JP2013131861A - Image reader, image forming apparatus, read image data processing method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To suppress degradation in picture quality such as color shift caused by switching, even if switching of white/black of a read image occurs at any timing relative to illumination period of a light source when color reading is performed by a light source sequential lighting reading method.SOLUTION: Based on pixel values D1, D2, and D3 of respective color components R, G, and B which are color pixel values of one line, it is determined whether a pixel to be processed is monochrome pixel or color pixel by a pixel of the line before and after the pixel to be processed which is taken as an interested pixel (S201-206). Under a condition in which the interested pixel is a color pixel and before and after pixels are determined to be monochrome pixels, the interested pixel is determined to be an abnormal monochrome pixel, and a pixel value of that pixel is corrected to be a pixel value as a correct monochrome pixel. A correction process detects n-th of lighting period of respective R, G, and B when white/black of an original switches, and according to a detection result, defines a color to be corrected and a correction value (substitution with a pixel value of the same color of before or after line)(S207-212).

Description

  The present invention relates to an image reading apparatus that performs color reading of a document image by a scanning method, and more specifically, performs color reading of a document image that is sub-scanned by an image sensor that sequentially turns on light sources that emit light of different colors and shares the colors. Image reading apparatus having a function of correcting a color attached to image data obtained by reading a monochrome image, an image forming apparatus provided with the image reading apparatus, a read image data processing method, and a program used for processing of the read image data processing method About.

Conventionally, a method that sequentially controls lighting of a plurality of color light sources corresponding to color components and reads a color image from a document that is sub-scanned using a line image sensor shared by each color (hereinafter referred to as “light source sequential lighting reading method”). There is a color image scanner that employs " In such a color image scanner, red, green and blue LEDs (light emitting diodes) are used as light sources.
In a color image scanner that employs a light source sequential lighting reading method, the line image sensor reads the original while always moving relative to the original at a constant speed in the sub-scanning direction, and light sources of red, green, blue, and each color. Since the lighting timings are sequentially shifted, the reading positions in the sub-scanning direction do not overlap.

Image reading by such a light source sequential lighting reading method is advantageous in terms of time, but since the reading positions of red, green, and blue colors are different, poor image quality such as color misregistration and color blurring, particularly at the boundary of the image. May occur.
Specifically, during the reading operation of pixels for one line in the main scanning direction, that is, the reading operation of three main scanning lines performed by turning on the red, green, and blue light sources of each color component, This is a case where a monochrome image portion that changes from white to black is read. For example, if a white image portion is read with a red light source, and then a black image portion is moved and read with a green light source and a blue light source, the pixel consisting of the red, green and blue color components of this line The color becomes red. Further, if the white image portion is read with a red light source and a green light source, and then the black image portion is moved and read thereafter with a blue light source, the information on this line becomes yellow.

Patent Document 1 (Japanese Patent Laid-Open No. 2000-13564) can be cited as a conventional example that has proposed means for solving the above-described problems such as color misregistration.
Japanese Patent Application Laid-Open No. 2004-228561 describes that the deviation of the irradiation position on the subject can be reduced and the color deviation can be reduced by emitting light so that the light emission times of light sources of different colors approach in time series. .
However, in the method of controlling the light emission timing between the light sources of the respective colors as in Patent Document 1, there is a case where image quality defects such as color misregistration cannot be suppressed depending on the relationship with the timing at which the monochrome image changes from white to black. In such a case, a sufficient effect cannot be obtained. A conventional reading operation for reading a color image by the light source sequential lighting reading method in which such a problem occurs will be described in more detail below.

FIG. 6 is a diagram for explaining a conventional reading operation in which color misregistration occurs in reading a color image by the light source sequential lighting reading method.
FIG. 6 shows the “original image” to be read in order from the top, the “main scanning synchronization signal” that is the operation signal of the image sensor, and the “Red” light source lighting control signal that is the red (Red) light source lighting control signal. “Lighting signal”, “Green lighting signal” which is a lighting control signal of a green (Green: G) light source, “Blue lighting signal” which is a lighting control signal of a blue (Blue: B) light source, R, G, B “Reading value of each color” which is a reading value when the light source of each color is turned on, and “Reading value of pixels in which RGB are aligned” which are reading values of pixels for one line in the main scanning direction are shown.

Note that the “original image” in FIG. 6 is placed on the time axis, so if the relative speed of the image sensor and the original changes, the white and black periods expand and contract on the time axis, but the original image is an image. Since the sensor moves relative to the sensor at a constant speed during the reading operation, the “document image” in the figure is an image (pixel array) in a moving direction of a certain point on the main scanning line as it is at a constant magnification. The
The “original image” shown in FIG. 6 is a monochrome image, and changes from white to black during the period T1 indicated by the arrow as the period for detecting the color pixel value of one line in the figure, and then the next period. During T2, the case where black continues is shown.
Further, the order of lighting the light sources of the respective colors in FIG. 6 is the order of R, G, and B, and the respective light sources are turned on for a predetermined period determined according to the difference of the light sources of the respective colors from the main scanning synchronization signal. Therefore, in order to read the RGB pixel values normally used as one line of color pixel values, the light sources of the respective colors are sequentially turned on for a predetermined period, and three line scans (main scan) by the image sensor are performed each time. Thus, the pixel value of each color is read.

In the light source sequential lighting reading method, as shown in FIG. 6, since the originals are read by sequentially turning on the LEDs that are the light sources of the respective colors in the order of R, G, and B in a state where the originals are conveyed. , G, and B are read with a shift of one line. Therefore, as shown in the period T1 ′ for reading color pixel values in which one line of RGB is aligned in FIG. 6, an image including white is displayed only in the lighting period of the R light source that is initially turned on at the position where white is switched to black. In the case of reading, the G light source and the B light source, the black image is read and “R = 150, G = 20, B = 20” is read as the pixel value.
That is, since it is black in the period T1, it should be “R = 20, G = 20, B = 20” (see the black pixel value read in the next period T2), but the R component is strong. The color is close to red instead of black.
In this example, the order of R → G → B is described, but even if the lighting order is changed, the boundary between white and black affects the reading result when the light sources of the respective colors are turned on, and “R = If the balance of 20, G = 20, B = 20 "is lost, a color is added at the black-and-white boundary.

As described above, when a portion where white is switched to black is read, the color is read and the image quality is deteriorated. Therefore, a technique for suppressing the following image quality degradation has been proposed.
In this method, by changing the order of the lighting period of the light source and the timing of the lighting period in the order of R → G → B, the change in the position where the color is switched from white to black has an effect on the reading value of each color. The lighting order is determined according to the lengths of the different lighting periods of the light sources of R, G, B colors, and the timing of each lighting period is set so that the lighting period of each light source falls within a short period. In particular, the possibility that the switching from white to black is affected by the reading value of each color is reduced.
Specifically, as will be described below with reference to the drawings, the light source with the longest lighting period is turned on second, and the negation timing of the light source lighting period with the first lighting color is set to the next line. It is set to be performed immediately before the main scanning synchronization signal is asserted.

FIG. 7 is a diagram for explaining a conventional reading operation to which a technique for suppressing color misregistration is applied in reading a color image by the light source sequential lighting reading method.
FIG. 7 is a diagram for explaining the operation when a method of changing the order of the lighting periods of the light sources of R, G, and B and the timing of the lighting periods is applied. It is drawing similar to FIG. Therefore, the description of the drawing common to FIG. 6 is omitted.
The optimum values of the lighting periods of the RGB light sources are determined by adjustment, and the color with the longest lighting order is turned on secondly according to the determined length of each lighting period. As described above, the lighting of the R light source having the longest lighting period is performed second in the lighting performed based on the main scanning synchronization signal output three times in the period T1 in which the color pixel value of one line is detected.

The G light source is turned on at the beginning of the main scanning synchronization signal output three times. At this time, the lighting period of the G light source is set to be negated immediately before the next main scanning synchronization signal is asserted.
By taking the order of the lighting periods of the light sources of each RGB color and the timing of the lighting periods, the period T1 ′ for reading the color pixel values in which one line of RGB is aligned is shortened by the period Td in FIG. (Refer to the period T1 ′ shown in FIG. 6).
In the example of FIG. 7 adopting the above configuration, assuming that the switching from white to black occurs at the same timing from the start of the period T1 for detecting the color pixel value of one line, as shown in FIG. By shortening the period T1 ′ for reading the color pixel values in which RGB are aligned, the portion where white is switched to black does not occur within the lighting period of the G light source that is initially turned on, and an image including white can be read. Since the black image is read during the lighting period of all the G, R, and B light sources, “R = 20, G = 20, B = 20” becomes the read pixel value, and color is added at the black and white border. Can read without any problem.

  As described above, by adopting the configuration of FIG. 7, it becomes difficult to switch from white to black within the period T1 ′, that is, to the read values of the respective colors that occur when switching is performed within the period T1 ′. The possibility of being affected is reduced, and color misregistration can be suppressed.

However, the solution shown in FIG. 7 can suppress the color misregistration because the switching from white to black is difficult to occur within the period T1 ′ in which the color pixel values in which one line of RGB is aligned are read. However, if switching from white to black occurs within the period T1 ′, the problem of color misregistration still remains.
In addition, Patent Document 1 cited as a conventional example does not describe a means for solving the above problem, and is a problem that the prior art is left unsolved.

  The object of the present invention is to perform the color reading by the light source sequential lighting reading method, regardless of the timing when the reading target image is switched from white to black or from black to white during the light source lighting period. In other words, it is possible to suppress deterioration in image quality such as color shift caused by black switching.

  The present invention is an image detection means having a plurality of types of light sources and image sensors for irradiating light of different colors to a read target image, the read target image and the image detection means relatively move, and a predetermined When the lighting operation of the light source, which is sequentially controlled in accordance with a predetermined order at a timing and a predetermined lighting time for each light source, is repeatedly performed, the light that has acted on the reading target image irradiated at each lighting time is Image detection means for detecting an image signal of each color component corresponding to each of the light sources detected by an image sensor, and a predetermined attention based on the pixel value of each color component obtained from the output signal of the image detection means A monochrome / color pixel determination unit that determines whether a pixel is a color pixel or a monochrome pixel, and the monochrome / color pixel determination unit; The predetermined pixel of interest is colored on the condition that the predetermined pixel of interest in the predetermined target pixel column is a color pixel, and the pixels read before and after the predetermined pixel of interest are determined to be monochrome pixels. An image reading apparatus comprising: an abnormal monochrome pixel determination unit that determines that the pixel is an abnormal monochrome pixel; and a pixel value correction unit that corrects the pixel value of the abnormal monochrome pixel determined by the abnormal monochrome pixel determination unit to a monochrome pixel value. is there.

  According to the present invention, when color reading is performed by the light source sequential lighting reading method, the white image is read regardless of the timing at which the reading target image is switched from white to black or from black to white during the lighting period of the light source. In addition, it is possible to suppress deterioration in image quality such as color shift caused by black switching.

1 is a diagram illustrating an outline of an overall configuration of an image reading apparatus according to an embodiment of the present invention. It is a figure which shows the processing flow of LED lighting period adjustment and AFE gain adjustment at the time of the power supply ON in the image reading apparatus which concerns on embodiment of this invention. It is a figure explaining an example of the reading operation | movement which applied the method which suppresses a color shift in the reading of the color image in the image reading apparatus (FIG. 1) of a light source sequential lighting reading system. FIG. 10 is a diagram illustrating another example of a reading operation to which a technique for suppressing color misregistration is applied in reading a color image in a light source sequential lighting reading type image reading apparatus (FIG. 1). It is a figure which shows the flow of the determination process of the abnormal monochrome pixel with a color, and the correction process of the determined said pixel. It is a figure explaining the reading operation of the prior art which a color shift produces in the reading of the color image by the light source sequential lighting reading system. It is a figure explaining the reading operation | movement of the prior art which applied the technique which suppresses a color shift in the reading of the color image by the light source sequential lighting reading system.

Embodiments of the present invention will be described with reference to the accompanying drawings.
In the following, the image reading apparatus of the present invention uses three light sources corresponding to tristimulus values for red, green, and blue (R, G, B) illumination light sources when reading a document image to be read by the light source sequential lighting reading method. A description will be given of an embodiment in which a light source is employed and a line image sensor is employed as means for detecting a pixel value of a reading target (original image) illuminated by the three light sources.
Note that the three light sources of red, green, and blue (R, G, and B) are not always used at fixed emission wavelengths, but in some cases, those that can change the reading characteristics by changing the emission wavelengths are used. The three light sources may be light sources that employ a method of outputting light of a desired wavelength by filtering a white light source.
In the image reading apparatus according to the present embodiment, the three light sources are LEDs, and the line image sensor employs a one-line CIS (Contact Image Sensor).

[Configuration of Image Reading Apparatus]
FIG. 1 is a diagram showing an outline of the overall configuration of an image reading apparatus according to an embodiment of the present invention.
The image reading apparatus shown in FIG. 1 is an apparatus that performs sheet-through reading. An image of a document D sent by a document transport roller 4 driven by a motor 5 that is operated by a motor drive circuit 51 is imaged at a fixed reading position A. It is detected by the detection unit 20.
The image detection unit 20 detects an image from a light source 21 for each of R, G, and B colors that are sequentially turned on in synchronization with a main scanning synchronization signal by a light source driving circuit 24 and an original that is illuminated by light emitted from the light source 21. It consists of one line CIS 22 that converts to a pixel signal.

The image reading apparatus also includes a controller 10 that controls the entire apparatus.
The controller 10 puts the AFE (analog front end) 30, the digital processing unit 40, the light source driving circuit 24, the motor driving circuit 51, and the operation unit 50 under control.
The AFE 30 receives an input of an image (luminance) signal detected as an electrical signal by the 1-line CIS 22, converts it into corresponding digital data, and transfers it to the digital processing unit 40. The image signal detected by the one-line CIS 22 is input to the AFE 30 in line-sequential synchronization with the main scanning synchronization signal because the light sources 21 of R, G, and B colors are sequentially turned on.

  The digital processing unit 40 includes an image data generation unit 41 that generates color image data by aligning image data of R, G, and B color components input line-sequentially as one line of data. In addition, as will be described in detail later, the digital processing unit 40 detects abnormal monochrome pixels (pixels that should originally be monochrome, pixels with image quality defects such as coloring) and correct them to the original monochrome pixels. As elements necessary for processing, each unit includes a color information detection unit 42, a monochrome / color determination unit 43, an abnormal monochrome pixel determination / correction unit 44, and a memory 45.

The light source drive circuit 24 drives each LED of R, G, and B at a predetermined timing in a period determined by the main scanning synchronization signal according to a predetermined order, and lights each LED. The controller 10 has light source control means for outputting a lighting control signal to the light source drive circuit 24, and the lighting operation is controlled by the light source control means.
The motor drive circuit 51 operates the motor 5 that drives the document transport roller 4 according to a control signal from the controller 10 that transports the document at a constant speed corresponding to the designated resolution.
The operation unit 50 includes a display unit and an operation unit such as an input key, and has an operation panel that provides a function as a user interface. The display unit of the operation panel uses display means for notifying the user of the state of the device and guiding the user's input operation, and the operation unit is a touch panel system that cooperates with the screen of the display unit. An input means capable of key operation is used.

Further, the image reading apparatus includes a CPU (Central Processing Unit) 11 as shown in FIG. 1, and the controller 10 is configured with a computer having the CPU 11 as an element as hardware. Thus, when the controller 10 is configured by a computer, the computer can use a program for controlling the AFE 30 to be controlled, the digital processing unit 40, the light source driving circuit 24, the motor driving circuit 51, and the operation unit 50. A computer can be made to function as a controller by preparing it by recording on various recording media and driving this program by the CPU 11.
In the configuration illustrated in FIG. 1, an example in which the AFE 30 and the digital processing unit 40 are configured by a circuit different from the controller 10 is employed. However, a program for realizing the processing functions of the AFE 30 and the digital processing unit 40 is stored in the controller. The controller 10 may function as the AFE 30 and the digital processing unit 40 by being driven by ten computers.

<Adjustment of CIS detection system>
Here, an adjustment operation performed when the power is turned on in the image detection unit 20 and the AFE 30 of the image reading apparatus in FIG. 1 will be described.
When the power is turned on, the controller 10 adjusts the light source (LED) lighting period of the image detection unit 20 and the gain of the AFE 30.
FIG. 2 is a diagram showing a processing flow of LED lighting period adjustment and AFE gain adjustment when the power is turned on in the image reading apparatus according to the embodiment of the present invention.
According to the processing flow of FIG. 2, first, the controller 10 starts adjustment of the LED lighting period (light quantity) (step S101), and provisionally sets a predetermined LED lighting period and a minimum value of the AFE gain as initial settings ( Step S102).

Next, the controller 10 detects the image data of the white reference plate provided at the reading position A in the state where there is no document D under the conditions that the image detection unit 20 is initialized, and based on the output of the AFE 30 at that time. The peak value of the read image data is acquired (step S103).
Further, the peak value acquired in step S103 is compared with a predetermined target value that has been set (step S104). If the peak value is lower than the target value (step S104-YES), an LED is turned on to set the peak value to the target value. The period is calculated, the calculated LED lighting period is set (step S105), and the peak value is acquired again (step S103).
If the peak value is greater than or equal to the target value in step S104 (step S104-NO), the process proceeds to AFE gain adjustment. The LED lighting period adjustment so far is performed for each of the R, G, and B LEDs. The lighting period after adjustment is not the same for R, G, and B due to the spectral sensitivity of one line CIS22 and the spectral characteristics of each R, G, and B LED.

  The process proceeds to AFE gain adjustment, and the process is started (step S106). First, the controller 10 sets the white reference plate provided at the reading position A where the original D is not present with the previously set LED lighting period. Is detected by the image detection unit 20, and the output of the AFE 30 at that time is acquired as a white reading value (step S107). When this white reading is performed, the lighting period of the LED is asserted in synchronization with the main scanning synchronization signal, and the white reading value when the LED is driven with the lighting periods adjusted to R, G, B respectively is a predetermined target. AFE gain adjustment is performed so as to obtain a value.

As an adjustment procedure, the white reading value acquired in step S107 is within the range of the set target value, or is compared with an upper limit value (target value +) and a lower limit value (target value-) (step S108). If the value does not fall within the range of the predetermined target value (step S108-NO), the AFE gain for setting the white reading value within the target range is calculated, and the calculated AFE gain is set ( In step S109, the white reading value is acquired again (step S107).
If the white reading value falls within the range of the predetermined target value in step S108 (step S108-YES), the AFE gain adjustment is completed, and the adjustment process of this flow is ended.
The LED lighting period and the AFE gain that are applied to the LEDs of the respective colors obtained in this adjustment processing flow are used for proper reading. In this image reading apparatus, the obtained LED lighting period is the reading target. This is information necessary for a reading operation to be described later that makes it possible to suppress deterioration in image quality such as color shift caused by switching between white and black in an image.

<Suppression of color shift>
Since the image reading apparatus reads the original image to be read by the light source sequential lighting reading method, the order of the light source lighting periods in the order of R → G → B described in the above “Background Art” with reference to FIG. According to the above, the probability of occurrence of color misregistration increases.
Therefore, the light source with the longest lighting period is turned on second, and the negation timing of the light source lighting period with the first lighting color is set to be performed immediately before the main scanning synchronization signal of the next line is asserted. Therefore, the period for reading the color pixel values in which one line of RGB is aligned is shortened to make it difficult for color misregistration to occur.

In the case of LEDs of R, G, and B colors, R usually has the longest lighting period and becomes shorter in the order of B and G. Therefore, the order is G → R → B, and immediately before the assertion of R is the G lighting period. Set. In addition, since it is possible to arbitrarily determine whether the shortest color in the lighting period is the first or last color, the length of the period for reading the color pixel values in which one line of RGB is aligned even in the order of B → R → G That does not change.
When the light source includes light other than visible light such as infrared light, the order of visible light is set before the lighting order other than visible light.

FIG. 3 shows an example of a reading operation in which a color image is read by a light source sequential lighting reading method, and a method of suppressing color misregistration by a lighting order that minimizes a period for reading color pixel values in which one line of RGB is aligned. It is a figure explaining.
FIG. 3 shows the “original image” to be read in order from the top, the “main scanning synchronization signal” which is the operation signal of the one-line CIS 22, the R light source lighting control signal “Red lighting signal”, and the G light source. Lighting control signal “Green lighting signal”, B light source lighting control signal “Blue lighting signal”, R, G, B light source lighting “each color reading value” and the main scanning direction “Read values of pixels in which RGB are aligned”, which is a read value of pixels for one line, is shown.

Since the “original image” in FIG. 3 is placed on the time axis, if the relative speed of the one line CIS 22 and the original D changes, the white and black periods expand and contract on the time axis. Is moved relative to one line CIS 22 at a constant speed during the reading operation, so that the “original image” in the figure is an image (pixel row) in a moving direction of a certain point on the main scanning line as it is at a constant magnification. It is represented by
The “original image” shown in FIG. 3 is a monochrome image, and changes from white to black during the period T1 indicated by the arrow as the period for detecting the color pixel value of one line in the figure, and then the next period. During T2, the case where black continues is shown.

Further, the order of lighting the light sources of the respective colors in FIG. 3 is such that R, which is the longest period in the adjustment of the LED lighting period in FIG. The negation timing of G is set immediately before the assertion timing of R, and B is asserted in the order of G, R, and B by asserting with a normal main scanning synchronization signal.
Therefore, in order to read the RGB pixel values that are normally used as one line of color pixel values, here, the G, R, and B light sources are sequentially turned on, and the image sensor performs three times each time. The pixel value of each color is read by scanning (main scanning).

By taking the order of the lighting periods of the light sources of R, G, and B colors and the timing of the lighting periods, the period T1 ′ for reading color pixel values in which one line of RGB is aligned is minimized, that is, in FIG. It can be shorter than the indicated period T1 ′.
Therefore, the period excluding the period T1 ′ in the period T1 in which the color pixel value of one line is detected is longer than before, and even if the “original image” changes from white to black during this period, the color becomes black with color. An advantage that the occurrence of the deviation can be avoided is generated (see FIG. 7).
Here, the order of the three RGB colors is determined. However, if there are more light source colors, the light is lit in the order from the longest lighting period to the center and before the longest lighting period. The color to be lit is such that the negation timing of the lighting period is immediately before the main scanning synchronization signal of the next line, and the interval with the long lighting period is made as short as possible to minimize the period T1 ′.

However, if only the order of the lighting periods of the light sources of R, G, and B and the timing of the lighting periods are as shown in FIG. 3, the white color is detected within the period T1 ′ for reading the color pixel values in which one line of RGB is aligned. If the switch to black occurs, a color shift occurs.
Since the “original image” in FIG. 3 is switched from white to black at the timing of the lighting period of the R light source, it occurs when it occurs within the period T1 ′. In this case, a white image is read during the entire lighting period of the G light source that is turned on first, a black image including white is read during the lighting period of the R light source, and a black image is read during the entire lighting period of the B light source. Therefore, “G = 240, R = 150, B = 20” becomes the read pixel value, and the read image becomes an image with a color (dark yellow) at the black-and-white boundary.

FIG. 4 is a diagram for explaining another example of the reading operation to which the same technique for suppressing color misregistration as in FIG. 3 is applied in reading a color image by the light source sequential lighting reading method.
FIG. 4 differs from FIG. 3 in that the “original image” is switched from white to black at the timing of the lighting period of the B light source after R. In the case of FIG. 4, a white image is read during the entire lighting period of the first G light source and the next R light source, and a black image including white is read during the lighting period of the G light source, so that “G = 240, R = 240, B = 150 "becomes the read pixel value, and the read image becomes an image with a color (light yellow) at the boundary between black and white.

<Correction to monochrome pixel>
As shown in FIGS. 3 and 4, when the switching from white to black in the “original image” occurs within a period T1 ′ in which color pixel values in which one line of RGB is aligned is read, the read image is a black-and-white boundary portion. As a result, the image quality is deteriorated.
In view of this, the present image reading apparatus has an image quality such as color misregistration caused by the switching between white and black regardless of the timing at which the reading target image is switched from white to black or from black to white during the lighting period of the light source. The reduction can be suppressed.
In the present embodiment, a method of determining the order of the lighting periods of the light sources of R, G, and B colors and the timing of the lighting periods in which the period T1 ′ for reading color pixel values in which one line of RGB is aligned is the shortest period (FIG. 3). ), And a correction process for changing a color pixel (hereinafter, also referred to as “abnormal monochrome pixel”) that is generated when the “original image” is switched from white to black within the period T1 ′ to a desired monochrome pixel. In order to prevent color misregistration, the output of an image obtained by reading a monochrome image is not colored.

In the correction process, out of the pixels detected every time period T1 for detecting the color pixel value of one line, an abnormal monochrome pixel that has become a color pixel instead of the original monochrome pixel is determined as a pixel to be corrected. In this process, the determined abnormal monochrome pixel is corrected to a desired monochrome pixel.
Specifically, based on the pixel values of the R, G, and B color components that are the color pixel values of one line, it is first determined whether the pixel to be processed is a monochrome / color pixel.
Next, using the pixel to be processed as a target pixel, the target pixel is a color pixel, and monochrome / color pixel determination results of pixels read before and after the target pixel are obtained, and these are determined to be monochrome pixels. Under these conditions, the target pixel is determined to be an abnormal monochrome pixel.
The correction of a pixel determined to be an abnormal monochrome pixel to a desired monochrome pixel differs depending on the timing at which the monochrome boundary portion of the document is in the detection (lighting) period of each color. For this reason, among the color components detected in accordance with the detection order, the detection order of the color component that is the pixel value of the abnormal monochrome pixel is detected, and the following 1. ~ 3. Apply the correction shown in.

1. When the black-and-white boundary portion of the document is the detection timing of the first color component (the G component pixel value in FIG. 3 is the color pixel value), the pixel value of the color (G) component is read one after. Correction to replace the pixel value of the same color component of the monochrome pixel to be performed is performed.
2. When the black-and-white boundary portion of the document is the detection timing of the color components in the middle order (the pixel value of the R component that is the second detection order in FIG. 3 is the color pixel value), the first color (G) component and Replace the pixel value of the intermediate color (R) component with the pixel value of the same color component of the monochrome pixel read one after, or change the pixel value of the intermediate color (R) component and the last color (B) component Correction that replaces the pixel value of the same color component of the monochrome pixel read immediately before is performed.
3. When the black-and-white boundary portion of the document is the detection timing of the last color component (the B component pixel value in FIG. 3 is the color pixel value), the pixel value of the color (B) component is set to the previous one. Correction that replaces the pixel value of the same color component of the read monochrome pixel is performed.
By performing the above correction, the color pixel component of the abnormal monochrome pixel is eliminated, so that the monochrome image is prevented from being colored, and the monochrome pixel that should be in accordance with the detection timing of each color component is set to either monochrome. By determining the above, it is possible to prevent the image quality from being deteriorated.

<Procedure for abnormal monochrome pixel determination and correction processing>
Next, a procedure of correction processing for abnormal monochrome pixels performed based on the method described in <Correction to monochrome pixels> will be described.
FIG. 5 is a diagram illustrating a flow of a determination process for a colored abnormal monochrome pixel and a correction process for the determined pixel.
The processing flow in FIG. 5 is executed by the digital processing unit 40 that receives the read image data that is detected by the image detection unit 20 by the light source sequential lighting reading method at the time of document reading and that is output via the AFE 30 in accordance with an instruction from the controller 10.
In the monochrome / color determination, abnormal monochrome pixel determination, and monochrome pixel correction processing described below, the color information detection unit 42 refers as necessary color information from the memory 45 that stores the read data of each line. It functions as means for detecting the pixel values of the R, G, and B color components and providing them to each processing unit.

According to the processing flow of FIG. 5, the digital processing unit 40 receives the input of the image data output from the AFE 30, and the image data generation unit 41 aligns the pixel values of the R, G, and B color components to provide one line of color pixel values. Based on the generated color pixel value, first, it is determined for each line whether the pixel to be read is a color pixel or a monochrome pixel. This determination is performed by the monochrome / color determination unit 43.
In the monochrome / color pixel determination, if the pixel is a color pixel, the difference between the pixel values of the R, G, and B color components is larger than a predetermined value. If it is a non-color pixel in this determination, it is determined as a monochrome pixel.

Specifically, the pixel value of the color component that is first turned on in the order of lighting of the LEDs of the R, G, and B colors (that is, the detection order) is D1, the color component that is turned on second in the middle is D2, and last. Let D3 be the pixel value of the color component that was lit up, and let (n) be the target line, and let D1 (n), D2 (n), and D3 (n) respectively. Further, the absolute value (| D1 (n) −D2 (n) |) of the difference between the pixel value D2 of the color component that is lighted second to the pixel value D2 of the color component that is lighted secondly is expressed as Da (n), Let Db (n) be the absolute value (| D1 (n) −D3 (n) |) of the difference between the pixel value D1 of the first lit color component and the pixel value D3 of the last lit color component.
The threshold value for determining monochrome / color pixels from Da (n) and Db (n) is Dth_c, and if either Da (n) or Db (n) is greater than Dth_c, it is determined as a color pixel. If both are small, it is determined as a monochrome pixel.

The monochrome / color pixel determination procedure in the processing flow of FIG. 5 is performed by determining the target line (n) and the preceding and subsequent lines, that is, the (n−1) line, (n) line, and (n + 1) line. Sequentially. By performing this procedure, an abnormal monochrome pixel is detected and a line that needs correction is detected.
Therefore, the monochrome / color determination unit 43 determines the absolute value of the difference between D1 and D2 and the absolute value of the difference between D1 and D3 in the (n−1) line immediately before the target line (n), that is,
| D1 (n-1) -D2 (n-1) | = Da (n-1)
| D1 (n-1) -D3 (n-1) | = Db (n-1)
(Step S201), and the threshold processing for each absolute value of the calculated difference,
Da (n-1) or Db (n-1)> Dth_c
Thus, the monochrome / color pixel is determined (step S202).

If Da (n-1) or Db (n-1)> Dth_c is satisfied in step S202 (YES in step S202), the (n-1) line is determined as a color pixel. Since the (n) line is not a correction target, the digital processing unit 40 ends this processing flow.
On the other hand, if Da (n-1) orDb (n-1)> Dth_c is not satisfied in step S202 (step S202-NO), the (n-1) line is a monochrome pixel and is therefore a target line ( Since there is a possibility that the n) line may be a correction target, the process proceeds to the monochrome / color pixel determination of the next (n) line.

(N) The monochrome / color pixel of the line is determined by the absolute difference between D1 and D2 with respect to the pixel values of the R, G, and B color components of the (n) line, as in the (n-1) line. The value Da (n) and the absolute value Db (n) of the difference between D1 and D3 are calculated (step S203), and monochrome / color pixel determination is performed (step S204) to obtain a result.
However, since the (n) line is the target line, if it is determined to be a monochrome pixel here (step S204-NO), it will not be subject to correction, so the digital processing unit 40 ends this processing flow.
On the other hand, if it is a color pixel in step S204 (step S204-YES), this (n) line, which is the target line, may be subject to correction, so the next (n + 1) line monochrome / color pixel. Move on to judgment.

The determination of the monochrome / color pixel of the (n + 1) line is the absolute difference between D1 and D2 with respect to the pixel values of the R, G, and B color components of the (n + 1) line, as in the (n-1) line. The value Da (n + 1) and the absolute value Db (n + 1) of the difference between D1 and D3 are calculated (step S205), and monochrome / color pixel determination is performed (step S206) to obtain a result.
In order for the (n) line, which is the target line determined to be a color pixel, to be subject to correction, the (n + 1) line must be determined to be a monochrome pixel, and here, the (n + 1) line is determined to be a color pixel. If this is done (step S206—YES), the (n) line, which is the target line, is not subject to correction, and the digital processing unit 40 ends this processing flow.

On the other hand, if it is determined that the pixel is a monochrome pixel in step S206 (step S206-NO), the pixel on the (n) line of interest that is determined to be a color pixel is determined to be an abnormal (colored) monochrome pixel, Since it becomes the object of correction, the process proceeds to the next correction process.
The abnormal monochrome pixel determination process performed in the steps S201 to S206 is performed by the abnormal monochrome pixel determination / correction unit 45 in cooperation with the monochrome / color determination unit 43.

  The correction of a pixel determined to be an abnormal (colored) monochrome pixel to a desired monochrome pixel differs depending on the timing at which the monochrome boundary portion of the document is in the detection (lighting) period of each color. For this reason, among the color components detected in accordance with the detection order, the detection order of the color component that is the pixel value of the abnormal monochrome pixel is detected, and correction according to the detected detection order is performed.

Therefore, at what number during the lighting period of each of the three colors the white / black color of the document is switched, that is, the detection period of G, R, or B at which the black and white boundary portion of the document is detected. Is detected.
Therefore, the absolute value of the difference between the pixel value of each color one line before the target line and the pixel value of each color of the target line, that is, (| D * (n−1) −D * (n) |) (where * 1, 2, and 3) are calculated as Dα (n), Dβ (n), and Dγ (n), respectively, and the white / black switching is performed by the change in these pixel values between the obtained line and the attention line. In which color detection period is detected. As processing, the obtained Dα (n), Dβ (n), and Dγ (n) are compared with the threshold value Dth_n, and when it is larger than Dth_n, the switching of white / black is known during the lighting period. In addition, by performing this threshold processing in the same order as the lighting order, it is possible to know where the switching has been performed and to specify the color.

As a procedure for detecting white / black switching of the document in the processing flow of FIG. 5, the digital processing unit 40 firstly compares the pixel values of the respective color components between the target line (n) and the previous line (n−1). , Dα (n), Dβ (n), Dγ (n) are calculated (step S207).
Next, the white / black switching of the document is performed by threshold processing in the order of Dα (n) → Dβ (n) with respect to the detection order of color components G → R → B, that is, the difference between the calculated pixel values of each color component. The detection procedure is performed, and if a change is detected, the monochrome pixel corresponding to each color component is corrected. In the following procedure, the threshold value processing for Dγ (n) in B in the third detection order is omitted according to the result of the threshold processing for Dβ (n) in R which is the second detection order. ing.

Therefore, threshold processing of “Dα (n)> Dth_n” of the first G color component is performed (step S208).
Here, when the pixel value difference Dα (n) of the G color component between the target line and the previous line exceeds the threshold value Dth_n, and white / black switching of the document is detected (YES in step S208). Since the pixels in the line are determined to be abnormal monochrome pixels, correction to the appropriate monochrome pixels is performed.
In this correction, the pixel value of this G color component is replaced with the pixel value of the same G color component of the monochrome pixel (determined to be monochrome in step S206) read one after, that is, “D1 ( n) ← D1 (n + 1) ”is corrected (step S209). In this correction process, appropriate correction can be performed considering that the subsequent R and B are monochrome pixels, and continuity is maintained by taking a value close to the pixel value of the next line.
After correcting this monochrome pixel, the procedure of this processing flow is finished.

On the other hand, when Dα (n) does not exceed the threshold value Dth_n and the white / black switching of the document is not detected (NO in step S208), the threshold value “Dβ (n)> Dth_n” of the R color component in the next detection order is detected. Processing is performed (step S210).
When the pixel value difference Dβ (n) of the R color component exceeds the threshold Dth_n and white / black switching of the document is detected (step S210-YES), the pixel of the target line is determined to be an abnormal monochrome pixel. Therefore, the correction to the desired monochrome pixel is performed.

In this correction, the pixel values of the R color component and the first G color component are replaced with the pixel values of the same R and G color components of the monochrome pixel read one after, that is, “D1 (n) ← D1 (n + 1) ), D2 (n) ← D2 (n + 1) ", or the pixel values of the R color component and the last B color component are set to the same R and B color component pixel values of the monochrome pixel read immediately before. Replacement is performed, that is, correction of “D2 (n) ← D2 (n−1), D3 (n) ← D3 (n−1)” is performed (step S211).
It is desirable that this correction process has selectivity and can be selected as appropriate depending on the characteristics of the device and user requirements. For example, a means for allowing the user or the administrator of the device to instruct selection from the operation unit 50 can be taken, and an appropriate processing result can be obtained in this way.
After correcting this monochrome pixel, the procedure of this processing flow is finished.

Further, when Dβ (n) does not exceed the threshold value Dth_n and the white / black switching of the document is not detected (step S210—NO), it is considered that the difference Dγ (n) of the B color component pixel values exceeds the threshold value Dth_n. It is determined that white / black switching of the document has been detected by detection of the B color component, and correction is performed on the pixel value of the B color component to an appropriate monochrome pixel.
The pixel value of the B color component is replaced with the pixel value of the same B color component of the previously read monochrome pixel (determined to be monochrome in step S202), that is, “D3 (n) ← D3 (n−1) ”is corrected (step S212). In this correction processing, appropriate correction can be performed considering that the immediately preceding R and G are monochrome pixels and continuity is maintained by taking a value close to the pixel value of the previous line.
After correcting this monochrome pixel, the procedure of this processing flow is finished.

  By performing the determination process of the abnormal monochrome pixel and the correction process of the determined pixel in FIG. 5, the color pixel component of the abnormal monochrome pixel is eliminated, thereby preventing the monochrome image from being colored, By determining which monochrome pixel should be monochrome according to the component detection timing, it is possible to prevent deterioration in image quality. Further, the correction to the monochrome pixel does not change the amount of the read data, and the same magnification can be ensured.

[Combination with other image processing functions]
The image reading apparatus can function as an input unit of image data to a destination image processing apparatus by connecting to an image processing apparatus using image information read from a document via a communication unit.
An image data input unit of an apparatus for forming an image by operating a printer engine by using image data for print output is a usage form of the image reading apparatus, and the use form allows the image forming apparatus to have a copying function. Can do.
The image forming apparatus configured as described above prevents the deterioration of the image quality of the copied image by inputting the original image by the image reading apparatus and performing the image forming process using the obtained print output image data. Thus, the performance can be improved.

  4 .. Document transport roller, 10 .. Controller, 11 .. CPU, 20 .. Image detector, 21 .. Light source of each color of R, G, B, 22 ... 1 line CIS, 24 .. Light source drive circuit 30..AFE, 40..Digital processing unit, D..Original.

JP 2000-13564 A

Claims (12)

An image detection unit having a plurality of types of light sources and image sensors for irradiating light of different colors to a read target image, wherein the read target image and the image detection unit move relative to each other, and at a predetermined timing and the light source When the lighting operation of the light source, which is sequentially controlled in accordance with a predetermined order for each predetermined lighting time, is repeatedly performed, the image sensor detects light applied to the reading target image irradiated at each lighting time. Image detecting means for outputting an image signal of each color component corresponding to each of the light sources;
A monochrome / color pixel determination unit that determines whether a predetermined pixel of interest is a color pixel or a monochrome pixel based on a pixel value of each color component obtained from an output signal of the image detection unit;
On condition that the predetermined pixel of interest in the predetermined target pixel column is a color pixel and the pixels read before and after the predetermined pixel of interest are determined to be monochrome pixels by the monochrome / color pixel determination means. Abnormal monochrome pixel determination means for determining that the predetermined pixel of interest is a colored abnormal monochrome pixel;
Pixel value correction means for correcting the pixel value of the abnormal monochrome pixel determined by the abnormal monochrome pixel determination means to a monochrome pixel value;
An image reading apparatus. The image reading apparatus according to claim 1,
The monochrome / color pixel determining unit is configured to turn on a pixel value of a color component detected when the first light source that is turned on first in the predetermined order is turned on and a second value that is turned on in an intermediate order in the predetermined order. The difference between the pixel value of the color component detected when the light source is turned on or the pixel value of the color component detected when the first light source is turned on and the third lighted last in the predetermined order The predetermined pixel of interest is a color pixel and any pixel other than the color pixel is a monochrome pixel on the condition that one of the differences from the pixel value of the color component detected when the light source is turned on is greater than the predetermined value Image reading device. The image reading apparatus according to claim 2,
A lighting control means for controlling lighting of the plurality of types of light sources;
The lighting control means, the longer the predetermined lighting time, the closer to the middle in the predetermined order,
The image reading apparatus, wherein the second light source is a light source that is in the middle in the predetermined order. The image reading apparatus according to any one of claims 1 to 3,
An abnormal pixel component detecting unit for detecting an abnormal color component among the color components of the pixels determined as the abnormal monochrome pixel by the abnormal monochrome pixel determining unit;
The image reading device, wherein the pixel value correcting unit corrects a color component detected by the abnormal pixel component detecting unit according to a detection order of the image detecting unit. The image reading apparatus according to claim 4,
The abnormal pixel component detection unit is a color in which a difference between pixel values of the same color component of the abnormal monochrome pixel and the monochrome pixel read immediately before the abnormal monochrome pixel detected by the image detection unit is larger than a predetermined value. An image reading apparatus that detects a component as an abnormal color component. The image reading apparatus according to claim 5, wherein
The pixel value correcting unit is a monochrome unit that reads the pixel value of the first color component in the first detection order when the color component in the first detection order is detected by the abnormal pixel component detection unit in the image detection unit. An image reading apparatus that corrects a pixel value by replacing it with a pixel value of the same color component. The image reading apparatus according to claim 5 or 6,
The pixel value correction unit reads the pixel values of the first and intermediate color components in the detection order when the color detection component in the detection order is detected by the abnormal pixel component detection unit in the image detection unit. By replacing the pixel values of the same color component of the monochrome pixel to be detected or by replacing the pixel values of the intermediate and last color components with the pixel value of the same color component of the previously read monochrome pixel. Image reading device. The image reading apparatus according to any one of claims 5 to 7,
The pixel value correcting unit reads the pixel value of the last color component in the detection order one time before when the abnormal pixel component detection unit detects the last color component in the detection order in the image detection unit. An image reading apparatus that corrects a pixel by replacing it with a pixel value of the same color component of the monochrome pixel. The image reading apparatus according to any one of claims 1 to 8,
The plurality of light sources are three light sources corresponding to tristimulus values.   A program for causing a computer to function as each unit of the monochrome / color pixel determination unit, the abnormal monochrome pixel determination unit, and the pixel value correction unit of the image reading apparatus according to claim 1. An image reading device according to any one of claims 1 to 9,
An image forming unit that forms an image using print output data generated based on input image data;
An image forming apparatus. A read image data processing method in an image reading apparatus provided with an image detection unit having a plurality of types of light sources and image sensors for irradiating light of different colors to images to be read,
While relatively moving the image to be read and the image detecting means, the lighting operation of the light source, which is sequentially controlled in accordance with a predetermined order at a predetermined timing and a predetermined lighting time for each light source, is repeatedly performed to irradiate each lighting time. An image detecting step of detecting light acting on the read target image by the image sensor and outputting an image signal of each color component corresponding to each of the light sources;
A monochrome / color pixel determination step for determining whether a predetermined pixel of interest is a color pixel or a monochrome pixel based on a pixel value of each color component obtained from the output signal detected in the image detection step;
On the condition that in the monochrome / color pixel determination step, the predetermined target pixel in the predetermined target pixel column is a color pixel, and pixels read before and after the predetermined target pixel are determined to be monochrome pixels. An abnormal monochrome pixel determination step for determining that the predetermined pixel of interest is a colored abnormal monochrome pixel;
A pixel value correction step of correcting the pixel value of the abnormal monochrome pixel determined in the abnormal monochrome pixel determination step to a monochrome pixel value;
A read image data processing method comprising:

Images