JPH0573668A - Mosaic processor - Google Patents

Abstract

PURPOSE:To form mosaic picture data having mosaic pictures in plural kinds of sizes in one picture on a real time basis. CONSTITUTION:A density change detection means 701 detecting a density difference between the picture elements of original picture information in picture element units obtained by dividing the picture into the picture elements in a main scan direction and an auxiliary scan direction, a size information generation means 702 generating size information obtained by permitting either the main scan direction or the auxiliary scan direction to designate the size of the picture element matrix of plural picture elements in accordance with the density difference, size setting means 703 and 704 setting the size of the picture element matrix corresponding to size information and a picture information conversion means 704 converting original picture information on the respective picture elements constituting the picture element matrix into picture information representing the picture element matrix by the division of the picture element matrix in the size which the size setting means 703 and 704 set are provided.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a mosaic processing apparatus, which is installed in a digital copying machine or the like, for performing predetermined processing on digital image data input serially to form mosaic image data, and particularly to one image. The present invention relates to a mosaic processing device that forms image data in which mosaic images of a plurality of sizes are mixed in real time.

[0002]

2. Description of the Related Art Recently, in an electronic image recording means such as a digital color copying machine and a digital color printing machine, there is an increasing demand for extremely faithfully reproducing an original image on a sheet, and one of typical examples thereof. One example is a mosaic image. Conventionally, as a method of forming a mosaic image, an original image is divided into a large number of rectangular regions having a constant size (hereinafter referred to as “unit block”), a representative point is set in the block, and the color of the representative point is It was common to replace the entire area of a unit block with the concentration.

[0003]

In the conventional mosaic function (a function of forming a mosaic image) such as a total scanner for plate making or DTP (Desk Top Publishing), image data of all original images are stored in a large-scale frame memory (2). Once stored in a three-dimensional image memory), the representative points in each unit block are accurately determined through complicated calculations by the built-in CPU. For this reason, the apparatus becomes large and expensive, and there is a problem that the rear time processing for data becomes impossible.

As a device made in view of the above, two lines of line memories are provided, and these line memories are alternately used, and when lines are read out, the data corresponding to the unit block is replaced and transmitted (the same data is repeated). A digital image processing apparatus (for sending out) to form a mosaic image in real time has been proposed (Japanese Patent Laid-Open No. 3-65873), but it has a drawback that only one kind of mosaic size can be formed in one image.

An object of the present invention is to form mosaic image data having mosaic images of a plurality of sizes in one image in real time.

[0006]

A mosaic processing apparatus according to a first aspect of the present invention is a density change detection for detecting a density difference between pixels in original image information in pixel units obtained by dividing an image into pixels in a main scanning direction and a sub scanning direction. Means (701); size information generating means (702) for generating size information designating the size of a pixel matrix of at least one of a plurality of pixels in at least one of the main scanning direction and the sub scanning direction in correspondence with the density difference; Size setting means (703, 70) for setting the pixel matrix size corresponding to the information
4); and image information conversion means for converting the original image information of each pixel forming the pixel matrix in the pixel matrix division of the size set by the size setting means (703, 704) into image information representative of the pixel matrix (704); The symbols in parentheses are the corresponding elements of the examples described later.

The mosaic processing apparatus of the second invention is a random information generating means (801) for generating size information whose information content is random; a pixel matrix of a plurality of pixels in at least one of the main scanning direction and the sub scanning direction. Size setting means (703, 704) for setting the size in correspondence with the size information; and pixel matrix division of the size set by the size setting means (703, 704) for displaying the original image information of each pixel forming the pixel matrix. Image information conversion means (704) for converting the pixel matrix into image information representative of the pixel matrix.

Furthermore, the mosaic processing apparatus of the third invention is
Density change detecting means for detecting a density difference between pixels in original image information in pixel units obtained by dividing an image into pixels in the main scanning direction and the sub scanning direction; a pixel matrix in which at least one of the main scanning direction and the sub scanning direction has a plurality of pixels First size information generating means (701) for generating the first size information designating the size of the first density information corresponding to the density difference; random information generating means (801) for generating the second size information having random information content; Size information selecting means (802) for selecting one of the first size information and the second size information; the size information selecting means selects the size of the pixel matrix of a plurality of pixels in at least one of the main scanning direction and the sub scanning direction. The size setting means (703, 704) that is set in accordance with the selected size information; and the size setting means (703, 70)
Image information conversion means (70) for converting the original image information of each pixel forming the pixel matrix into image information representative of the pixel matrix in the pixel matrix division of the size set by 4).
4);

[0009]

According to the first invention, the size setting means (703, 70
4) sets the pixel matrix size corresponding to the size information generated corresponding to the density difference between pixels of the original image information, the image information conversion means (704), the size setting means (703, 704)
In the pixel matrix division of the size set by, the original image information of each pixel forming the pixel matrix is converted into image information representative of the pixel matrix. Therefore, although the pixel matrix size is selected based on the change in the inter-pixel density of the original image information, in the preferred embodiment of the present invention, a small matrix size is used at a large change amount and a large matrix size is used at a small change amount. Since the selection is made, it is possible to reproduce a variety of unique image information in which different matrix sizes are mixed. Since the amount of change is large in the edge portion of the image and the character image portion and a small matrix size is selected, image information can be obtained in these portions in a shape and density relatively close to the original.

According to the second invention, the size setting means
(703, 704), the information content corresponds to the random size information, at least one of the main scanning direction and the sub-scanning direction sets the size of the pixel matrix of a plurality of pixels, the image information conversion means (704), In the pixel matrix section of the size set by the size setting means (703, 704), the original image information of each pixel forming the pixel matrix is converted into image information representative of the pixel matrix. Therefore, image information in which a plurality of types of matrix sizes coexist can be randomly formed in one image,
Creates a special design effect on the playback image. Further, even for the same original, since the matrix size is randomly selected for each process, the reproduction image information is not the same.

Further, according to the third invention, the size information selecting means (802) has the pixel matrix size information corresponding to the size information generated corresponding to the density difference between pixels and the size information in which the information content is random. Choose one. The size setting means (703, 704) sets the size of the pixel matrix of a plurality of pixels in at least one of the main scanning direction and the sub-scanning direction in correspondence with the size information selected by the size information selecting means, and further The information conversion unit (704) converts the original image information of each pixel forming the pixel matrix into image information representing the pixel matrix in the pixel matrix section of the size set by the size setting unit (703, 704).
Therefore, the respective effects obtained by the first invention and the second invention can be selected and obtained according to the situation at that time. In the preferred embodiment of the present invention, the image data that is sequentially input is alternately stored for each line.
By alternately switching the write / read operation to the system memory, it is possible to form a matrix image in real time without using a large amount of room memory or complicated calculation, and it is effectively applied to digital copiers, etc. it can. Other objects and features of the present invention will become apparent from the following description of embodiments with reference to the drawings.

[0012]

1 is a block diagram showing a schematic configuration of an electronic image printing apparatus equipped with an image processing apparatus for performing mosaic processing according to the present invention. An image reading apparatus 1 for reading a document image as digital data while scanning the image, The image processing apparatus 2 is configured to perform predetermined processing / processing on the image data, and the image recording apparatus 3 for printing the image data supplied from the image processing apparatus 2 on a transfer sheet. The image processing apparatus 2 includes an image processing unit 2a that performs image processing, an operation unit 2c that inputs an operation instruction regarding image processing, and a control unit 2b that controls the image processing unit 2a and the operation unit 2c. There is.

FIG. 2 is a block diagram showing the configuration of the image processing unit 2a. The image processing unit 2a includes a change amount detector 701,
Decoder 702, latch 703, and processing circuit 704
It is composed of 3 shows the configuration of the change detector 701, FIG. 4 shows the configuration of the latch 703, and FIG. 5 shows the configuration of the processing circuit 704. Below, FIG. 2, FIG.
The operation of the image processing unit 2a will be described with reference to FIGS.

In the processing circuit 704 shown in FIG.
Reference numerals 1 and 302 denote memories for the first and second systems, respectively, which have a function of storing image data for one line in the main scanning direction, respectively. Image data is read. That is,
The first and second system memories 301 and 302 perform the writing and reading functions while being exchanged alternately. 101,1
Reference numerals 02 are input registers for the first and second systems, respectively, and output to the first and second system memories 301 and 302, respectively. 20
Reference numeral 1 is a write address counter that calculates the write address of the first and second system memories 301 and 302 in response to a pixel clock (hereinafter referred to as “CLK”), and 202 is a read address counter that calculates the read address. is there. Reference numerals 103 and 104 denote write address buffers for the first and second systems, respectively, and 502 is a read address conversion unit for thinning out and extracting the output value of the read address counter 202 every predetermined number of times (hereinafter referred to as "sampling"). Also, 1
07 and 108 are registers for the first and second systems, 109
Is a selector which selects one of the read values from 107 and 108 and supplies it to the next step (in this case, the image recording apparatus 3 shown in FIG. 1). Reference numeral 402 is a memory control unit that regulates the operation of the entire image processing unit 2a.

First, the operation when the mosaic function is canceled (normal image recording) will be described. At this time, the function of the read address conversion unit 502 is canceled, and the output of the read address counter 202 is directly input to the read address buffers 105 and 106 for the first and second systems. As described above, when the data writing to the first system memory 301 is in progress, the data already stored is read from the second system memory 302 and the function thereof is exchanged at predetermined intervals. Therefore, the operations of these two systems will be described in parallel below.

The original image is read as digital data by the image reading device (1 in FIG. 1) line by line, and is stored in the memory 301 (30) via the input register 101 (102).
Sequentially stored in 2). The write address counter 201 counts 1 when the data for one pixel is stored,
The calculated value (read address) is sent to the memory 301 (302) via the write address buffer 103 (104), and the write address of 301 (302) advances by one address.

In parallel with this, data is read from the memory 302 (301). That is, the read address counter 202 is the write address counter 201.
Similarly to, the counting operation is repeated in response to the CLK pulse, and the calculated value is the read buffer 502 (while the function is stopped).
Through the read address buffer 106 (10
After 5), it is transmitted to the memory 302 (301).
Therefore, the data read address from the memory 302 (301) advances by one address. This allows the memory 302
In parallel with the writing of data to (301), the memory 30
Data is read from 2 (301).

The data read from the memory 302 (301) pixel by pixel is output to the next step (image recording apparatus 3) via the output register 108 (107) and the selector 109. .. The selector 109 selects the output register 108 (107) when data is being read from the memory 302 (301). When the reading of the data for one line is completed, the selector 109 switches to select the output register 107 (108).

All the operations described above are controlled by control signals output from the memory control unit 402 to the respective units. The timing chart of this operation is as shown in the upper part of FIG. 14 (portion with (1) added).

Next, regarding the mosaic processing, the formation of a mosaic size signal will be described first. In FIG. 3, the image data is delayed by one line in the one-line data delay circuit 901 of the change amount detector 701.
The further delayed data is the 1-line data delay circuit 90.
At 2, the data is delayed by one line. This makes it possible to obtain data for three lines at the same time.
In addition, the above three lines of data are processed by the 1-pixel data delay circuit 9
A data delay of 1 pixel is made by 03, 904, and 905, and a data delay circuit 906 again makes a data delay of 1 pixel for the central line of the three lines. As described above, the upper, lower, left and right pixels can be obtained at the same time with respect to the output pixel A of the one-pixel data delay circuit 904.
FIG. 6 shows pixel data, and the outputs A, B, C, D, shown in FIG.
It shows the positional relationship of E. The image data is a signal having density information of 8-bit 256 gradations, and the target pixel is A
In such a case, the detector 907 detects the difference between B and C and the difference between D and E, and outputs the larger one as the variation a. The change amount a is also a signal having density information of 8-bit 256 gradations. In this way, the change amount detector 701 detects the amount of change in the density of the target pixel with respect to the adjacent upper, lower, left and right pixels.

The change amount a detected by the change amount detector 701 is input to the decoder 702 (FIG. 2). The decoder 702 is composed of a ROM in which a is an address and b is data, and as shown in FIG.
Output b = 0 or b = 1 or b = 2 or b = 3. That is, the value of b becomes the representative value (rank) of the change amount of the image data. Since b is a value of 0 to 3, it is represented by 2 bits.

The latch 703 (FIGS. 2 and 4) stores the signal b in the memory 402 at the timing of the pixel corresponding to X = 1 and Y = 1 of the block composed of the rectangular area of 32 × 32 pixels as shown in FIG. Write to. Also, between Y = 1 to 32, X = 1
The signal b written in the memory 402 is read out sequentially at the timing of the 2-bit block size signal BS (BS0,
BS1). Since BS0 and BS1 are signals whose output timing of the signal b is adjusted, the value is the same value of the signal b as shown in FIG. That is, the latch 70
3 performs an operation of fetching the change amount of the corresponding pixel only once for each 32 × 32 pixel block, and outputs the BS signal for every 32 pixels of each line while scanning the line for 32 pixels. In this embodiment, the mosaic size has a basic size of 32 × 32 pixels, and this size is 2 2 i (i = 1, 1).
A total of four types of mosaic sizes, which are divided into 2 and 3), are generated.

Next, using the mosaic size signal BS,
A case of obtaining a mosaic pattern will be described. In this case, in the processing circuit shown in FIG. 5, the read address conversion unit 502 restores its original function. Read address conversion unit 5
02 denotes the number of pixels N (N = 4, which is determined by the BS signal).
Every time (8, 16, 32) is counted, the count value of the read address counter 202 is fetched. In other words, the read address conversion unit 502 uses the read address counter 202.
The sampled value is sampled every N addresses, and the read address buffer 106 (105) is repeated N times to read the image data of the same address. Since the BS signal arrives at every 32 pixels, the value of N is changed every 32 pixels and the above operation is repeated. As described above, reading of one line in the main scanning direction is performed. The read data is stored in the register 108 (10
7) and the selector 109 to output to the image recording apparatus 3. The data for one line output at this time is sequentially taken into the one-line delay circuit 110.

Even when the data processing (reading) for one line is completed, the write / read operation of the memories 301 and 302 is not switched, and the data is read again from the memory 302 (301). During this time, the line data next to the original image is written in the memory 301 (302) (the old data is erased). Thereafter, this operation is repeated (for four pixels in the sub-scanning direction). That is, the data of the fifth pixel is stored in the memory 301 (302) in the sub-scanning direction, and the 1-line delay circuit 11
0 is the output state of the line data of the fourth pixel in the sub-scanning direction. Here, the writing / reading operation of the memories 301 and 302 is switched, and the data reading of the fifth pixel is started in the sub-scanning direction. At this time, when the value of N determined by the BS signal is smaller than 5, that is, when N = 4, data is read from the memory 301 (302) in the same manner as described above, but N = 8, 16, 32. In this case, the data is sequentially read from the 1-line delay circuit 110. The selector 109 switches between the memory 301 (302) and the 1-line delay circuit 110. Due to the difference in mosaic size, there is a mixture of where the data of the previous line is read as it is on one line and where the sample data is updated and read, so switching by the selector 109 is necessary.
The above operation is repeated for four pixels in the sub-scanning direction.

Similarly, when the ninth pixel is operated in the sub-scanning direction, if the value of N determined by the BS signal is smaller than 9, that is, N = 4,8, the memory 302
Data is read from (301) in the same manner as described above, but if N = 16, 32, the 1-line delay circuit 110 is used.
The data is sequentially read.

Further, when the 17th pixel is operated in the sub-scanning direction, if the value of N determined by the BS signal is smaller than 17, that is, if N = 4, 8, 16 then the memory 301 (302). From the above, data is read in the same manner as described above, but when N = 32, the 1-line delay circuit 110
The data is sequentially read.

As described above, the memories 301 and 302 are switched every 4 pixels in the sub-scanning direction, and the next arbitrary Mth (M
At the time of (pixel) operation, if the value of M is smaller than the value of N determined by the BS signal, data is read from the memory 301 (302) in the same manner as described above. Data is sequentially read from the line delay circuit 110. This operation is repeated as a set of 32 pixels in the sub-scanning direction (the (32 × n + 1: n = positive integer) pixel is considered to be the first pixel in the sub-scanning direction).

By performing the memory switching operation (control in the sub-scanning direction) as described above and simultaneously continuing the above-described processing in the main-scanning direction, as shown in FIG.
It is possible to form four types of mosaic images having different sizes for each block of 32 × 32 pixels. Note that, for example, the timing chart of the operation when forming a mosaic of 4 × 4 pixels is shown in the lower part of FIG. 14 (the part marked with (2)) and FIG.
As shown in FIG.

The logic actually used in the processing circuit shown in FIG. 5 will be described below. First, FIG. 10 shows a main part of the read address counter 202 shown in FIG.
Reference numeral 704 indicates a 13-bit counter, CLK indicates a pixel clock, and LSYNC with an overline indicates a synchronization signal.

FIG. 11 shows the memory control unit 4 shown in FIG.
5 shows a 5-bit counter 705 for controlling the block size in the sub-scanning direction. The signal FG is at level "1" while the image data for one pixel is being input, and is at "0" during other periods. The Y2 signal is stored in the memories 301 and 302.
Is used as the signal RW for controlling the read / write toggle operation, the memory 301 and the memory
02 toggle operation.

FIG. 12 shows the address conversion unit 50 shown in FIG.
2 shows the main part of the read address X0 to X12 and converts the lower address X0 to X4 into X0 'to X4', and this conversion controls the size of the mosaic size in the main scanning direction. 707 is a selector, S0 =
If “0” and S0 = “1”, Yi = Ai (i = 0, 1,
If S0 = "1" and S0 = "0" in 2), Yi = Bi
If S0 = "0" and S0 = "1", then Yi = Ci,
If S0 = "1" and S0 = "1", Yi = Di.
Therefore, for example, when BS = 1, X0 'to X3' =
0, X4 '= X4, which means that the read address is incremented by 16 at every 16 pixels, and when BS = 4, X0'-X1' = 0, X2'-X4 '= X2-X4, This means that the read address advances by 4 at a stretch for every 4 pixels. This controls the size of the mosaic size in the main scanning direction.

FIG. 13 shows a mosaic pattern in an output image when the mosaic size is selectively switched. As shown in the figure, an image in which mosaic sizes of four types are mixed in one image can be obtained. .. In particular, a small mosaic pattern is selected in the part where the image data changes drastically, and a large mosaic pattern is selected in the part where the change does not change much. A small mosaic pattern is selected for the character image portion and the like inside, and a unique effect such that the character is easy to read is obtained.

FIG. 16 is an embodiment showing another example of the image processing section 2a shown in FIG. 2, in which the block size signal BS is pseudo-randomly generated. Reference numeral 801 is a 4-bit shift register. The image data is the image reading device 1
The analog signal read by the A / D converter is quantized (LSB) with 8-bit 256 gradations for each pixel, and the LSB hardly depends on the density of the original image, and the illumination optical system and the reading circuit. It is effective as a pseudo random signal because it depends on the noise component of the system. The shift register 801 allows the L of each of two pixels separated by two pixels.
A 2-bit random number is obtained by combining SBs. The latch 703 has the same configuration and operation as the example shown in FIG. 2, and generates the block size signal BS (BS0, BS1) in the block of 32 × 32 pixels. The processing circuit 704 is similar to the example shown in FIG.

By randomly generating BSs in this way, four kinds of mosaic sizes can be randomly mixed in one image, a special design effect is produced in a reproduced image, and even in the case of the same original document. There is an effect that an image with a different mosaic arrangement can be obtained for each reading, image processing, and reproduction.

FIG. 17 shows that the selector 802 selects one of the two types of block size signals BSA and BSB to generate the final block select signal BS. 701 and 702 are the same as those shown in FIG. With the same configuration, 801 has the same configuration as the example shown in FIG. That is, there are two types of block signal generation sources, which are switched by the selector 802. The select signal SEL for switching the sectator 802 is the operation unit 2c of the apparatus.
The control unit 2b generates based on an input instruction from (FIG. 1). The latch 703 and the processing circuit 704 in the figure are also similar to the example shown in FIG. Therefore, by inputting a selection instruction to the operation unit 2c, one of the image data formed by the image processing unit shown in FIG. 2 and the image data formed by the image processing unit shown in FIG. 16 is selected. Obtainable.

As described above, the mosaic processing according to the present invention has been described. However, the image data input to this apparatus may be limited to that immediately after being read from the manuscript surface, or a facsimile apparatus. It may be transmitted from a remote place. That is, the digital image processing apparatus of the present invention can be widely applied to all electronic image recording means.

Further, the circuit diagram as the realizing means is not limited to the embodiment, and various modifications can be made. Further, the size and type of the mosaic are not limited to those in the embodiment, and the mosaic size information can be changed without being limited to the example shown in FIG. For example, there are various known techniques for automatically discriminating a copy region and a character region in an image from the characteristics of image data. By applying this technique, only the copy region is set to the mosaic image mode, or the photo region is used. The application such as switching the mosaic size between the character area and the character area is also possible.

[0038]

As described above, according to the first invention, the size setting means (703, 704) sets the pixel matrix size corresponding to the size information generated corresponding to the density difference between pixels of the original image information. Then, the image information conversion unit (704) converts the original image information of each pixel forming the pixel matrix into image information representative of the pixel matrix in the pixel matrix division of the size set by the size setting unit (703, 704). To do. Therefore, although the pixel matrix size is selected based on the change in the inter-pixel density of the original image information, in the preferred embodiment of the present invention, a small matrix size is used at a large change amount and a large matrix size is used at a small change amount. Since the selection is made, it is possible to reproduce a variety of unique image information in which different matrix sizes are mixed. Since the amount of change is large in the edge portion of the image and the character image portion and a small matrix size is selected, image information can be obtained in these portions in a shape and density relatively close to the original.

According to the second invention, the size setting means
(703, 704), the information content corresponds to the random size information, at least one of the main scanning direction and the sub-scanning direction sets the size of the pixel matrix of a plurality of pixels, the image information conversion means (704), In the pixel matrix section of the size set by the size setting means (703, 704), original image information of each pixel forming the pixel matrix is converted into image information representative of the pixel matrix. Therefore, image information in which a plurality of types of matrix sizes coexist can be randomly formed in one image,
Creates a special design effect on the playback image. Further, even for the same original, since the matrix size is randomly selected for each process, the reproduction image information is not the same.

Further, according to the third invention, the size information selecting means (802) has the pixel matrix size information corresponding to the size information generated corresponding to the density difference between the pixels and the size information in which the information content is random. Choose one. The size setting means (703, 704) sets the size of the pixel matrix of a plurality of pixels in at least one of the main scanning direction and the sub-scanning direction in correspondence with the size information selected by the size information selecting means, and further The information conversion unit (704) converts the original image information of each pixel forming the pixel matrix into image information representing the pixel matrix in the pixel matrix section of the size set by the size setting unit (703, 704).
Therefore, the respective effects obtained by the first invention and the second invention can be selected and obtained according to the situation at that time. In the preferred embodiment of the present invention, the image data that is sequentially input is alternately stored for each line.
By alternately switching the write / read operation to the system memory, it is possible to form a matrix image in real time without using a large amount of room memory or complicated calculation, and it is effectively applied to digital copiers, etc. it can.

[Brief description of drawings]

FIG. 1 is a block diagram showing a schematic configuration of an electronic image printing apparatus equipped with an image processing apparatus that executes a mosaic process of the present invention.

FIG. 2 is a block diagram showing an example of a schematic configuration of an image processing unit 2a shown in FIG.

3 is a block diagram showing a configuration of a change amount detector 701 shown in FIG.

4 is a block diagram showing a configuration of a latch 703 shown in FIG.

5 is a block diagram showing a configuration of a processing circuit 704 shown in FIG.

FIG. 6 is a block diagram showing an example of image data,
This is for explaining the detection of the change in the image data.

FIG. 7 shows a decoder 7 including a ROM shown in FIG.
3 is a bit assignment diagram showing the output relationship of 02 with respect to address a and data b. FIG.

FIG. 8 is a block diagram showing an example of a mosaic composed of rectangular regions of 32 × 32 pixels.

FIG. 9 is a block diagram showing a mosaic pattern in an output image when the mosaic size is selectively switched.

FIG. 10 is a read address counter 202 shown in FIG.
3 is a block diagram showing the configuration of the main part of FIG.

11 is a 5-bit counter for controlling the block size in the sub-scanning direction in the memory control unit 402 shown in FIG. 5 (a counter that controls the read / write toggle operation of the memories 301 and 302 by the output Y2). ) Is a block diagram showing

12 is a block diagram showing a selector included in the address conversion unit 502 shown in FIG.

FIG. 13 is a plan view showing an example of a mosaic pattern obtained by the present invention.

FIG. 14 is a time chart showing the operation timing of each unit shown in FIG. 5 during normal image formation and during formation of a 4 × 4 pixel mosaic image.

FIG. 15 is a time chart showing the operation timing of each unit shown in FIG. 5 when forming a 4 × 4 pixel mosaic image.

16 is a block diagram showing a schematic configuration of another example of the image processing unit 2a shown in FIG. 2, which is a block size signal BS.
It shows a configuration in which is generated in a pseudo-random manner.

FIG. 17 is a block diagram showing a schematic configuration of still another example of the image processing unit 2a shown in FIGS. 2 and 16, and one is selected from two types of block size signals BSA and BSB to obtain a mosaic image. It is a composition.

[Explanation of symbols]

1: Image reading device 2: Image processing device 2a: Image processing part 2b: Control part 2c: Operation part 3: Image recording device 101: Input register for first system 102: Input register for first system 103: First System write address buffer 104: Second system write address buffer 105: First system read address buffer 106: Second system read address buffer 107: First system output register 108: Second System output register 109: Selector 201: Write address counter 202: Read address counter 203: Sub-scanning direction address counter 301: First system memory 302: Second system memory 401: Output register control unit 402: Memory control Part 502: Read address conversion part 701: Change detector (density change detecting means) 702: Decoder ( Size information generating means, first size information generating means) 703: latch 704: processing circuit (703, 704: size setting means, 704: image information converting means) 705, 706: counter 707: selector 801: shift register (random information) Generating means, second
Size information generating means) 802: Selector (size information selecting means)

Claims (3)

[Claims] 1. A density change detecting means for detecting a density difference between pixels of original image information in pixel units obtained by dividing an image into pixels in the main scanning direction and the sub scanning direction; at least one of the main scanning direction and the sub scanning direction. Size information generating means for generating size information designating a size of a pixel matrix of a plurality of pixels corresponding to the density difference; size setting means for setting a pixel matrix size corresponding to the size information; and the size setting means A mosaic processing apparatus comprising: an image information conversion unit that converts original image information of each pixel forming the pixel matrix into image information representative of the pixel matrix in a pixel matrix section having a size set by. 2. Random information generating means for generating size information whose information content is random; a size of a pixel matrix having a plurality of pixels in at least one of a main scanning direction and a sub scanning direction is set corresponding to the size information. Size setting means; and image information conversion means for converting the original image information of each pixel forming the pixel matrix in the pixel matrix division of the size set by the size setting means into image information representative of the pixel matrix. A mosaic processing device provided. 3. A density change detecting means for detecting a density difference between pixels in original image information in pixel units obtained by dividing an image into pixels in the main scanning direction and the sub scanning direction; at least one of the main scanning direction and the sub scanning direction. First size information generating means for generating first size information designating a size of a pixel matrix of a plurality of pixels corresponding to the density difference; random information generating means for generating second size information having random information content; Size information selecting means for selecting one of the first size information and the second size information; the size of the pixel matrix having a plurality of pixels in at least one of the main scanning direction and the sub scanning direction is set to the size information selected by the size information selecting means. Size setting means for correspondingly setting; and original image information of each pixel forming the pixel matrix in the pixel matrix division of the size set by the size setting means, A mosaic processing device comprising: image information conversion means for converting a pixel matrix into representative image information.