JPH01220575A - Color image processor - Google Patents

Abstract

PURPOSE:To realize a real time processing and to prevent an image from being deteriorated by deciding a threshold value data from the maximum value and the minimum value on a line. CONSTITUTION:The spectral characteristic of a dichroic mirror 34 is selected so that the red color separation image of the image information of an original 21 can be separated from the prism 32 side of a spectral system 31 and the cyan color separation image from the prism 33 side. Next. the images are photoelectrically transduced at CCDs 35 and 36, and are converted to digital signals by A/D converters 37 and 38. The chrominance signals VR and VC normalized by a white signal are separated at every image element by a color discrimination ROM39 to color code data and density data, and a multileveled signal on which a prescribed signal processing is applied is outputted from an image processing circuit 40, and it is recorded as a color image by a recorder 78 or is stored in a memory 78. At the circuit 40, the optimum threshold value for a single chrominance signal is calculated at every line in a real time, then, the density data is multileveled In such a way, since the threshold value data is decided from the maximum value and the minimum value, the real time processing can be realized and the image can be prevented from being deteriorated.

Description

Detailed Description of the Invention [Field of Industrial Application] The present invention relates to an image processing device suitable for application to a color image recording device such as a color copying machine or a laser printer.
The present invention relates to a color image processing device that determines a multivalue threshold for a single color signal corresponding to each line.

[Background of the Invention] In an image recording device such as a color copying machine or a laser printer, a binary image signal, for example, black and white, corresponding to an original image is formed, and the image No. 48 is supplied to the main body of the recording device to reproduce the original image. I try to record it.

FIG. 10 shows a conventional example of an image processing apparatus 20 of this type.

In the figure, an original image 21 is supplied to an image reading means, for example a CCD 35, via an optical system 22, whereby the optical image is converted into a predetermined image signal for each scanning line. The image signal is converted into a digital signal of a predetermined number of bits by the A/D converter 37, and then supplied to the multi-value conversion means 70 in order to match the writing format of the recording device 78.

In this example, the image signal is converted into a black and white binary signal (hereinafter referred to as a binary signal).

The multi-value conversion means 70 is composed of a binarization circuit 76 and a threshold value output means 77. A reference signal having a predetermined threshold level is supplied from the threshold value output means 77, and the image signal is converted into a binary value based on this. be converted into

The binary code can be supplied to a recording device 78 or stored in an internal memory 79.

In the case of a color copying machine that uses a laser as the recording device 78, the binary code can be supplied as a laser modulation signal, and the optical signal modulated by this is transmitted in a predetermined direction on the photoreceptor drum provided in the color copying machine. By scanning at a predetermined speed, a binary electrostatic latent image is formed on the photosensitive drum.

The electrostatic image is developed and fixed to form an original image on recording paper.

If the document to be recorded is in color, it is separated into a plurality of color separation bodies, each of which undergoes photoelectric conversion, and then the signal processing described above can be performed. Then, each of them is multi-valued using the same threshold value.

By the way, the predetermined threshold value outputted from the threshold value output means 77 used in the multi-value conversion means 70 described above can generally be created based on the following criteria.

The first example is a case where the threshold value T is determined from the density histogram.

To do this, the original image is preliminarily scanned to obtain density information of the original image. That is, a density histogram is created. An example of the density histogram is shown in FIG.

From the density histogram, peak value nickel density data VA and VB for the white side and the black example are calculated. Next, the threshold value T is determined by substituting these peak concentration data VA and VB into the following equation.

T=1/3 (VB-VA)+VA For fil/3, a value of 1/3 to 2/3 may be preferable depending on the document.

The second example is the minimum value and maximum value V as shown in FIG.
This method calculates the threshold by finding A and VB and substituting them into the following equation.

T=1/2 (VB-VA)+vA Here, (VB-VA) represents the density width DW of the original image.

[Problems to be Solved by the Invention] By the way, when calculating the threshold value T for multi-value conversion as described above, various problems arise.

In order to calculate the threshold value T using the first example, it is necessary to create a density histogram as a premise. Therefore, it is necessary to separately prepare a circuit for the density histogram, which leads to an increase in circuit scale.

A Briscan is required to create a density histogram. Therefore, there is a drawback that the copying time becomes long when copying a -sheet.

Furthermore, in the case of this first example, when the resolution at the time of imaging is low, the density histogram does not have a bimodal characteristic as shown in the figure, and in this case, it is difficult to determine appropriate values for VA and VB. becomes difficult. If these values are not appropriate, the recorded image quality will deteriorate.

Even when an original image of a chromatic writing instrument is binarized, the same rWJ problem as mentioned above occurs because the peak characteristics of the density histogram are not remarkable.

On the other hand, in the second example, there is no need to create a density histogram, and therefore no Briscan is required.

However, when an original image of a chromatic writing instrument is binarized, the minimum and maximum values V^ and VB may not be clear. This causes the drawbacks described in the first example, such as difficulty in determining appropriate values of VA and VB.

Even in the case of a color original, the threshold values used for each color signal are the same, so determining the threshold value is especially problematic when recording a color image.

Therefore, the present invention solves these conventional problems by determining a threshold value for one color signal per line and sequentially executing this for each line to obtain a particularly appropriate color image. This paper proposes a color image processing device that is capable of determining a multivalue threshold for an effect.

[Means for Solving the Problems] In order to solve the above-mentioned problems, in the present invention, a signal for each color obtained by performing image processing on a signal obtained by photoelectric conversion from image information is set to a predetermined threshold value. When the number of color signals is n (nLt integer), in a color image processing device that forms an image signal for multilevel conversion by comparing This method is characterized in that equivalence values for multivalue quantization are determined.

[Operation] An appropriate threshold value is set according to the separated color signal. Therefore, the threshold value for multi-value conversion is different for each color.

The threshold value Ti (i refers to the separated color) is calculated by multiplying the density width of the original image by a constant coefficient ki, and then adding or subtracting a predetermined correction value αi to that value.

The coefficient m k f and the correction value αi are both used for fine adjustment for determining the threshold value. If the optimal coefficient ki and correction value αi are determined in advance so as to be able to deal with multiple colors and multiple originals, image signals can be multiplied with the optimal threshold even if the colors and types of originals are different. Value processing is possible.

As a result, it is possible to record clear color images without skipping or missing images.

Threshold determination for each color signal is performed for one color per line.

[Example] Next, a case where this example of the color image processing bag according to the present invention is applied to the above-mentioned image recording apparatus will be described in detail with reference to FIG. 1 and subsequent figures.

In FIG. 1, image information of the original FI 21 is guided to an optical system 22 and a spectroscopic system 31. The spectroscopic system 31 is composed of a pair of prisms 32 and 33 and a dichroic mirror 34.

In this example, the spectral characteristics of the dichroic mirror 34 are selected so that a red color-separated image is separated from one prism 32 side, and a cyan color-separated image is separated from the other prism 33 side.

The color separated images are supplied to the corresponding CCDs 35 and 36 for photoelectric conversion. Each color signal that has been photoelectrically converted is A/
Predetermined bits in the D converter 37 and 38, in this example 6
It is converted into a bit digital signal.

The color signals VR and VC which have been digitized and normalized by the white signal are used as a color discrimination device RO which functions as a color discrimination means.
The color signal is supplied to M39, and the color signal is separated for each pixel into data indicating color information, that is, color code data, and density data, which is density information of the color signal, and output as described later in this example. .

Color discrimination here refers to a process in which each pixel is assigned to one of a plurality of specific colors based on the image signal of each pixel obtained by reading a document. This specific color is mainly determined by the recording color (for example, black, red, blue) of the recording device.

Therefore, the address of the ROM storing the color code data and density data can be referenced by the color signals VR and VC.

A predetermined number of bits of color code data (for example, 2-bit data) and density data (for example, 6-bit data) are
By being supplied to the image processing circuit 40 and subjected to predetermined signal processing, a multivalued signal (signal of two or more values) is finally output.

This multivalued signal is supplied to the recording device 78 to record a color image or stored in the memory 79.

The color discrimination ROM 39 mentioned above is the 21st! The corresponding color segments so that separation data as shown in I is output.

Akebono data is stored. In this example, red, helmet, black, and white data (color code data and density data, respectively) can be referenced by a pair of color signals VR and VC.

Further, the relationship between color signals and color code data can be set as shown in FIG. Therefore, for example, the color signal V
When R and VC are signals indicating a black image, the color discrimination ROM 39 outputs not only density data corresponding to the density but also color code data corresponding to the black image.

As the above-mentioned recording device 78, an electrophotographic copying machine can be used.

FIG. 4 is a configuration diagram showing an example of the main part. This copying machine attempts to record a color image by separating the color information of a color original into about three types of color information. In this example, the color information to be separated is black BK, red R, and blue B.
The following three colors are illustrated.

In the figure, reference numeral 1 indicates a drum-shaped image forming member (photoreceptor drum), on the surface of which a photoconductive photoreceptor surface layer of selenium, OPC (organic semiconductor), etc. is formed, and a static image forming member corresponding to an optical image is formed. It is designed so that an electric image (electrostatic latent material) can be formed.

The surface of the image forming body 1 is negatively charged by the charger 2. A light signal (binarized signal based on each color separation body) 4 is irradiated onto the surface of the charged image forming body 1 for deep exposure.

After exposure, the film is developed using a predetermined developing device.

The developing devices are arranged in a number corresponding to the number of color separation bodies.

In this example, the developing device 5 is filled with red toner developer.
, a developing device 6 filled with a blue toner developer, and a developing device 7 filled with a black toner developer are connected to the image forming body 1 .
The image forming member 1 can be sequentially arranged facing the surface of the image forming member 1 in this order in the direction of rotation of the image forming member 1 .

The developing units 5 to 7 are sequentially selected in synchronization with the 10 rotations of the image forming body, and by selecting the developing unit 7, for example, a black color separation 61 (ordinary monochrome image) is developed.

A pre-transfer charger 9 and a pre-transfer exposure lamp 11 are provided on the developing device 7 side.
These make it easy to transfer the color image onto the recording medium P.

The pre-transfer charger 9 and the pre-transfer exposure lamp 11 are provided as necessary.

The color image developed on the image forming body 1 is transferred onto the recording body P by a transfer device 12.

The transferred recording material P is subjected to a fixing process by the fixing device 13 at the subsequent stage, and then the recording material P is discharged.

Note that the static eliminator N14 includes one or a combination of a static eliminator lamp and a corona discharger for static elimination.

The cleaning device 15 is composed of a cleaning blade and a fur brush, and is used to remove residual toner adhering to the drum surface after the color image of the image forming member 1 has been transferred.

FIG. 5 is a system diagram of essential parts showing an example of the image processing circuit 40. As shown in FIG.

In this invention, as mentioned above, in real time,
An optimal threshold value for a single color signal is calculated for each line, and density data is multi-valued based on this threshold value.

Therefore, the threshold value of one color can be calculated sequentially for one line.

First, the above-mentioned color code data and density data supplied to the terminals 41 and 42 are sent to the selector 43, where the color code data and density data corresponding to the line to be scanned are selected.

For example, if the first line determines the black threshold, the second line determines the red threshold, and the third line determines the blue threshold, then when the current scan line is the first line, Only color code data and density data related to black can be selected.

Therefore, the selector 43 can be supplied with the selection signal C1 output from the control circuit 44 as shown in FIG.

As shown in FIGS. 7A and 7B, the control circuit 44 is supplied with a line signal HV synchronized with the horizontal scanning line (main scanning line), and in synchronization with this, the corresponding color information is sequentially selected. It is designed so that

The color information selected by the selector 43 is supplied to the multivalue converting means 70 via the buffer memory 45. ``Such threshold value determination is particularly effective when recording is performed in one color with the apparatus shown in FIG. 1, regardless of the color of the document.

For example, this is the case when a color original is recorded using only one ring color. At this time, since the threshold values are sequentially determined for each line, the black, red, and blue signals obtained by reading the color original are sequentially multivalued using each threshold value, and are recorded completely in black.

In other words, recording is performed simultaneously by scanning the original once.

If a threshold value for each color is determined for each scan and printing is performed, three scans would be required, but according to this example, only one scan is required.

In order to perform multi-value conversion, a threshold value for multi-value conversion is required, but in this invention, a threshold value corresponding to the color can be set for each line using the means described below.

Therefore, among the selected color information, the density data is first supplied to the threshold value determination circuit 50, and the threshold value for the current scanning line is calculated.

In the threshold value determination circuit 50 described above, the threshold value for each color is determined as follows.

The threshold must be determined between the background level and the black level. The background level varies greatly depending on the color of the original. The information that we want to reproduce in the original image also shows large fluctuations in density.

On the other hand, although the background levels of originals vary widely, density unevenness at the background level does not vary greatly depending on the original. For example, when image information of a document is expressed in 6 bits (64 levels), the density unevenness of the background is only about 6 to 8 levels.

This can be roughly determined by checking the lowest level of the manuscript.

Incidentally, in the first example described as a conventional example, the background level is calculated by detecting the peak level of the density histogram.

The threshold value can be determined between this background level and the black level, that is, the black line level, but as mentioned above, if the threshold value is set at a level between the two, it will be difficult for high-contrast originals such as type printed on a white background. On the other hand, when printing on a colored original such as a newspaper, the characters are recorded in a slightly blurred manner.

For the above reasons, it is not preferable to use the middle level of the density range of the original as the threshold value. It is necessary to determine the threshold value based on other levels.

In this embodiment, the point determined from the density width DV (=DH-DL) of the density histogram as shown in FIG. 8 is the same, but the conditions for determining the threshold value are changed depending on the color. That is, the threshold value T1 is set as follows.

Ti=ki (DH-DL)+DL+αi here, ki: Which level of the original density range Corrected image to decide whether to select In numbers, 0≦ki≦1 ■＝Subscript indicating color αi: Correction value, 1α11≦1 A preferable value of the correction coefficient ki is as follows: 0.2≦≦0.8 It has been found through various experiments that this is the case.

The value of αi is usually around 1α11≦10 because the recording resolution varies depending on the system configuration of the image recording device used, especially the system configuration of the optical system and the laser drive system when performing imagewise exposure with a laser beam. You can select .

In order to perform the above-mentioned processing, the density data is first supplied to the maximum value calculation circuit 52 and the minimum value calculation circuit 53, and the maximum value DH and minimum value DL for that line are calculated, and the calculated maximum value DH and the minimum value DL are supplied to an arithmetic circuit 65, where the arithmetic processing of the above equation is performed.

FIG. 9 shows a specific example of the threshold value determination circuit 50.

Since the maximum value calculation circuit v852 and the minimum value calculation circuit 53 are the same in content, the configuration of the maximum value calculation circuit 52 will be described.

The density data of the current pixel and the density data of the previous pixel latched by the latch circuit 55 are supplied to the switching circuit 9 . Then, the density data of the current pixel and the density data of the previous pixel are supplied to a comparator 56 for comparing the levels, and the comparison output is used to compare the density data of the current pixel and fil.
The slope of each previous concentration data is selected. When the density data of one pixel before is larger, the density data of one pixel before is selected by the comparison output as shown in the figure.

Such a magnitude comparison operation is performed for all pixels having the same color code of the line, and the maximum value DH of the line is detected.

Similarly, in the minimum value calculation circuit 53, the comparator 5
The minimum value DL of that line is detected by the comparison output indicating the minimum value obtained in step 9. 57 is a switching means, and 58 is a latch circuit.

Maximum and minimum values DH obtained at the end of one line,
DL is a subtracter 60, and a calculation process of a density width DW of (DL-DH) is executed.

The density width DIJ is determined by the first ROMa that stores the correction coefficient ki.
After supplying the signal to l, the multiplication process with the correction coefficient ki is executed. The correction coefficient ki is selected in synchronization with the recording color of the color copying machine. Therefore, a BBR signal indicating which color is currently being recorded is supplied to the coefficient ROM 61.

As the BBR signal, a signal generated by a system control circuit (not shown) composed of a CPU is used.

The multiplied output ki (DH-DL) and the maximum value DH are added together by an adder 62.

On the other hand, the data of the coefficient ROM 63 in which the correction value αi has been stored and the addition output are supplied to the second addition M64, so that the final threshold value Ti is obtained at the output terminal 65a.

The correction value αi can also be selected by the BBR signal.

Such arithmetic processing is real-time processing.

In this case, it is necessary to consider a delay corresponding to the threshold value determination processing time, so to perform real-time processing, a line memory or the like may be used for processing time adjustment.

Note that the minimum value can be output to the other output terminal 65b, and the density width DIJ can be output to the output terminal 65c.

Here, when the calculated density width DV is below a certain value,
Since it is considered to be only background information, in this case, for convenience, it is desirable to use the minimum value as the threshold value at that time.

Therefore, a determination circuit 80 is provided after the threshold determination circuit 50.

This judgment circuit 80 includes a switching means 81 and a comparator 82.
and a latch circuit 83, the data of the density width DW is compared with the reference level VREF in the comparator 82,
When it is below the reference level VREF, the minimum value DL can be selected based on the comparison output.

On the other hand, when it is equal to or higher than the reference level VREF, the determination threshold Ti is selected as usual.

Further, in the process of calculating the maximum or minimum value, it is possible that noise or the like is mixed in and the threshold value fluctuates considerably, but as a countermeasure against such a situation, threshold value averaging processing is performed. Reference numeral 90 indicates interpolation means for this purpose.

In the embodiment, the average value of the previous and subsequent threshold values is used.

Therefore, the interpolation means 90 includes a selector 91, registers 92 to 94 for storing threshold values corresponding to a plurality of colors, and an averaging circuit 95.

Since the threshold value is determined for each line and also for each color, the threshold value for each line regarding the same color is output once every four lines, so registers 92 to 94 are set for each color. The register to be stored is selected by the selector 91.

For this reason, the selector 91 is supplied with a selection signal C2 as shown in FIGS. 6 and 7C from the control circuit 44.

Here, the reason why the contents of the selection signals C1 and 02 are different is because the threshold value of one line cannot be calculated until the scanning of that line is completed. This is because it is necessary to store the threshold data of the previous line in the register.

The averaging circuit 95 calculates the average value of threshold data regarding the same color obtained once every four lines.

Therefore, the output threshold data of the latch circuit 83 and the output threshold data of the registers 92 to 94 regarding the same color as this output threshold data are averaged.

Such processing avoids sudden fluctuations in the calculated threshold value.

The threshold value data that has been averaged is supplied to the multivalue converting means 7°, but in this case as well, selectors 71 and 75 at the front and rear stages and registers 72 to 74 corresponding to each color are provided, so that the average value processing can be completed. The threshold data is stored in the corresponding register.

Therefore, the selector 71 is supplied with the selection signal C2.

Therefore, as shown in FIG. 7, the maximum and minimum values of the colors corresponding to the lines are calculated in the sections I, Il [and V, respectively, and the threshold values are determined in the sections tl and ■, and these are stored in the corresponding registers. 72 to 74, and the threshold data is updated.

The color code data outputted from the buffer memory 45 can be supplied to the selector 75 at the subsequent stage as its selection signal, and the same threshold value data as the color code data selected by this color code data is sent to the multi-value circuit, e.g. The signal is supplied to a binarization circuit 76.

In this binarization circuit 76, the density data can be binarized based on the output threshold data. Therefore, the binary signal obtained at the output terminal 76a is fully supplied to the recording device 78 shown in FIG. 1, and a color image is recorded.

[Effects of the Invention] As explained above, according to the present invention, the threshold data is determined from the maximum value and minimum value in the line, so real-time processing becomes possible.

Furthermore, since the E value data is determined for each line and for each color, multi-value processing can be performed using a threshold close to the optimal threshold. Therefore, there is a practical benefit in that deterioration in recorded image quality due to collapse of the outline of a color image can be easily eliminated.

Therefore, the color image processing device according to the present invention is extremely suitable for application to the color image recording device as described above.

[Brief explanation of the drawing]

FIG. 1 is a system diagram showing an example of the configuration of essential parts of a color image processing device according to the present invention, FIG. 2 is a diagram showing an example of a color discrimination map, FIG. 3 is a diagram showing an example of color code data,
FIG. 4 is a block diagram showing an example of a color copying machine suitable for use in the present invention, FIG. 5 is a system diagram of main parts showing an example of an image processing circuit, and FIG. 6 is a diagram showing a color code and a selection signal. 7 is a diagram showing the timing relationship of the selection operation, FIG. 8 is a characteristic diagram of the density histogram, and FIG. 1
Figure 0 is a system diagram of a conventional image processing circuit, Figures 11 and 1
2[i4 are characteristic diagrams of density histograms. 39... Color discrimination ROM 40... Image processing circuit 50... Threshold value determination circuit 52... Maximum value calculation circuit 53... Minimum value calculation circuit 65... Arithmetic circuit 70... Multi-value conversion Means 80... Judgment circuit 82... Comparison circuit 90... Interpolation means 95... Averaging circuit Patent applicant Konica Corporation Figure 8 Figure 10 Figure 8. I, I, R/(J Figure 17 Figure 12

Claims (2)

[Claims] (1) A color image processing device that performs image processing on signals obtained by photoelectric conversion from image information and compares the signals for each color with a predetermined threshold value to form an image signal for multi-value conversion. is characterized in that, when the number of color signals is n (n is an integer), the threshold value for multi-value conversion for the corresponding color is determined once for every n horizontal lines. color image processing device. (2) The color image processing apparatus according to claim 1, wherein the color signals are selected from red, blue, black, and white signals.