing the difference calculation for the first color, and the difference coding is performed for the second and subsequent colors. A value is calculated, and the difference value is encoded and transfer-encoded. For example, in the fourth embodiment, if image data is transferred in the order of Y, M, C, and K, the density value of the Y image data is transferred and output, and then the M image data and the Y image The difference (M−Y) for each pixel from the data is calculated, encoded, and transfer encoded. Then, similarly, the differences CM and KC are sequentially calculated and coded. As a method of encoding the difference value in the encoding unit 603, for example, the code assignment as described in the above-described first to third embodiments may be performed.

The coded data held in the image memory 604 as described above is sent from the printer driver 3000 to the LBP 2000 via the output I / F 605 to the Y, M-
It is transmitted in the order of Y, CM, and KC.

In the printer driver 3000, 6
Reference numeral 06 denotes a CPU, which stores components in the driver in a ROM 60
7 is controlled overall according to the control program stored in the control program 7. Reference numeral 608 denotes a RAM, which is used as a work area of the CPU 606.

Next, the printer control unit 213 in the LBP2000 for receiving the image data as described above.
0 is shown in FIG. 20 and will be described.

In FIG. 20, reference numeral 2140 denotes a printer unit, which includes a configuration for performing image forming processing after the laser driver 2106 in the configuration shown in FIG.

In the printer control unit 2130, reference numeral 810 denotes an image memory for storing the transferred image data, and has a capacity enough to store gradation image data of a single color. Reference numeral 804 denotes a printer unit 214
0 according to the timing signal generated by the image memory 81.
The print sequencer controls output of image data on the printer unit 2140 to the printer unit 2140. Reference numeral 801 denotes a reception buffer for temporarily storing received data, and 802 denotes an expansion unit for expanding received data compressed by a normal method associated with transfer. Reference numeral 803 denotes a color restoration unit,
The density information of the next color is restored by performing an operation based on the difference value information obtained in step 2 and the density value information of the corresponding pixel on the image memory 810, and written back to the image memory 810.

Reference numeral 800 denotes a printer control unit 213.
0 is a system sequencer that controls the entire operation. The system sequencer 800 stops the operation of the expansion unit color restoration unit 803 when the first color data is received, and controls the density data simply expanded by the expansion unit 802 to be written to the image memory 810. . When the image data of the second and subsequent colors is received, the decompression unit 802
, The difference information is output, so that the color restoration unit 803 is controlled to operate.

Further, the system sequencer 800 controls the amount of data received from the printer driver 3000, that is, a reception request to the printer driver 3000, based on information such as the usage rate of the reception buffer 801 and the amount of paper movement in the printer unit 2140. For example, the reception buffer 80
When 1 is full, the reception request is stopped, and when there is free space in the reception buffer 801, the reception request is sent.

In FIG. 20, reference numeral 210 denotes image data transferred from the printer driver 3000.
11 is a control signal for a reception request, 212 is a reception buffer 80
Reference numeral 213 denotes a control signal for monitoring the amount of data in 1, reference numeral 213 denotes an operation control signal of the color restoration unit 803, and reference numeral 214 denotes address information of an image memory output from the print sequencer 804. Also, thick arrows in the figure indicate the flow of image data.

Next, the color restoration processing in the color restoration unit 803 will be described. In the image memory 810, the image data already in the memory area that has been sent to the printer unit 2140 is added to the difference value sequentially input from the decompression unit 802 by the color restoration unit 803 to obtain the density of the next color. The data is restored. For example, in the fourth embodiment, first, the Y color information is stored in the image memory 81 as the first color information.
Expands to zero. Then, the Y color information is sequentially output to the printer unit 2140, and at the same time, the color restoration unit 803 uses the MY difference data to replace the Y color data of the already output area with the M color data. As described above, the Y color data on the image memory 810 is stored in the printer unit 2.
Since the data is sequentially replaced with M color data in accordance with the output to 140, the capacity of the image memory 810 may be one color.

The target area for the next color data restoration in the color restoration section 803 is stored in the image memory 810 from the printer section 21.
If it is almost impossible to catch up with the transmission area to 40, the system sequencer 800 needs to stop the calculation and write-back processing of the difference data and the previous color density data in the color restoration unit 803 to prevent overtaking. For this purpose, for the image area held in the image memory 810, an address designation pointer (sending point pointer) indicating a sending point to the printer unit 2140, and an address designation of the next color restoring point in the color restoring unit 803. Pointer (restoration point pointer)
Is required. Since the sending point pointer is inevitably provided in the print sequencer 804, a reading means for reading the image memory 810 according to the sending point pointer may be prepared. Similarly, a restoration point pointer is necessarily provided in the color restoration unit.

Here, an image transmission procedure in the printer driver 3000 is shown in the flowchart of FIG. 21 and will be described below.

The color image data generated by the host computer is color-separated by the color separation unit 601, and the difference value calculation unit 602 calculates a difference value between colors according to the color order to be transmitted next (S 201 to S 203). ). Then, the calculated color difference value is further compressed by the encoding unit 603 (S204), and then sent to the LBP2000 (S205 to S207).

Since the LBP2000 needs image data in the order of Y, M, C, and K as described above, the printer driver 3000 sends the compressed data of Y color and the density difference of MY from the image memory 604. Value compressed data,
The compressed data of the CM density difference value and the compressed data of the KC density difference value are transmitted in this order.

The format of the transfer encoding in the encoding unit 603 is as follows: the input I / F unit 211 on the LBP2000 side.
For convenience of controlling the flow of received data in step 4, it is desirable to use a format that can be partially decompressed sequentially. When a compression format that cannot partially restore data is employed, one page of image data is further divided into small areas, and compression of each small area is not performed. It will be difficult.

In the fourth embodiment, an example has been described in which density information is directly transferred and coded for the first color data to be transferred. By assuming the 0th color data having a pixel value of “0”, image data of all color components can be represented by difference information for each color. As a result, the amount of transfer data can be further reduced. In this case, if the image memory 810 is cleared to 0 at the start of the transfer on the LBP side, it is not necessary to control the operation of the color restoration unit 803 according to the order of the color data.

As described above, according to the fourth embodiment,
In an image processing system that performs frame sequential processing of a color image, a high compression rate can be obtained by encoding and transferring a difference value between each color, that is, a transfer data amount can be reduced and a transfer time can be reduced. Is done.

Further, when developing the differential data transferred on the printer side, it is only necessary to prepare a memory capacity for one color.

<Fifth Embodiment> Hereinafter, a fifth embodiment according to the present invention will be described.
An embodiment will be described.

In the fifth embodiment, an example will be described in which the operation of the printer control unit 2130 in the above-described fourth embodiment is realized mainly by software.

The image processing apparatus in the image processing system according to the fifth embodiment employs the LBP2000 shown in FIG.
FIG. 22 shows a block configuration of a printer control unit 2130 in the fifth embodiment.
In the LBP2000 of the fifth embodiment, it is assumed that a drawing command and a character code received from the host computer can be developed and output.

Referring to FIG. 22, reference numeral 550 denotes a CPU.
Printer control unit 2 shown in the fourth embodiment described above
The operation in 130 is executed according to a control program stored in ROM 650. Note that the CPU 550 may include various hardware for the purpose of improving performance. A RAM 750 is used as a work area of the CPU 550, for example, as the image memory 810 described in the fourth embodiment. A bus line 805 connects the CPU 550 and other components.

Although the configuration shown in FIG. 22 is intended to realize the operation in the printer control unit 2130 only by software, it is needless to say that software processing is performed by replacing time-consuming processing with appropriate hardware. , Faster processing becomes possible.

The image which can be processed by the LBP2000 of the fifth embodiment and is generated by a drawing command such as a character or graphics has a small data size, and it is difficult to predict a generated color, and the image is generated over a plurality of colors. affect. Therefore, in the method of calculating the difference value for each color as described in the above-described fourth embodiment, it is difficult to clearly reproduce an image formed based on a drawing command. In the fifth embodiment, an example in which these drawing commands and simple image data are processed separately will be described.

First, in the printer driver 3000 on the host computer side, it is necessary to separate simple image data and various drawing commands, and control the transmission order. Hereinafter, the control method will be described.

As a first method of separating the simple image data and the drawing command in the printer driver 3000 and sending them, there is a method of separating the drawing commands for each color and sending the drawing commands for each color. In this method,
For each color, it is necessary to check the overlap between the drawing range affected by the drawing command to be sent and the simple image area. If there is an overlap, it is determined which is ultimately overwritten and remains on the output image, and the remaining is sorted so that it is later and then sent.

In this method, that is, for each color: Drawing command partially overwritten by image data Image data 3. Sent in the order of the drawing commands that overwrite part of the image data. The drawing command that does not interfere with the image data may be sent after the image data or before it.

According to this method, one drawing command is transmitted for, for example, each of four colors, so that the transfer amount of the drawing command is quadrupled. However, since the data size of the drawing command is generally smaller than the image data size, the compression effect of the image data is higher in a processing system which mainly handles the image data, and the data amount of the drawing command is increased. Can be ignored. However, LBP2
In an image processing apparatus based on the electrophotographic method such as 000, once the image forming operation is started, the development of all the colors must be performed within a specified time, so that the processing of the drawing command can be executed at a sufficiently high speed. Some processing performance is required.

As a second method of transmitting the simple image data and the drawing command separately in the printer driver 3000, there is a method of transmitting all the drawing commands first, and then transmitting the image data for each color. In this case, a flag according to the degree of interference between the drawing area and the image data is added to all drawing commands. That is, LB
On the P2000 side, it is determined based on the flag whether the received drawing command is developed before processing the image data or after processing.

In this method, since a drawing command can be obtained in advance in the LBP2000, for example, even the last drawing command to be drawn is converted in advance to a simple format that can be processed at high speed by the printer unit 2140. It becomes possible. Therefore, even when the processing performance of the CPU 550 is relatively low with respect to the printer unit 2140 at a constant transport speed, image data output to the printer unit 2140 can be completed in time.

In both the first and second methods described above, it is necessary to detect the overlap between the image area of the image data and the image area by the drawing command. Hereinafter, a method of determining the overlapping area will be described.

The simplest determination method is to calculate the maximum value and the minimum value of the coordinates in which each image area is included when both are drawn and developed, and to calculate the area group based on the image data and the image data based on the drawing command. There is a method of performing a comparison with a region group. That is, assuming that a two-dimensional space in which image data is drawn is represented by x and y coordinates, if there is a common range in both the x and y directions in both image areas, that is, if the inclusion areas overlap, , There is a possibility that both areas overlap. On the other hand, when there is no common value in either the x direction or the y direction, it is determined that there is no possibility of overlapping. In the case of processing a drawing command such as a hatched drawing command in which most of its included region does not actually overlap with image data, a more accurate and efficient overlap determination is performed. To do
It is preferable to prepare an individual determination routine for each drawing command.

The flowcharts of FIGS. 23 and 24 show the printer driver 300 for implementing the second method described above, that is, a processing system for transmitting a drawing command before image data.
The procedure of image generation and data transmission on the 0 side is shown.

In the flowchart, the drawing objects output from the application in the host computer are classified into image data and drawing commands while the output order information is stored in the printer driver 3000. Then, a flag corresponding to the degree of interference between the drawing area and the image data is added.
The processing shown in the flowchart is stored in the ROM 607 in the printer driver 3000 as a control program, and is executed by the CPU 606.

First, a variable i indicating the order of objects is initialized to “0” (S301). Then, it is determined whether or not the object i is image data, and the object i is stored in the RAM 6 so that identification according to the determination is possible.
08 in a predetermined area of the working memory (S30).
2 to S304). Hereinafter, an object determined to be image data is referred to as an image object, and an object determined to be a drawing command is referred to as a command object. The image memory 60 is used as a temporary storage memory.
4 may be used.

Then, the variable i is incremented, and if there are no more objects to be processed, the subsequent flag addition processing is started (S305, S306).

Hereinafter, the flag adding process will be described.
First, after the variable i is initialized to “0” again (S307), if the object i is a command object,
It is determined whether or not the drawing area of the command object overlaps with the drawing area of each image object (S308, S309). If there is an overlap, the order j of the duplicated image object is compared with the order i of the command object. If the command object is output later (i> j), the command object i A "prefix flag" indicating that the drawing command is developed after the image data, that is, in the formed image, that the drawing by the drawing command is placed on the front side than the drawing by the image data is added. On the other hand, if the command object is output first, the drawing command is expanded before the image data for the command object i, that is, the drawing by the drawing command is performed in the formed image. A “postfix flag” indicating that the image is to be placed on the rear side of the drawing by the image data is added (S310 to S31)
4). If there is no overlap, no flag is added to the command object i, and the drawing order is arbitrary.

Then, the variable i is incremented, and the flag adding process is performed on all command objects (S315, S316).

Then, all the drawing commands subjected to the flag addition processing by the above processing are transmitted to the LBP2000 side (S317 to S319).

When transmission of the drawing command is completed, next,
After all the image data is subjected to difference compression for each color in the same manner as the procedure shown in FIG.
The message is transmitted to P2000 (S320 to S325).

As described above, the drawing command group to which the flag is added and the image data group are sent from the printer driver 3000 to the LBP 2000.

In the LBP2000, especially when the printer unit 2140 performs low-speed paper conveyance, the drawing command received before the image data in the printer control unit 2130 is delayed by a delay time with respect to the operation of the printer unit 2140. Is developed in advance to a simple form that does not cause the

Then, after drawing the drawing command to which the post-flag is added, image data is received and expanded, and thereafter, the drawing command to which the pre-flag is added is drawn. As a result, an image of each color is formed in the printer control unit 2130, and is output to the printer unit 2140 on a color-by-color basis.

For image data in an area overwritten by a drawing command to which a prefix flag has been added, the original color density value cannot be faithfully restored based on the difference value for each color. Since each is overwritten, there is no problem in the actually formed color image even if the density value faithful to the original image data is not restored. However, a density command may be rewritten before receiving a difference value for restoring a next color by a drawing command to which a postfix flag has been added, and a correct color may not be expressed. Accordingly, when processing a drawing command to which a post-fix flag is added, clipping processing is performed for the second and subsequent colors, and control is performed so that writing by the drawing command is not performed in a drawing area based on image data. There is a need.

As explained above, according to the fifth embodiment,
In an image processing system that performs frame-sequential processing of a color image, a drawing command such as characters and graphics is separated from simple image data and processed, so that the drawing data information based on the drawing command is kept clear and the fourth embodiment is performed. Image data transfer with a high compression ratio similar to that of the embodiment can be realized.

In the above-described first to third embodiments, an example in which code data is held in the page memory in the image processing apparatus will be described. In the fourth and fifth embodiments, a number of operations between an external device and the image processing apparatus will be described. The example of transferring the value image data has been described. The present invention can, of course, be realized by combining these embodiments. For example, the code based on the difference value for each color of the multi-valued color image data may be stored in the page memory in the image processing apparatus, or the multi-valued image data may be stored in the image processing system by the difference value from the previous pixel. Information may be encoded and transferred.

The present invention is not limited to the method of encoding the difference value from the previous pixel used in the above embodiments. For example, the encoding method may be performed by using a plurality of adjacent peripheral pixels. A known peripheral prediction method may be used.

[0262]

[Other Embodiments] Even if the present invention is applied to a system including a plurality of devices (for example, a host computer, an interface device, a reader, a printer, etc.), an apparatus (for example, a copying machine) Machine, facsimile machine, etc.).

Further, an object of the present invention is to provide a storage medium storing a program code of software for realizing the functions of the above-described embodiments to a system or an apparatus, and to provide a computer (or CPU) of the system or apparatus.
And MPU) read and execute the program code stored in the storage medium.

In this case, the program code itself read from the storage medium implements the functions of the above-described embodiment, and the storage medium storing the program code constitutes the present invention.

As a storage medium for supplying the program code, for example, a floppy disk, hard disk, optical disk, magneto-optical disk, CD-ROM, CD
-R, a magnetic tape, a nonvolatile memory card, a ROM, or the like can be used.

When the computer executes the readout program codes, not only the functions of the above-described embodiments are realized, but also the OS (Operating System) running on the computer based on the instructions of the program codes. ) May perform some or all of the actual processing, and the processing may realize the functions of the above-described embodiments.

Further, after the program code read from the storage medium is written into the memory provided in the function expansion card inserted into the computer or the function expansion unit connected to the computer, based on the instruction of the program code, It goes without saying that a CPU or the like provided in the function expansion card or the function expansion unit performs part or all of the actual processing, and the processing realizes the functions of the above-described embodiments.

[0268]

As described above, according to the present invention, the difference information from the previous pixel is encoded and stored for each pixel of the multi-valued image information, so that the memory saving effect is high and the restoration can be performed at high speed. Compression is possible. Therefore, it is possible to provide an image processing apparatus capable of forming a low-cost and high-quality image at a high speed.

Further, in the image processing system for transferring color image data, by encoding and transferring the difference value between each color, the data transfer amount can be reduced and the transfer time can be shortened.

[Brief description of the drawings]

FIG. 1 is a side sectional view of an image processing apparatus (LBP) according to an embodiment of the present invention.

FIG. 2 is a block diagram showing a basic configuration for realizing compression processing in the embodiment.

FIG. 3 is a block diagram illustrating a detailed configuration of a density value calculation unit and an address control unit when density information is restored from a page memory in the embodiment.

FIG. 4 is a block diagram illustrating a configuration for realizing a page memory update process according to the embodiment;

FIG. 5 is a flowchart of address control in a page memory update process according to the embodiment.

FIG. 6 is a diagram illustrating an example of a code table that defines code assignment in the present embodiment.

FIG. 7 is a block diagram illustrating a configuration for restoring and outputting density information based on code assignment according to the present embodiment.

FIG. 8 is a flowchart showing sign extension control in the sequencer of the embodiment.

FIG. 9 is a flowchart of address control in a page memory update process according to the embodiment.

FIG. 10 is a flowchart of address control in a page memory update process according to the embodiment.

FIG. 11 is a diagram illustrating an example of a code table that defines code assignment according to the second embodiment of the present invention.

FIG. 12 is a block diagram illustrating a detailed configuration of an address control unit and a density value calculation unit according to a second embodiment.

FIG. 13 is a flowchart of address control in a page memory update process according to the second embodiment.

FIG. 14 is a flowchart of address control in a page memory update process according to the second embodiment.

FIG. 15 is a diagram showing a concept of memory division in a third embodiment according to the present invention.

FIG. 16 is a flowchart of address control in a page memory update process according to the third embodiment.

FIG. 17 is a flowchart of address control in a page memory update process according to the third embodiment.

FIG. 18 is a side sectional view of an image processing apparatus (LBP) according to a fourth embodiment of the present invention.

FIG. 19 is a diagram illustrating a block configuration of an image processing system according to a fourth embodiment.

FIG. 20 is a diagram illustrating a detailed configuration of a printer control unit 2130 according to the fourth embodiment.

FIG. 21 is a flowchart illustrating an image transmission procedure in a printer driver 3000 according to the fourth embodiment.

FIG. 22 is a block diagram illustrating a detailed configuration of a printer control unit 2130 according to a fifth embodiment of the invention.

FIG. 23 is a flowchart illustrating a procedure of image generation and data transmission when a drawing command is transmitted before image data in the fifth embodiment.

FIG. 24 is a flowchart illustrating a procedure of image generation and data transmission when a drawing command is transmitted before image data in the fifth embodiment.

[Explanation of symbols]

 100 page memory 101 storage capacity change unit 200 difference value calculation unit 300 address control unit 301 counter 302 delay unit 303 flip-flop (F / F) 304, 402 register 305, 406 AND gate 400 density value calculation unit 401 adder 403 sequencer 404 , 405 Latch 500 CPU 601 Color separation unit 602 Difference value calculation unit 603 Encoding unit 604,810 Image memory 800 System sequencer 801 Receive buffer 802 Decompression unit 803 Color restoration unit 804 Print sequencer 2000 Laser beam printer (LBP) 2130 Printer control unit 2140 Printer section 3000 Printer driver

Claims (36)

[Claims] An input unit for inputting multi-valued image data; an encoding unit for encoding the multi-valued image data input by the input unit, and a difference value between the multi-valued image data and a predicted value thereof; Holding means for holding the code data according to the above, decoding means for decoding the code data held by the holding means to restore multi-valued image data, and forming the multi-valued image data restored by the restoration means into a predetermined image. An image processing apparatus, comprising: output means for outputting to an output means. 2. The method according to claim 1, wherein the encoding unit encodes, for each pixel of the multi-valued image data, difference information from density information for a previous pixel, which is the predicted value.
The image processing apparatus according to any one of the preceding claims. 3. The image processing apparatus according to claim 2, wherein the previous pixel is a immediately preceding pixel adjacent to the image forming unit in the main scanning direction. 4. The image processing apparatus according to claim 3, wherein the encoding unit encodes each pixel of the multi-valued image data based on first-order difference information of density information with respect to a previous pixel. apparatus. 5. The image processing apparatus according to claim 1, wherein the prediction value is a value obtained by prediction using peripheral pixels. 6. The image processing apparatus according to claim 1, wherein the encoding unit ignores an overflow that occurs when calculating the difference information. 7. The image processing apparatus according to claim 6, wherein said encoding means uses, for each pixel of said multi-valued image data, difference information thereof in a two's complement representation. 8. The multi-valued image data is decoded by decoding the code data held in the holding unit into difference information for each pixel, and integrating the difference information. 3. The image processing device according to 2. 9. The image processing apparatus according to claim 2, wherein the encoding unit calculates a difference value of the first pixel, assuming that density information of a pixel preceding the first pixel is a specific value. apparatus. 10. The encoding unit, for each pixel column of the multi-valued image data, assuming that density information of a pixel preceding the first pixel in the pixel column is a specific value, and The image processing apparatus according to claim 9, wherein is calculated. 11. The image processing apparatus according to claim 9, wherein the specific value is 0. 12. The image processing apparatus according to claim 2, wherein said encoding unit performs encoding in accordance with an appearance frequency of said difference information. 13. The image processing apparatus according to claim 12, wherein said encoding means assigns a fixed-length code according to an appearance frequency of said difference information. 14. The image processing apparatus according to claim 13, wherein the encoding unit assigns an exceptional code to difference information having a low appearance frequency. 15. The image processing apparatus according to claim 12, wherein the encoding unit assigns a code having a shorter code length to the difference information having a higher appearance frequency. 16. The image processing apparatus according to claim 15, wherein the encoding unit holds the density information in the multi-valued image data while holding the density information discretely. 17. The image processing apparatus according to claim 16, wherein the restoration unit integrates the difference information with the density information discretely held by the holding unit as a starting point. 18. The apparatus according to claim 1, further comprising a holding area control means for controlling an area for holding the code data by said encoding means in said holding means.
8. The image processing apparatus according to any one of 7. 19. An input unit for inputting color image data, an encoding unit for encoding a difference value between color components of the color image data input by the input unit, and code data obtained by the encoding unit Holding means for holding the image data; decoding means for decoding the code data held by the holding means to restore color image data; and output for outputting the color image data restored by the decoding means to a predetermined image forming means. And an image processing apparatus. 20. An image processing system in which an external device and an image processing device are connected, wherein, on the external device side, difference value calculating means for calculating difference value information of multi-valued image data; An encoding unit that encodes the value information; a transmission unit that transmits the encoded image data; a receiving unit that receives the encoded image data on the image processing apparatus side; Decoding means for decoding the decoded image data to obtain difference value information; restoration means for restoring image data based on the decoded difference value information; and image formation for forming an image based on the restored image data. Means, and an image processing system comprising: 21. The external device, wherein the multi-valued image data is color image data composed of a plurality of color components, further comprising a color separation unit that separates the multi-valued color image data for each color, The calculating means calculates difference information between each color of the image data separated for each color, and the transmitting means transmits the image data for each color encoded by the encoding means in a frame-sequential manner, On the image processing apparatus side, the receiving unit receives the encoded image data for each color in a frame-sequential manner, and the restoring unit sets the image data for each color based on the difference value information decoded by the decoding unit. 21. The image processing system according to claim 20, wherein the image forming unit forms an image in a frame-sequential manner based on the restored image data for each color. 22. The image processing system according to claim 21, wherein the transmission order of each color in the transmission unit is the same as the image formation order in the image forming unit. 23. The image processing system according to claim 21, wherein said difference value calculating means calculates a difference value with respect to image data of a previous color for image data of a second color and thereafter. 24. The image processing system according to claim 23, wherein said difference value calculating means does not calculate a difference value for the image data of the first color and outputs the image data as it is. 25. The image processing apparatus according to claim 25, wherein the color restoring unit restores the image data of the second and subsequent colors by adding the difference value information decoded by the decoding unit to the image data of the previous color. The image processing system according to claim 24, wherein: 26. The image processing apparatus further includes a holding unit for holding image data output to the image forming unit, wherein the holding unit has a size capable of holding image data for one color. 26. The image processing system according to claim 25, wherein: 27. The image processing apparatus according to claim 27, wherein the color restoring unit sequentially updates the image data already held in the holding unit when outputting the image data of the second and subsequent colors to the holding unit. 27. The image processing system according to 26. 28. The external device separates a drawing command and image data, and determines a drawing order of each drawing command and image data according to whether or not a drawing area of each drawing command and image data overlaps. The image processing system according to any one of claims 20 to 27, further comprising a drawing command control unit that performs the drawing command. 29. The image processing system according to claim 28, wherein the drawing command is a character code or graphic drawing command. 30. The image processing system according to claim 29, wherein said drawing command control means adds a flag indicating a drawing order with said image data to said drawing command. 31. The image processing system according to claim 30, wherein the transmitting unit transmits the drawing command to which the flag is added first, and subsequently transmits image data in a frame-sequential manner. 32. An image processing apparatus, comprising: a development control unit for performing development by considering a drawing order with image data before output to the image forming unit by referring to a flag added to a received drawing command. The image processing system according to claim 31, further comprising: 33. The image processing system according to claim 20, wherein the external device is a host computer. 34. The image processing system according to claim 33, wherein said image forming means forms an image by an electrophotographic method. 35. An inputting step of inputting multi-valued image data, an encoding step of encoding the input multi-valued image data based on a difference value thereof, A holding step of holding; a decoding step of decoding code data held by the holding unit to restore multi-valued image data; and forming and outputting a multi-valued image based on the restored multi-valued image data. An image processing method, comprising: an image forming step. 36. An image processing method in an image processing system in which an external device and an image processing device are connected, wherein, on the external device side, a difference value calculating step of calculating difference value information of multi-valued image data; An encoding step of encoding the calculated difference value information; and a transmitting step of transmitting the encoded image data, wherein the image processing apparatus side receives the encoded image data. A decoding step of decoding the received image data to obtain difference value information; a restoration step of restoring image data based on the decoded difference value information; and An image processing method, comprising: forming an image.