JPH0884227A - Color image processor - Google Patents

Abstract

(57) [Summary] [Object] To provide a color image processing apparatus capable of reducing the capacity of a connection memory used for marker editing processing. [Structure] The data read by the read sensor 1 and the output data of the noise removal circuit are combined to determine a plurality of output codes, and the correction circuit 61 corrects the output codes according to conditions. The connection memory 62 connected to the correction circuit 61 stores the correction information of the previous band and reduces the number of stored line information to be smaller than the number of pixels in the line direction.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a color image processing apparatus such as a color copying machine for reading a color-marked original on a monochrome original and performing marker editing processing.

[0002]

2. Description of the Related Art In a marker editing process using color marking, a plurality of processes such as filling a closed section or changing a character color are automatically recognized and processed according to a marking method. Therefore, the marking method has a certain rule for distinguishing between them, and by reading a marker document according to this marking rule, accurate marker editing is performed.

Conventionally, in this type of marker editing processing, for example, an original in which a monochrome original is color-marked is read by a BJ method color copying machine which is a serial scan method, and stored data of processed pixels in the main scanning direction and reading are performed. The print color is determined from the obtained input pixel data.

Then, the marker document is divided into a plurality of bands and read by a serial scan method. In this case, it is necessary to connect the data between the respective bands in order to maintain the continuity of the processing. Therefore, a connection memory is provided and all the connection data are stored in this connection memory.

[0005]

As described above, in the conventional marker editing process, since all the connection data are stored in the connection memory, there is a problem that the capacity of the connection memory increases.

In view of the above-mentioned conventional problems, it is an object of the present invention to provide a color image processing apparatus capable of reducing the capacity of the connection memory used for marker editing processing.

[0007]

In order to achieve the above object, the present invention provides a reading sensor which divides an original into a plurality of bands and reads the original and outputs multi-valued data, and the multi-valued data that expresses colors. Coding means for classifying into a plurality of color codes, area processing means for performing two-dimensional area processing on the codes classified into a plurality of color codes, storage means for storing processing information of one line before, Reading means for reading the stored contents of the storing means; output code determining means for synthesizing the data read by the reading means and the output data of the area processing means to determine a plurality of output codes;
The correction code for correcting the output code of the output code determination means according to the conditions, the correction information storage means connected to the correction code for storing the correction information of the preceding band, and the processed information of the sensor arrangement direction are Information retrieval means for retrieving processed information in the sensor arrangement direction for the beginning of a non-existing band from the correction information storage means, and information writing means for writing the correction information of the next band in the correction information storage means,
Storage control means for storing the correction code output from the correction means in the storage means as processing information of one line before, and output conversion means for converting the output code corrected by the correction means into multivalued data and outputting the multivalued data. In the color image processing apparatus including the above, the number of line information stored in the correction information means is smaller than the number of pixels in the line direction.

In the first aspect of the invention, as compared with the line direction pixels used in the area processing means,
It is desirable to reduce the number of pixel units used by the correction information storage means.

Further, in the first invention, it is desirable that the information to be read from and written to the correction information storage means is coded.

[0010]

According to the present invention having the above structure, the number of line information stored in the correction information storage means is made smaller than the number of pixels in the line direction, so that the capacity of the correction information storage means can be reduced.

Further, for example, correction is made in comparison with the pixels in the line direction used in the area processing means for performing two-dimensional area processing such as removal and complementation of isolated points for codes classified into a plurality of color codes. By reducing the number of pixel units used by the information storage unit, the capacity of the correction information storage unit can be reduced easily and accurately.

Further, by encoding the information to be read from and written to the correction information storage means, the capacity of the correction information storage means can be further reduced.

[0013]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing a schematic configuration of image processing of a full-color copying machine according to an embodiment of a color image processing apparatus of the present invention, FIG. 2 is a diagram showing marker editing specifications, and FIG. 3 is a marking rule. It is a figure explaining.

First, the marker editing process will be described with reference to FIGS.

When a marker is edited by coloring a monochrome manuscript with a color marker, for example, as shown in FIGS.
An output as shown in (c) is obtained.

(1) As shown in FIG. 2A, when it is desired to paint a predetermined closed section of a document, by marking the inside of the closed section, an output in which the closed section is filled with color is obtained. be able to. (2) As shown in FIG. 2B, when it is desired to change the color of the black line of the original, marking is performed so as to include the black line, and the black line is replaced with the color of the marker. (3) As shown in FIG. 2C, when it is desired to perform both of the above cases (1) and (2), the closed section is filled with the composite form of (1) and (2) above. Then, the color of the portion included by the marker is replaced with the marker color.

In order to automatically distinguish and process the above (1), (2) and (3) at the same time, a marking rule as shown in FIGS. 3 (a) and 3 (b) is required.

In the case of (1) (filling), the inside of the black frame line 101 is marked as shown in FIG. 3 (a) (marking portion 102), but at this time, the black frame line 101 is marked.
The gap 103 formed on the inside of the gap or the protrusion 104 formed on the outside of the black frame line 101 is allowed within a predetermined range (for example, the width 103a of the gap 103 is within 1 mm,
The width 104a of the protrusion 104 is within 0.3 mm).

In the case of the above (2) (color conversion),
As shown in FIGS. 3 (a) and 3 (b), the surrounding 106 is marked so as to include the black character 105. At this time, the surrounding 106 must be filled over a predetermined range (for example, from the black character 105). Overhang 106a
Is 1 mm or more).

In addition to this, the black 107 inside the closed section must be separated from the marking portion 102 by a predetermined range or more (for example, the marking portion 102 and the adjacent black line 107).
And the distance 107a is 1 mm or more). Further, the width 101a of the black frame line 101 serving as the boundary of the closed section needs to be a predetermined value (for example, 0.3 mm) or more, and the coating width of the marker (width of the marking portion 102) 102a is also a predetermined value (for example, 0.3 mm).
mm) or more is required.

Documents satisfying these conditions are subjected to accurate marker editing processing.

In order to realize such marker editing processing, the image processing of the full-color copying machine of this embodiment is constructed as follows.

In FIG. 1, reference numeral 1 in the drawing is a CCD line sensor for reading an original image and outputting RGB data, and an amplifier circuit 2 for amplifying the RGB data and an amplified RGB data are provided on the output side thereof. A / D converter 3 for quantizing into an 8-bit digital value, and quantized R
A shading correction circuit 4 that performs shading correction of GB data, a color misregistration correction circuit 5 that corrects positional deviation of read RGB data, and a black character detection circuit 6 that detects black characters from this RGB data and generates a black character signal.
And a marker editing circuit 7 for performing marker editing described later are sequentially connected.

Further, on the output side of the marker editing circuit 7, a scaling signal for scaling the scaling and a control signal for generating control signals used in a spatial filter circuit 13 and a binarization circuit 15 which will be described later. A generation circuit 9; a LOG conversion circuit 10 that performs LOG conversion according to a LOG table;
Minimum value extraction circuit 11 for extracting the minimum value in the converted CMY (cyan, magenta, yellow) data, and masking / UCR for performing masking and UCR by matrix calculation
A circuit 12, a spatial filter circuit 13 that performs edge enhancement or smoothing processing, a gamma conversion circuit 14 that performs gamma conversion according to a gamma table, and a binarization circuit 15 that binarizes 8-bit multivalued data by a dither method or the like. , CM
A head timing adjustment circuit 16 for adjusting the heads for the four YK (cyan, magenta, yellow, black) colors;
A head driver circuit 17 that drives the adjusted head is sequentially connected, and a four-color (CMYK) BJ ink head 18 is connected to the output side of the head driver circuit 17.

Next, the marker editing circuit 7 will be described in detail with reference to FIG.

The RGB data read by the CCD line sensor 1 is in units of one pixel (data is 8 bits (0
.About.255)) is input to the marker editing circuit 7.

The marker editing circuit 7 comprises a marker color discriminating circuit 20, a sub-scanning direction noise removing circuit 30, a main scanning direction noise removing circuit 40, a region determining circuit 50, a region error determination correcting circuit 60 and an output converting circuit 70. Has been done.

First, the RGB data is a marker-dedicated LO.
After being converted into CMY data by the G conversion circuit 21, it is input to the black and white determination circuit 22 and the color determination circuit 23 in the marker color determination circuit 20. In the black and white judgment circuit 22, CM
If the minimum value of Y is above a certain value, it is black, and if the maximum value is below a certain value, it is white. In the color determination circuit 23, the minimum value of CMY is subtracted from each value to determine the color based on the ratio of the remaining two components. The white, black, and color determined by these are the coding circuit 24.
Is converted into a color code corresponding to Table 1 below.

[0030]

[Table 1] In addition, since halftone output is performed for black, the black density generation circuit 25 determines the density from the black signal from the output of the black and white determination circuit 22 and the color determination circuit 23 and the raw CMY data. produce.

The color code is the noise removal circuit 3 in the sub-scanning direction.
If the number of consecutive pixels of the same color is less than or equal to a certain value, it is determined to be noise and removed. Delay circuit 32
Is a circuit for synchronizing the black density with a delay caused by the processing of the sub-scanning direction noise removal circuit 31.

The main scanning direction noise removal circuit 40 uses the input data and the data delayed by one line in the FIFO 41 to calculate the continuity of the pixels of the same color in the logic circuit 42 and store it in the FIFO 43. In addition, the FIFO43
Of the data is used to determine which of the input data and the output data of the FIFO 41 is to be output as the color code.

Next, the area determination circuit 51 will be described.

The circuit 51 determines which of the four areas each pixel of the image is. These four areas are (1) normal area, (2) paint area,
(3) line area and (4) in-paint line area.

The normal area is an area that does nothing,
The initial value is set in this area. The paint area is
This is a region where the closed section is painted over. However, the normal area is set on the black line at the boundary. Further, the line area is an area in which black is replaced with a marker color. The in-paint line area is a composite type of the paint area and the line area, and is a line area in the paint area.

As shown in FIG. 5, the area determination circuit 51 has 128 pixels of sensors and each pixel (target pixel P1).
The read data and the storage data of the pixel P2 immediately before in the main scanning direction are stored in the FIFO memory 52.

The memory data stores the paint-determined color that determines the color in the paint area, the line-determined color that determines the color in the line or in-paint line area, the area, and the distance from the black pixel in the main scanning direction. Horizontal black counter, horizontal color white counter that stores the distance from the color or white pixel, the previous color that is the color of the pixel read one line before, and the current color that has changed from the main scanning direction There is a history color that remembers whether or not the color has changed.

The print color is determined by logical synthesis of these data and the read data. The details are shown in Tables 2 and 3 below.

[0039]

[Table 2]

[0040]

[Table 3] In Tables 2 and 3, YBK1: clear with black (clearance pixel number comparator), YBK2: clear with black (extruded pixel number comparator), YIROWHH: clear condition is changed depending on the area (normal: clear with color, (Painting: Clear with color, other than: No clear), N: Paint color, * 1: Processing in parentheses in line only mode, black, color, and white counters are always 1
Set to (Full).

As a simple example, referring to Tables 2 and 3, when the previous color is white in the paint area and the current color which is the input data is the marker color, looking at the vertical and horizontal black distance counters, If it is determined that the color is closer to black, the area becomes the paint area, the current color is printed, and the current color is set as the paint-determined color.

A point to be noted in the above Tables 2 and 3 is that a counter is used, whereby the marking is adjacent to black as long as it is within a certain condition even if the marking is not strictly adjacent to black. It can be treated as if it were. The reason for adopting the allowance of a gap instead of the protrusion as the marking condition is that the protrusion cannot be colored with respect to something like a pie chart, and the gap can be corrected.

Further, since the counter only looks at the vertical and horizontal directions, even if the marker color is adjacent to black in the diagonal direction, it cannot be detected by the vertical and horizontal counters. In this case, since a malfunction occurs, a tailing process as shown in FIG. 6 is added to the area determination circuit 51 to avoid it. This method replaces the pixel of interest P1 with the previous color if the pixel of interest P1 is white, the previous color is a marker color and is a paint area, and either the vertical or horizontal direction is closer to black. That is, if any of the distances A, B, and C from the black pixel is within 1 mm and the above condition is satisfied, the pixel of interest P1
The white of is the color pixel of the previous line. As a result, the white gap between black and the marker color is processed as if it were the marker color, and it is possible to deal with the gap in the diagonal direction.

Further, the reason why the stored data does not include the vertical distance counter is that there is no need to store the data because it can be calculated in the sensor currently being read. After the area determination, the color code is input to the area determination correction circuit 61. The area determination correction circuit 61 compares the area determination result of the area determination circuit 51 with surrounding pixels and corrects the surrounding pixels when it is determined that the area is included in an incorrect area (details will be described later). ). Further, the connection memory (SRAM) 62 is a memory for storing original data for correcting the previous band due to the band straddling. The contents stored in the connection memory 62 are the history codes described in Tables 4 and 5 below.

[0045]

[Table 4]

[0046]

[Table 5] The color code after the above-described processing is input to the output conversion circuit 70. This data can be switched to a color code such as data before area determination or data before noise removal by the selector 71, and is input to the print color determination circuit 72.

The print color judgment circuit 72 is a circuit for converting a 4-bit color code into CMYK data in a preset table. As for black, the density is generated by the density generation circuit 73 using the black density generated by the marker color discrimination circuit 20 in order to realize halftone.

The output conversion circuit 70 has three output modes: (1) standard back mode, (2) standard blue back mode, and (3) special blue back mode. The standard back mode is a mode in which the color code generated by the area determination circuit 51 is converted by the CMYK output table shown in Table 4 below and output. In the standard blue back mode, the pixels whose output is determined to be white in the standard back mode are converted into blue and the pixels whose output is determined to be black are converted into white, and are output. can do. The special blue back mode is almost the same as the standard blue back mode, but only the black in the paint area is output as it is.

[0049]

[Table 6] The CMYK data that has undergone the above processing is subjected to the processing after the scaling circuit 8 shown in FIG. 1 and printed by the printer section.
However, when the marker editing process is performed, the RGB data is converted into C in the marker editing process of the marker editing circuit 7.
Since it is converted to MYK data, the LOG conversion circuit 1
0, the minimum value extraction circuit 11, and the masking / UCR circuit 12 do nothing. When the marker editing is not performed, the marker editing circuit 7 also does not perform any processing.

Next, the correction circuit 61 will be described in detail.

As described above, when an original document that violates the marking rule shown in FIG. 3 is processed, the original document shown in FIGS.
The output is as shown in.

This will be described in detail with reference to Tables 2 and 3 above. In FIG. 7A, Z1 is accidentally entered into the line area at the entrance of the paint area. When the color changes from white in the normal area to black, there is no black in the surrounding area, so the area enters the line area. Further, the line area is continued in the order of color → black → color, and when the color → white is changed, the normal area is restored. Therefore, only a part of the paint area returns to the normal area, and a blank area appears as indicated by Z1 '. Similarly, in the case of Z2, there is a blank area as indicated by Z2 '.

In FIG. 7B, Z3 is a problem at the exit of the paint area. When the color of the paint changes to black, it returns to the normal area. However, since it changes from normal black to color again, it enters the paint area. After that, since the paint end condition does not exist, the color is continuously printed. In Z4, since the white gap is too wide when the color of the paint changes to black, it is judged to be black in the paint area to continue the paint area. Continue to print colors.

The problem with this approach is that the region information is processed only in the main scan direction and is completely independent of the sub scan direction. For this reason, even though the input data of the adjacent two are the same white, one is the paint area and the other is the normal area, which is not the actual state.

Therefore, the correction circuit 60 corrects the area processed only in the main scanning direction by the area determination circuit 50 in the sub scanning direction.

As shown in FIG. 8, the pixel of interest (history color, color,
Region P1 for the left pixel P in the main scanning direction
The area is determined using the stored data of 2. The stored data is the information shown in FIG. Next, the upper pixel P in the sub-scanning direction
The history code of No. 3 (coded a combination of history color, color, and area) is compared with the determined area, and when a certain condition is satisfied, the area is replaced with the area of the upper pixel.

At the beginning of the band, since there is no history code in the upward direction, the history code is taken out from the connection memory 62 and used. The connection memory 62 stores a history code required at the beginning of the next band determined by the previous band processing.

Here, the history code will be described using the history code determination conditions in Table 4.

The history code has a value of 0 to 5 in 3 bits. When the read color of the pixel of interest (current color in Table 5) is white and the area is normal, and the history color in the processed storage data in the main scanning direction is black, the history code of that pixel is "0". . This code is used as the upper pixel history code shown in the correction condition of Table 5 when performing the area correction of the pixel one below. Hereinafter, history codes 1 to 5 are assigned under the conditions shown in Table 5.

Next, a method for reducing the information of the history code (band connection data) stored in the connection memory 62 will be described with reference to FIGS. 9 (a) and 9 (b) and FIGS. 10 (a) and 10 (b).
An explanation will be given using (c).

In FIG. 9A, the inside of a square frame written in black is colored with a green marker (simply described as color in the figure).
FIG. 4 is an enlarged view of a manuscript that has been prepared. However, band processing is designed to enter near the center of a square frame written in black (indicated by a dotted line L in the figure). FIG. 9B is a diagram showing the output of the area determination correction circuit 61 when the document shown in FIG. 9A is read. As shown in the figure, the isolated points are removed by the effect of the noise removing circuits 30 and 40 in the sub-scanning direction and the main scanning direction, and the remaining color information is continuous in a plurality of pixels in the main scanning direction and the sub-scanning direction. I understand.

As shown in FIG. 10A, the band connection data portion of FIG. 9B is extracted and numbered in the main scanning direction, and then as shown in FIG.
The connection data of the bands corresponding to the odd numbers among the numbers given in the main scanning direction in (a) are extracted. And FIG.
As shown in FIG. 10C, when each pixel is doubled in the main scanning direction, the band connection data of FIG.
It can be seen that the same data as 0 (a) is obtained.

On the other hand, when a manuscript in which the inside of a square frame written in black is colored with a green marker is read,
As shown in FIG. 1, in the case where the width of the square frame written in black and a part of the width of the green marker are thick, if the same process as above is performed, the processes shown in FIGS. It will be as shown in c). 12 (a), (b), (c)
Is the same as in FIG. 10 (a), (b),
The figure corresponding to each of (c) is shown.

The banding data shown in FIG. 12A in which the banding data portion of FIG. 11 is extracted and numbered in the main scanning direction, and FIG. 12 obtained by performing the same processing as that of FIG. Comparing with the band connection data shown in (c), the connection data of the 10th band is “3” in the case shown in FIG. 12 (a), whereas it is “3” in the case shown in FIG. 12 (c). 5 ”, which is different.

However, this is not a problem for the following reason. That is, in the correction method shown in FIG. 8, if the pixel for which the tenth connection data is used is the pixel of interest P1, the history color to the left of the pixel of interest P1 is black, the current color is color, and the upper pixel history code is "5." , And Table 5
Area correction processing is not performed because it does not correspond to the correction condition of
The paint mode is executed with the color of the pixel of interest. As a result, even if the connection data is as shown in FIG.
The result of the marker processing is the same as the case of using the connection data of FIG.

As described above, the marker editing is performed based on the image signal from which the isolated points are removed by the effect of the noise removing circuits 40 and 30 in the main scanning direction and the sub-scanning direction, and regarding the band connection data generated as a result. It can be seen that does not need to store all lines. That is, it becomes possible to store the connection data in the connection memory 62 in the form shown in FIG. 10B or 12B.

Next, using FIG. 13, FIG. 14 (a), (b), and FIG. 15 (a), (b), a method for reading / writing the above-mentioned connection data from / to the connection memory 62 is realized. A configuration example will be specifically described.

FIG. 13 is a diagram showing the reading position of the document, and FIGS. 14A and 14B are diagrams showing the relationship of the timing signals relating to the image reading operation. Note that FIG. 14 (b)
14 is an enlarged view of a part shown in FIG.
Further, FIGS. 15A and 15B are control block diagrams for generating timing signals necessary for reading and writing band connection data to the connection memory 62.

In the present embodiment, the reading sensor 1 is 1
28 pixels (1 is the beginning, 128 is the end, and 1 pixel is R
It is composed of a pair of GB). As shown in FIG. 13, the first read operation of band 1 is line 1
From the first to the line n, the information of the original is read in units of 128 pixels for each line. Next, read sensor 1 to band 2
Then, the information of the original corresponding to the band 2 is read.

BVE shown in FIGS. 13 and 14 is a signal indicating the image effective period in the main scanning direction, and while reading one band (1st pixel of line 1 to 128th pixel of line n), It becomes "H" level. VE is a signal indicating an effective image area in the sub-scanning direction, and is at "H" level while reading one line. VD is image data. CLK is "H" / "L" in the cycle corresponding to one pixel
The clock signal which repeats a level is shown.

The counter 81 shown in FIG. 15A is a counter for making timing in the sub-scanning direction, counts the number of rising edges of the clock CLK, and outputs the count result as an m-bit signal. . When the clear terminal is at the "L" level, the count result is cleared and all m-bit outputs are at the "L" level.

The coincidence detecting circuit 82 compares the output value of the counter 81 with two predetermined values, and when one of them coincides, the output VE signal becomes "H" level, and when the other coincides. The VE signal is set to "L" level. The coincidence detection circuit 83 compares the output value of the counter 81 with two predetermined values, and when one coincides, the output is set to "H" level, and when the other coincides, the output is set to "L" level. To

Further, the coincidence detection circuit 84 includes the counter 8
The output value of 1 is compared with a predetermined value, and when the values match, the output is set to the "L" level. The AND circuit 85 outputs "L" when either the input signal BVE or the output of the coincidence detection circuit 84 becomes "L" level.
Become a level. The OR circuit 86 sets the output signal WR to the “L” level when both the input signal CLK and the output of the coincidence detection circuit 83 become the “L” level.

The address counter 87 shown in FIG.
Is a counter for making timing in the main scanning direction, counts the number of rising edges of the VE signal, and outputs the count result as an n-bit signal. Also,
When the clear terminal is at "L" level, the count result is cleared and all n-bit outputs are at "L" level.

Further, the connection memory 62 stores the band connection data as described above, and the OR circuit 89, when both the input signal WR and the least significant bit of the address counter 87 become "L" level. , Output signal SW
Set R to "L" level. The S-WR signal becomes "L" level at the timing of 128 pixels on the odd line, and S-WR signal
When the WR signal rises, the signal of the data bus is written in the connection memory 62.

The inverter 90 inverts the WR signal and supplies it to the RD terminal of the connection memory 62, and the bidirectional bus controller 91 responds to the S-WR signal by the connection memory 62.
It controls the data bus signal. That is, S-WR
When the signal is at the “L” level (timing of processing 128 pixels), the history color, the current color and the area are connected to each other and the memory 62
This history color, now color and area are output to the data bus of
At the rising timing of the S-WR signal, it is stored in the connection memory 62 as band connection data. When the S-WR signal is at the “H” level, the history color stored in the connection memory 62 is the history color, the current color, and the buffer that outputs the area are in a high impedance state on the data bus of the connection memory 62. The current color and area data are sent to the correction circuit 61 that processes the correction conditions.

Although not shown in the block diagram of FIG. 15, a circuit for using the history color, present color, and area data stored in the connection memory 62 only for the first pixel of each line, and There is a circuit for setting the correction conditions for the first pixel of the first band at the start of the reading operation to black for the history color, white for the current color, and normal for the area.

The connection data stored in the connection memory 62 includes history color (black, white, current 2 bits), current color (black, white, color 2 bits), area (normal, paint, line). , 2 bits of the line in the paint), a total of 6 bits.

Next, the correction conditions will be described.

From the left side of Table 5, the history color of the pixel of interest, the current color which is the color of the pixel of interest read, the area determined by the area determination in the main scanning direction, and the sub-scan already determined by the history code determination condition. 4 of the history code of the previous upper pixel in the direction
The area is corrected using one of the conditions, and the history code of the pixel of interest is changed.

Only the case where correction is applied in Table 5 will be described. When no correction is applied, it means that the area is correctly determined.

When it is determined that the history color is black, the pixel of interest is white, and the area is the paint area in the main scanning direction, and the upper history code is "0", the area of interest is too large for the area of interest to correctly paint the area. It corresponds when it cannot be completed. That is, it is unlikely that the pixel one above is white and is in the normal region, but the pixel below it is not in the paint region, so the pixel is corrected to the normal region.

Next, when it is determined that the history color is the marker color, the target pixel is white, and the normal region is in the main scanning direction, and the upper history code is "1", the target pixel correctly enters the paint region. If you can't do it. In other words, if the pixel one level above is in the paint area with white as the history color and is in the paint area, the pixel below it cannot be the normal area, so the pixel is corrected to the paint area.

Next, in the case where it is determined that the history color is the marker color, the target pixel is white, and the area is the paint area in the main scanning direction, and the upper history code is "0", the paint end marker is too large. It corresponds to the case where the paint area cannot be finished. That is, it is not possible that the pixel above is white and is a normal area, but the pixel below it is a paint area, so the pixel is corrected to a normal area.

Finally, when the pixel of interest is in the line area with the marker color and the upper history code is 3, the line area and the paint area are adjacent to each other. to correct.

By performing the above-described processing, a gap or protrusion exceeding the allowable amount as described with reference to FIG. 7 is temporarily in an erroneous area, but is corrected after that to obtain a correct image. Output is possible.

FIG. 16 is a diagram showing another embodiment of the color image processing apparatus of the present invention.

As described above, the band connection data is
It consists of a total of 6-bit signals: history color (black, white, color 2 bits), current color (black, white, color 2 bits), area (normal, paint, line, line within paint) There is. When a general-purpose product is used as the SRAM of the connection memory 62 that stores this connection data (for example, 8 Kbytes or 3
An SRAM composed of 2 Kbytes), generally a data bus is composed of 8 bits. Therefore, when a general-purpose SRAM is used as an element for storing 6-bit concatenated data, there is a wasteful portion where 2 bits are not used for each address.

With respect to this point, in recent years, when image processing is carried out by a digital hardware circuit, it is generally performed by using an ASIC technique to form a dedicated integrated circuit. Particularly, a small-scale SRAM can be incorporated as one element in an integrated circuit, and it is not necessary to use a general-purpose product. Under such circumstances, SRA
When storing the same information in M, how to reduce the capacity is becoming important.

From the above point of view, the present embodiment is based on the connection memory 6
In order to reduce the capacity of 2, the connection data is coded and stored. That is, the connection data includes history color (black, white, and 2 bits of color), current color (black, white,
2 bits of color, area (normal, paint, line,
It is not necessary to use a total of 6 bits (2 bits of the line in the paint), and it is only necessary to know the states of the 6 types of history codes shown in the history code determination conditions in Table 4, so store them as 3 bit data in the memory. Is possible.

Therefore, as shown in FIG. 16, a connection data encoding circuit 92 is provided, and history color (black, white, 2 bits of color), current color (black, white, 2 bits of color), area (normal). , Paint, line, line within paint)
6-bit signal is input, converted into a 3-bit history code, and sent to the bus controller 91.

The connection data encoding circuit 92 is composed of a ROM, and the history color, current color, and area signals are connected to the ROM address. This ROM stores information for converting all combinations of previously determined history colors, current colors, and areas into history codes. The history code is output from the data bus of the ROM as band connection data and stored in the connection memory 62 via the bus controller 91.

[0093]

As described in detail above, according to the present invention, since the number of line information stored in the correction information means is smaller than the number of pixels in the line direction, the capacity of the correction information storage means can be reduced. As a correction information storage means,
It is possible to use the SRAM incorporated as one element in the integrated circuit.

In addition, the pixel unit used by the correction information storage means is compared with the pixels in the line direction used by the area processing means for performing two-dimensional area processing on the codes classified into a plurality of color codes. By reducing the number, it is possible to easily and accurately reduce the capacity of the correction information storage unit.

Moreover, since the information to be read from and written to the correction information storage means is coded, the capacity of the correction information storage means can be further reduced.

[Brief description of drawings]

FIG. 1 is a block diagram showing a schematic configuration of image processing of a full-color copying machine according to an embodiment of a color image processing apparatus of the present invention.

FIG. 2 is a diagram showing a marker editing specification.

FIG. 3 is a diagram illustrating a marking rule.

FIG. 4 is a diagram showing an internal configuration of a marker editing circuit 7.

FIG. 5 is a diagram showing stored data at the time of area determination.

FIG. 6 is a diagram showing a tailing process.

FIG. 7 is a diagram showing a malfunction due to a rule violation.

FIG. 8 is a diagram showing a correction method.

FIG. 9 is a diagram showing a relationship between an image and a history code (No. 1).

FIG. 10 is a diagram showing a relationship between an image and a history code (No. 2).

FIG. 11 is a diagram showing another relationship between an image and a history code (No. 1).

FIG. 12 is a diagram showing another relationship between an image and a history code (No. 2).

FIG. 13 is a diagram showing a reading position of a document.

FIG. 14 is a diagram showing a relationship of timing signals regarding an image reading operation.

FIG. 15 is a control block diagram for generating a timing signal required for a read / write operation to the connection memory 62.

FIG. 16 is a diagram showing another embodiment of the color image processing apparatus of the present invention.

[Explanation of symbols]

 1 CCD line sensor 7 Marker editing circuit 22 Black and white determination circuit 23 Color determination circuit 24 Coding circuit 25 Black density generation circuit 30 Sub-scanning direction noise removal circuit 40 Main scanning direction noise removal circuit 51 Area determination circuit 52 FIFO memory 61 Area determination correction Circuit 62 Connection Memory 72 Print Color Judgment Circuit 73 Density Generation Circuit 92 Connection Data Encoding Circuit

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification number Reference number within the agency FI technical display location H04N 1/46 H04N 1/46 Z (72) Inventor Eiichi Motoyama 3-30 Shimomaruko, Ota-ku, Tokyo No. 2 Canon Inc. (72) Fumio Mikami 3-30-2 Shimomaruko, Ota-ku, Tokyo Canon Inc. (72) Inventor Shigeo Yamagata 3-30-2 Shimomaruko, Ota-ku, Tokyo Kya Non-Incorporated (72) Inventor Koji Arai 3-30-2 Shimomaruko, Ota-ku, Tokyo Canon Inc. (72) Inventor Takashi Nonaka 3-30-2 Shimomaruko, Ota-ku, Tokyo Canon Inc.

Claims (3)

[Claims] 1. A document is divided into a plurality of bands and read,
A reading sensor that outputs multi-valued data, a coding means that classifies the multi-valued data into a plurality of color codes that represent colors, and a two-dimensional area processing is performed on the codes that are classified into the plurality of color codes. Area processing means, storage means for storing processing information of one line before, reading means for reading the storage contents of the storage means, and data read by the reading means and output data of the area processing means Output code determining means for determining a plurality of output codes by
The correction code for correcting the output code of the output code determining means according to the condition, the correction information storage means for storing the correction information of the band before one, which is connected to the correction means, and the processed information of the sensor arrangement direction are Information retrieval means for retrieving processed information in the sensor arrangement direction for the head of a non-existing band from the correction information storage means, and information writing means for writing the correction information of the next band in the correction information storage means,
Storage control means for storing the correction code output from the correction means in the storage means as processing information for one line before, and output conversion means for converting the output code corrected by the correction means into multivalued data and outputting the multivalued data. In the color image processing apparatus including the above, the number of line information stored in the correction information unit is smaller than the number of pixels in the line direction. 2. The color image processing apparatus according to claim 1, wherein the number of pixel units used by the correction information storage unit is smaller than the number of pixels in the line direction used by the area processing unit. 3. The color image processing apparatus according to claim 1, wherein the information to be read from and written to the correction information storage means is coded.