JPH05130438A - Picture processing unit - Google Patents

Abstract

PURPOSE:To improve the picture quality when a color is patterned by preventing mixing existence of plural patterns at a color change point. CONSTITUTION:The processing unit is provided with a color discrimination means 109 discriminating a specific color from an input picture signal, a pattern generating means 110 generating a relevant graphic pattern decided for each color in response to the color discrimination result by the color discrimination means 109 and a misdiscrimination prevention means 140 preventing an error in the color discrimination by using plural color discrimination results by the color discrimination means 109.

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, for example, an image processing apparatus such as an image scanner, a printer, a digital copying machine, a facsimile machine or the like.

[0002]

2. Description of the Related Art Conventionally, for example, in a digital copying machine, an original is irradiated with a light source such as a halogen lamp, and light reflected from the original is photoelectrically converted by using a solid-state image pickup device such as CCD (charge coupled device). After being converted into a digital signal and subjected to predetermined correction processing and the like, a recording image is formed by using the signal with a recording device such as a laser beam printer, a liquid crystal printer, a thermal printer, an inkjet printer.

Further, in such a digital copying machine, an output having a large amount of information is required due to colorization of an original to be copied, and a digital full color copying machine or a one-point color copying machine has been developed. .. On the other hand, when outputting a color original with a black and white output means, the color original is read by the color sensor, the color is discriminated, and the pattern corresponding to the color is output from the color discrimination signal.
Techniques for expressing the difference in color depending on the difference in pattern have also been proposed.

[0004]

However, in a device that discriminates colors, generates a pattern corresponding to the colors, and outputs the patterns, a point where two kinds of colors are mixed or a color is continuously changed. Has a problem that two kinds of patterns are mixed at the change point, and the output is hard to see. Further, in the case of an intermediate color between the two types of judgment colors, there is a problem that two types of patterns are mixed and it becomes difficult to see.

An object of the present invention is to solve the above-mentioned problems, and to provide an image processing apparatus capable of preventing the mixture of patterns at a color change point and improving the image quality when patterning a color. There is something to do.

[0006]

In order to solve the above problems, the image processing apparatus of the present invention provides a color discrimination means for discriminating a specific color from an input image signal and a color discrimination result of the color discrimination means. An erroneous determination preventing means for preventing an error in color determination by using a plurality of color determination results by the color determining means, and a pattern generating means for generating a corresponding graphic pattern previously determined for each color. It is characterized by having.

[0007]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The embodiments of the present invention described below consider the image information in a certain block in a color discriminating means for discriminating the color of an input image from an input color image signal, and when several types of patterns are mixed therein. Is to output one of the patterns with priority.

[0008]

Embodiments of the present invention will now be described in detail with reference to the drawings.

First Embodiment (Overall Configuration) FIG. 1 shows an example of the overall circuit configuration of an image processing apparatus of a digital copying apparatus to which the present invention is applied. Figure 1
In the figure, 101 is an image sensor (color reading sensor) of an image pickup means for converting a color original image formed through an optical system such as a rod lens into G (green), B (blue) and R (red) electric signals. ) As CCD
(Charge coupled device) and has a color separation filter. 1
Reference numeral 02 is an amplifier circuit for amplifying the output signal of the CCD 101, 1
Reference numeral 03 is a coaxial cable, and 104 is a sample hold (S / H) that samples and holds (S / H) the color image signal output from the amplifier circuit 102 via the coaxial cable 103 to separate it into three colors of G, B, and R. ) Circuit. Reference numeral 105 denotes an analog / digital (A / D) conversion circuit for converting the analog color image signal sample-held by the sample-hold circuit 104 into a digital color image signal, and 106.
Reference numeral 107 denotes a shift correction circuit that electrically corrects a shift in the reading position between the channels of the CCD 101, and reference numeral 107 denotes a black correction / white correction circuit that performs a black level correction and a white level correction (shading correction) described later on the digital image signal. Is.

Reference numeral 108 denotes a luminance signal generating section, which generates a luminance signal from the black-corrected and white-corrected digital color image signal. Reference numeral 109 denotes a color discrimination circuit, which performs color discrimination for each pixel of the digital color image that has been black-corrected and white-corrected, and outputs a color determination signal according to the hue signal in this example. Reference numeral 110 denotes a pattern generation circuit including a RAM or ROM storage device, which generates a predetermined graphic pattern corresponding to the color of each pixel based on the color discrimination result of the color discrimination circuit 109. In this example, the pattern generation circuit 110
Reads out and outputs one of the graphic patterns stored in advance using the color determination signal corresponding to the hue signal as a read address.

Reference numeral 111 denotes a pattern synthesizing circuit for synthesizing the luminance signal generated by the luminance signal generating unit 108 and the graphic pattern generated by the pattern generating circuit 110 and indicating the number of colors. Reference numeral 112 denotes a Log conversion unit that converts the output signal of the pattern synthesis circuit 111 into a density signal and outputs it to various connected printers. The portion A surrounded by a chain line in FIG. 1 corresponds to a video image processing circuit of an image reading device (image scanner).

The image copying apparatus of the present example exposes a full-color original document with an illumination source such as a halogen lamp or a fluorescent lamp (not shown), picks up a reflected color image with a color image sensor such as a CCD, and obtains an analog image. It is a device for obtaining an image by digitizing the signal with an A / D converter or the like, processing and processing the digitized full-color image signal, and outputting it to a thermal transfer printer, an inkjet printer, a laser beam printer or the like (not shown). The details are as described below.

(Color Reading Sensor) An original is first illuminated by an exposure lamp (not shown), and the reflected light is color-separated and read by the color reading sensor 101 for each image, and amplified by an amplifier circuit 102 to a predetermined level. Here, the driving of the CCD 101 is generated by a system pulse generator (not shown).

2 and 3 show a color reading sensor and driving pulses. FIG. 2 shows a color reading sensor used in this example. This sensor uses 63.5 μm as one pixel (400
dot / inch (hereinafter referred to as dpi) 1024 pixels, that is, 1 pixel in the main scanning direction as G, B, R
Since it is divided into three, total 1024 × 3 = 307
It has an effective pixel count of 2. On the other hand, the chips 58 to 62 of this sensor are formed on the same ceramic substrate, the 1st, 3rd and 5th (58a, 60a, 62a) of the sensor are on the same line LA, and the 2nd and 4th are 4 and LA. Line ((6
3.5 μm × 4 = 254 μm) apart from each other on the line LB, and the document is scanned in the direction of arrow AL when reading.

Of the five CCDs, the first, third, and fifth CCDs are driven by the drive pulse group 0DRV118a, and the second and fourth CCDs are driven by the drive pulse group EDRV119a independently and synchronously. The pulse signals O01A, O02A, ORS included in the drive pulse group 0DRV118a and the pulse signal E01 included in the drive pulse group EDRV119A
A, E02A, and ERS are charge transfer clocks and charge reset pulses in each sensor, respectively, so that mutual interference between 1st, 3rd, 5th and 2nd and 4th, and noise limitation are prevented so that they are not in jitter with each other. It is generated in perfect synchronization. Therefore, these pulses are generated by one reference oscillation source 0S (not shown).
It is generated from C.

FIG. 4A shows the above-mentioned drive pulse group 0DRV.
118a and a circuit block for generating EDRV 119a are shown, and FIG. 4B shows the timing. This circuit block is included in a system control pulse generator (not shown). A clock K0135 obtained by dividing the source clock CLK0 generated from the single reference oscillation source 0SC.
a is a sensor drive pulse 0DRV and a sensor drive pulse ED
Reference signals SYNC2, S that determine the generation timing of RV
This is a clock for generating YNC3. These reference signals SYNC2. SYNC3 is a presettable counter 64 set by the signal line 22 connected to the CPU bus.
The output timing is determined according to the set values of a and 65a. The reference signals SYNC2 and SYNC3 are divided by the frequency divider 66.
a, 67a and drive pulse generators 68a, 69a are initialized.

That is, the reference signals SYNC2 and SYNC
3 is a horizontal cycle signal HSYNC1 input to this block
Since all are generated by the source clock CLK0 output from one transmission source 0SC558a and all divided clocks generated in synchronization with 18 as a reference,
The respective pulse groups of RV118a and EDRV119a are obtained as synchronized signals without any jitter, and the disturbance of signals due to interference between sensors can be prevented.

Here, the sensor driving pulse 0DRV118a obtained in synchronization with each other is applied to the first, third and fifth sensors 58a, 60a and 62a, and the EDRV 119a is applied to 2,4.
The second sensor 59a, 61a is supplied to each sensor 58
Video signals V1 to V5 are independently output from a, 59a, 60a, 61a, and 62a in synchronism with the drive pulse, and the amplifier circuits 501- shown in FIG.
1 to 501-5 is amplified to a predetermined voltage value, and the signals V1, V3 and V5 are sent out at the timing of 00S129a in FIG. 3 through the coaxial cable 101a, and E0S134a.
The signals V2 and V4 are sent out at the timing of and input to the video image processing circuit.

(Sample and Hold Circuit) A color image signal obtained by reading an original input to the video image processing circuit in five divisions is sample and hold circuit S / H1.
In 04, it is separated into three colors of G (green), B (blue), and R (red). Therefore, after S / H, 3
Signal processing of x5 = 15 systems is performed.

(A / D conversion circuit) The analog color image signal sampled and held for each color R, G, B by the S / H circuit 104 is 1 in the next stage A / D conversion circuit 105.
Digitized every 5 channels and output to the next stage in parallel independently for each 1-5 channel.

(Displacement Correction Circuit) In this embodiment, as described above, four lines (63.5 μm × 4 = 254 μ) are used.
m) is provided in the sub-scanning direction and the original is read by the five staggered sensors divided into five areas in the main scanning direction. In 5, the reading position is misaligned. Therefore, in order to correctly connect this, the shift correction circuit 106 including memories for a plurality of lines performs the shift correction.

(Black Correction / White Correction Circuit) Next, FIGS.
The black correction operation in the black correction / white correction circuit 107 will be described with reference to FIG.

As shown in FIG. 6, the black level outputs of channels 1 to 5 have large variations between chips and between pixels when the amount of light input to the sensor is small. When this is output as it is and the image is output, lines (streaks) and unevenness (spots) appear in the data part of the image.
Occurs.

Therefore, it is necessary to correct the output variation of the black portion, and the variation is corrected by the black correction circuit as shown in FIG. Prior to the document reading operation, the document scanning unit is moved to the position of the black plate having a uniform density arranged in the non-image area at the front end of the document table, the halogen is turned on, and the black level image signal is input to this circuit.

Regarding the blue signal B _IN , one line of this image data is stored in the black level RAM 78a.
The selector 82a selects A (d), closes the gate 80a (a), and opens 81a. That is, the data line is 151a
→ 152a → 153a, while RAM78a
The address input 155a is initialized by inverted HSYNC, c is output to the selector 83a so that the output 154a of the address counter 84a that counts VCLK is input, and the black level signal for one line is stored in the RAM 78.
It is stored in a (hereinafter referred to as a black reference value capture mode).

At the time of image reading, the RAM 78a is in the data reading mode and the data line 153a → 157a.
In the path of, the line B is read out and input for each line and pixel to the B input of the subtractor 79a. That is, at this time, the gate 81
a is closed (b) and gate 80a is open (a). Further, the selector 86a has an A output. Therefore, the black correction circuit output 156a outputs B _IN (i) -DK (i) = B in the case of a blue signal, for example, with respect to the black level data DK (i).
_It is obtained as _OUT (i) (hereinafter referred to as the black correction mode).

Similarly, green G _IN and red R _IN are 7
Similar control is performed by 7aG and 77aR. Also,
Control lines a, b, c of each selector gate for this control
d and e are controlled by the CPU by the latch 85a assigned as I / 0 of the CPU (not shown). The RAM 78a can be accessed by the CPU by selecting B from the selectors 82a, 83a and 86a.

Next, the white level correction (shading correction) in the black correction / white correction circuit 107 will be described with reference to FIGS. The white level correction corrects variations in sensitivity of the illumination system, the optical system, and the sensor based on white data obtained when the original scanning unit is moved to a uniform white plate position and irradiated. The basic circuit configuration is shown in FIG. Although the basic circuit configuration is the same as that of FIG. 5, the subtractor 79a performs the correction in the black correction, while the multiplier 79 'in the white correction.
Since only the point of using a is different, description of the same parts is omitted.

CCD for reading the original at the time of color correction
When 101 is at the reading position (home position) of the uniform white plate, that is, before the copying operation or the reading operation, the exposure lamp (not shown) is turned on, and the uniform white level image data is stored in the correction RAM 78'a for one line. Store. For example, if the width in the longitudinal direction of A4 in the main scanning direction is 16 p / mm, 16 × 297 mm.
= 4752 pixels, that is, at least the RAM capacity is 4
It is 752 bytes, and as shown in FIG. 8A, if the white plate data Wi of the i-th pixel is Wi (i = 1 to 4752), R
As shown in FIG. 8B, the AM 78 'stores data for the white plate for each pixel.

On the other hand, for Wi, the corrected data Do = Di × for the read value Di of the normal image of the i-th pixel
It should be FF _H / Wi.

From the CPU, the latch 85'aa ',
Gates 80'a and 81'a are opened for b ', c', and d ', and the selectors 82'a, 83'a, 86'a output so that B is selected, and the RAM 78'a is C.
Enables PU success.

Next, in the procedure shown in FIG. 9, the CPU sequentially calculates FF _H / W _{O for} the _first pixel W _O , FF / W ₁ for W _1, and replaces the data. When the blue component of the color component image is finished (Step B in FIG. 9), similarly, the green component (Step G) and the red (Step R) are sequentially performed, and thereafter, D _O = for the input original image data Di.
Gate 80'a so that Di × FF _H / Wi is output
Is opened (a '), the gate 81'a is closed (b'), and the selectors 83'a and 86'a are selected as A, whereby the RAM is selected.
The coefficient data FF _H / Wi read from 78'a passes through the signal line 153a → 157a, is multiplied by the original image data 151a input from one side, and is output.

As described above, the black level and the white level are corrected based on various factors such as the black level sensitivity of the image input system, the dark current variation of the CCD, the sensitivity variation between the sensors, the optical system light amount variation and the white level sensitivity. Then, the image data B _OUT 121, G _OUT 122, and R _OUT 123 that are uniformly corrected for each color of white and black are obtained over the main scanning direction.

(Luminance signal generator) Black-corrected and white-corrected image data R _OUT 121, G _OUT 122, B _OUT 123
Is input to the luminance signal generation unit 108 and the color discrimination circuit 109.

In the brightness signal generator 108, the sensor 10
ND the averaged filter image read in 1.
I'm making an image. FIG. 10 is for explaining the above-mentioned operation, and the input image data R _OUT 121,
First, the adder 201 adds the G _OUT 122 and the B _OUT 123, respectively. After that, the divider 202 divides it by 1/3 and outputs it.

(Color Discrimination Circuit) In this embodiment, a hue signal is used as the color discrimination method. This is for accurate determination even when the same color has different saturation or brightness.

First, the outline of the color discrimination method will be described.

The input R, G, B data are each 8 bits and have information of a total of 2 ²⁴ colors. Therefore, using such enormous information as it is becomes expensive from the circuit scale. In this embodiment, the hue described above is used. This is exactly
Although it is different from the hue normally displayed, it is referred to as a hue here. The color space is known as Munsell's solid, etc.,
It is known to be represented by saturation, lightness, and hue.
First, it is necessary to convert the R, G, B data into plane, that is, two-dimensional data. Common part of R, G, B,
That is, since the minimum value min (R, G, B) of R, G, B is an achromatic component, min (R, G, B) is subtracted from each R, G, B data, and the remaining information is The three-dimensional input color space was converted into a two-dimensional color space by using it as a chromatic color component. As shown in FIG. 11, the converted plane is divided into 6 from 0 ° to 360 °, and the order of the sizes of R, G, and B to be input, that is, R>G> B, R>.
B> G, G>B> R, G>R> B, B>G> R, B> R
The hue value is obtained by using a LUT (look-up table) or the like based on the information of> G and the maximum value and the intermediate value of R, G, and B that are input. Next, the actual operation of the color discrimination circuit 109 will be described with reference to FIG.

The input R, G, B data is first of all a max / mid / min detection circuit 14 for discriminating its magnitude.
01 is input. This compares each input data using a comparator, and depending on the comparison result, the max value, mid
The value and the min value are output. Also, the output value of the comparator is output as a ranking signal. Output max, m
As described above, the id and min values subtract the achromatic color component from the max value and the mid value.
3 which is the minimum value from the max value and the mid value.
The n value is subtracted and input to the hue detection circuit 1404 together with the ranking signal. The hue detection circuit 1404 is a randomly accessible storage element such as RAM or ROM.
In this embodiment, the look-up table is constructed by using the ROM. A value corresponding to the angle of the plane as shown in FIG. 11 is stored in advance in the ROM, and the hue value is output according to the input order signal (max-min) value and (mid-min) value. It The output hue value is then input to the window comparators 1405 and 1406. To these comparators 1405 and 1406, color data that is originally desired to be patterned is input by a data input means (not shown), and a hue data value that matches the color is input to the CPU 1.
A desired offset is given by 407 and set in comparators 1405 and 1406. Comparator 14
In 05, assuming that the set value is a ₁ , “1” is output for (hue data) <a ₁ for the input hue data, and the comparator 1406 sets the set value to a ₂
Then, “(1)” is output when (hue data)> a ₂ .

[0040] Thus, the subsequent AND gate 1410, when a ₁ <(hue data) <a _2, "1" is outputted from the color detection circuit 1410.

Similarly, the circuits 14 (b) and 14 (c) output color determination signals for other hues.

(Majority Decision Circuit) The decision signal for color decision is
The majority circuit 140 determines which color is determined to be the closest within a certain area. Here, the majority circuit will be described with reference to FIGS. In FIG.
C1, C2 and C3 are color determination signals determined in FIG. 12, VCK is a video transfer clock, AREA is a signal indicating a certain area, and 2001, 2002 and 20 are shown.
The counter 03 is cleared for each fixed area. In other words, the 2001 counter is within a certain area,
This is a circuit that counts how many colors are determined as C1, and 2002 and 2003 similarly count the number of colors of C2 and C3. The counting result is input to FIGS. 14A, 14B and 14C. 2101, 2102 and 2103 are both comparators, and 2104 to 2109 are logical products, 2110 to 21
12 is a logical sum, 2113 is an encoder circuit, 2114 is a D flip-flop circuit, and AREA is AR in FIG.
Similar to the EA signal, it is a constant area signal for making a majority decision. For each A, B, and C input, the logical product circuit 2104 outputs 1 when A ≧ B ≧ C, and similarly, the logical product 2105 outputs 1 when A>C> B. .. The logical sum circuit 2110 outputs "1" only when A is the largest among A, B, and C. Similarly, the OR circuit 2111 outputs 1 when B is maximum, and the OR circuit 2112 outputs 1 when C is maximum. 2113 is an encoding circuit, which is 1 when A is maximum and 2 when B is maximum.
When C is maximum, 3 is output. The DF / F 2114 latches this code signal for each constant area.
Outputs which color is most determined for each fixed area.

(Pattern Generating Circuit) Pattern Generating Circuit 1
10 will be described with reference to FIG. R for pattern
In the OM 803, for example, a dot pattern corresponding to each color as shown in FIG. 16 is written in advance. Each graphic pattern has 16 × 16 dots as one pattern. The pattern ROM 803 performs pattern generation processing by repeatedly outputting this pattern in the main scanning direction and the sub scanning direction according to the color determination signal. The main scanning counter 802 operates by counting the video clock CLK in synchronization with the horizontal synchronizing signal HSYNC, and the sub-scanning counter 801 operates by synchronizing with the ITOP signal and counting the horizontal synchronizing signal HSYNC.

The outputs of the counters 801 and 802 are 4 bits each, and the color determination signal is 5 bits, for a total of 13 bits.
t is input as the address of the pattern ROM 803. That is, it is configured such that 32 types of patterns of 16 dots × 16 dots can be generated for (the type of) the read color.

The output from the pattern ROM 803 is 8
It has a data length of bit, and the MSB (most significant bit) therein is used as a control signal (HIT signal) in the pattern synthesizing unit 111 described later, and is stored in the ROM.
Normally 0 for 803, and always MS when generating a pattern
Data is written so that B becomes 1.

Naturally, the above-mentioned pattern ROM 803 is
RAM or the like may be used. Even if a RAM or the like is used, its capacity and bit allocation of addresses are the same as those of the ROM.

(Pattern Synthesis Unit) Pattern Generation Circuit 11
The pattern signal output from 0 and H indicating the judgment area
The IT signal and the brightness signal from the 108 brightness signal generation unit are input to the selector 1601 in FIG. That is, the selector 1601 selects a pattern signal based on the color obtained as a result of the majority decision for each area of a certain size in the determination area, and selects a luminance signal in the outside of the determination area.

(Log Transforming Unit) The image data that has undergone the addition process is converted into a density signal by the Log converting unit 112 in FIG. 1 so as to perform the brightness-density conversion. This Log converter 11
In No. 2, a lookup table using a ROM is used. The signal converted into the density signal by the Log conversion unit 112 is output to a monochrome printer unit (for example, a laser beam printer) which should form an image.

The effect of the majority voting system will be described with reference to FIG. Here, a majority decision is made for a region having a size of 3 pixels in the vertical direction and 3 pixels in the horizontal direction, and a pattern is selected. 2201,
A and B in 2202 and 2203 respectively indicate different color determination results. For example, if it is determined that all the colors are A within a certain area such as 2202, patterning like 2205 is performed. Further, if it is determined that the color in the constant area is B as in 2203, patterning in 2206 is performed. Further, when there is a mixture of a portion determined to be A and a portion determined to be B as in 2201, 1
When pixel-by-pixel patterning, two types of patterns coexist like 2204, but if a majority method is used, for example, 2201 has more B colors, resulting in 2207.
The pattern B is selected as shown in. Generally, for example, in the case of making a red judgment in a red color area, the erroneous judgment is at most one pixel in several pixels or tens of pixels. Therefore, by adopting the majority decision method, the mixture of patterns due to the erroneous judgment is considerably reduced. I can do things.

[Second Embodiment] As another embodiment for saving erroneous judgment of color patterning, a system for referring the judgment result of the previous pixel to the judgment of the next pixel will be described. The circuit configuration is basically the same as that of the first embodiment, but the majority circuit 140 in FIG. 1 is deleted and a DF / F is added instead. This part will be described with reference to FIG. The color of a certain pixel is determined by the color determination circuit 2301 and the result is input to the DF / F 2302. Where V
CK is an image transfer clock. The color determination result delayed by one pixel by the DF / F 2302 is returned to the color determination circuit 2301. In addition, this returned information is specifically shown in Figure 1.
It is input to the second hue detection circuit 1404. Since the hue detection circuit 1404 is composed of the look-up table of the RAM or the ROM as described above, the returned information is input to this address. In other words, when the previous pixel is determined to be red, it is possible to prevent misjudgment from occurring by writing in the lookup table information that makes it easy to determine the next pixel to be red.

[Third Embodiment] As another embodiment for saving an erroneous judgment of color patterning, for pixels around the judgment pixel,
An example in which a histogram of the color determination signal is obtained and the color determination of the pixel of interest is performed with reference to the result will be described.

FIG. 20 is a block diagram for explaining this embodiment well. In this embodiment, the majority circuit 140 is deleted from the first embodiment and the color discrimination circuit 109 is replaced. The color of the input color image information is determined by the 2501 color determination circuit from the ratio of R, G, and B. This determination method is similar to that of the color determination circuit of FIG. 12 in the first embodiment, and in this embodiment, the determination result is encoded.
It is output as a color code signal of bit. The determination result is decoded by the multiplexer 2502, and 2503
The counters 2506 through 2506 count the number of color determination pixels in a certain area. The clear signal in the figure is an area signal indicating a certain area, and the color determination information is counted for each area. Reference numerals 2511 to 2514 are comparators, and 2507 to 2510 are registers that determine the slice level of the comparator. The register set value is set according to the size of the area, and the slice level for each color can be changed according to the characteristics of the CCD. The outputs of the comparators 2511 to 2514 are 25
It is input to the 15-color conversion circuit. Further, the output of the 2501 color determination circuit is configured by the 2516 delay circuit so as to match the timing with the image delay of the counter register comparator described above. The 2515 color conversion circuit is a portion for correcting the 2501 color determination result with reference to the histogram, and the details will be described later. The finally determined result is converted from a color code signal into a bit-corresponding signal by the 2517 decoding circuit, and then input to the 110 pattern generation circuit in the first embodiment.

The 2515 color conversion circuit will be described with reference to FIG. FIG. 21A is a table showing a color conversion method, and FIG. 21B shows a color space of 0 to 3 and 4 colors. In FIG. 21A, C ₀ , C ₁ , C ₂ , and C ₃ respectively indicate the outputs of the comparators 2511 to 2514, and IN is the input color determination signal. The basic idea is that if the histogram value of the same color as the 2501 color determination result for the pixel of interest is equal to or higher than the slurry level, then the determination result is not converted and the final color determination result is given as is. , If the histogram value is below the slurry level, refer to the histogram values of other colors, and if the histogram of adjacent colors in the color space is other than a certain value, convert the determination result to the adjacent color, and otherwise In the case of, the judgment result is directly used as the final color judgment result.

As described above, according to the above-described embodiment, when the color determination is performed, the color determination signal is majority-determined within a fixed size to determine the color, or the determination result of the previous pixel is By adopting the method that is referred to when determining the pixels, it is possible to prevent the mixture of patterns due to erroneous determination of colors.

[0055]

As described above, according to the present invention, it is possible to prevent the mixture of patterns at the color change points and improve the image quality when patterning the colors.

[Brief description of drawings]

FIG. 1 is an overall block diagram of a first embodiment of the present invention.

FIG. 2 is a diagram illustrating a CCD sensor.

FIG. 3 is a diagram illustrating CCD driving.

FIG. 4 is a diagram illustrating CCD driving.

FIG. 5 is a diagram illustrating black correction.

FIG. 6 is a diagram illustrating black correction.

FIG. 7 is a diagram illustrating white correction.

FIG. 8 is a diagram illustrating white correction.

FIG. 9 is a diagram illustrating white correction.

FIG. 10 is a diagram showing luminance signal generation.

FIG. 11 is a diagram for explaining hue discrimination.

FIG. 12 is a block diagram of a color discrimination circuit.

FIG. 13 is a diagram showing a majority decision circuit.

FIG. 14 is a diagram showing a majority decision circuit.

FIG. 15 is a block diagram of a pattern generation circuit.

FIG. 16 is a diagram showing correspondence of color patterns.

FIG. 17 is a block diagram showing pattern synthesis.

FIG. 18 is a diagram showing an effect of a majority decision method.

FIG. 19 is a block diagram showing a second embodiment of the present invention.

FIG. 20 is a block diagram showing a third embodiment of the present invention.

FIG. 21 is a view showing the concept of the third embodiment of the present invention.

[Explanation of symbols]

 190 Color Discrimination Circuit 140 Majority Decision Circuit 110 Pattern Generation Circuit 111 Pattern Synthesis Circuit

Claims (4)

[Claims] 1. A color discriminating means for discriminating a specific color from an input image signal, and a pattern generating means for generating a corresponding graphic pattern predetermined for each color according to a color discriminating result of the color discriminating means. An image processing apparatus comprising: an erroneous determination prevention unit that prevents an error in color determination by using a plurality of color determination results by the color determination unit. 2. The erroneous judgment preventing means makes a majority decision of a plurality of color judgment results in a predetermined area,
The image processing apparatus according to claim 1, wherein the color in the area is determined. 3. The image processing apparatus according to claim 1, wherein the erroneous determination prevention unit takes the determination result of the previous pixel into consideration when determining the next pixel. 4. The image processing apparatus according to claim 1, wherein the erroneous determination preventing unit prevents the erroneous determination of the determination pixel by taking a histogram of the color determination signal around the determination pixel.