JP7353899B2 - Image reading device and method of controlling the image reading device - Google Patents

Description

The present invention relates to an image reading device and a method of controlling the image reading device.

Copying machines and multifunction printers are equipped with image reading devices for reading images from originals. Two known reading methods used by image reading apparatuses include a pressure plate reading method in which a document is placed on a document table and read while moving the reading unit, and a panning method in which the document is read while being transported by an automatic document transport mechanism.

The reading unit includes a light emitting section that illuminates the document, and a light receiving section that receives reflected light from the document. The image reading device generates image data representing an image of a document based on reflected light received by a light receiving section.

The light emitting section uses a light source such as an LED (Light Emitting Diode). The light receiving section includes a plurality of line sensors in order to read color images, and for example, R (red), G (green), and B (blue) color filters are applied on the light receiving surface of each line sensor. . The reading unit reads a document by having a light receiving section receive diffused light irradiated onto the document from a light emitting section. The image reading device generates image data from luminance data (hereinafter referred to as "read data") that is a result of light reception by a light receiving section of a reading unit.

Hereinafter, an image reading device including a reading unit using a light emitting section and a light receiving section will be described as an example. In the reading unit of this image reading device, a plurality of light receiving elements are lined up in the main scanning direction, with the main scanning direction being a direction perpendicular to the direction in which the document is conveyed or the direction in which the reading unit moves. In such an image reading device, when changing the reading resolution in the sub-scanning direction and increasing the transport speed of the automatic document transport mechanism, or increasing the transport speed of the reading unit to perform high-speed reading, it is necessary to change the reading resolution of R, G, and B. The amount of deviation is, for example, the amount of deviation of a non-integer pixel such as 0.5 pixel.

Patent Document 1 discloses a method of generating pseudo pixels at integer pixel positions by performing interpolation processing.

Furthermore, in Patent Document 2, by using multicolor light sources (for example, R light source, G light source, and B light source) in the light emitting part and controlling the lighting timing of each light source, the position of the image is determined by alignment processing for an integer number of pixels. A method for performing the alignment is disclosed.

Japanese Unexamined Patent Publication No. 1-109966 Japanese Patent Application Publication No. 2005-322990

However, in Patent Document 1, interpolation processing is performed for a line sensor that is shifted by a non-integer number of pixels with respect to the reference line sensor position, so pixels are generated in a pseudo manner, so the signal level of the light receiving element is determined by the light receiving area. The situation is the same as when the area is widened, resulting in blurring and deterioration of image quality. Further, in Patent Document 1, when high-speed reading is performed to generate a low-resolution image, aliasing noise occurs if the document contains high-frequency components. For example, if the reading position of each line sensor is shifted by 0.5 pixel in G with respect to R and B, the aliasing noise in the image generated by the G line sensor will be the same as in the image generated by the R and B line sensors. The aliasing noise occurs at a different position. Therefore, the RGB values at the position where the aliasing noise occurs do not satisfy R=G=B, and a chromatic pixel occurs. Therefore, when a monochrome original is read, false colors occur, resulting in deterioration of image quality and erroneous determination of automatic color selection (ACS: Auto Color Select) (determining a monochrome original as a color original).

Further, in Patent Document 2, the light source needs to be an RGB light source including an R light source, a G light source, and a B light source, and cannot be used for a white light source, which limits the usable light sources. With a white light source, light is emitted by arranging multiple light sources in the main scanning direction, but when using the same configuration with an RGB light source, the number of light sources must be tripled to obtain the same illuminance, and the area required will increase. If the number of light sources is reduced to increase the amount of light, uneven bright spots will occur and the arrangement of LEDs will appear in the image data. For this reason, RGB light sources employ a method of diffusing and irradiating light using a light guide, but when the reading speed increases, it is difficult to secure a sufficient amount of light.

An object of the present invention is to suppress deterioration in image quality due to misalignment of reading positions of a plurality of light receiving element arrays.

The image reading device of the present invention has a plurality of light receiving element rows in which a plurality of light receiving elements are arranged in a first direction in a second direction perpendicular to the first direction, and reads an image of an object. A reading means for generating read data, a control means for independently controlling charge accumulation timing of the plurality of light receiving element rows, and correcting a reading position shift of the read data of the plurality of light receiving element rows generated by the reading means. the plurality of light-receiving element rows include a first light-receiving element row, a second light-receiving element row, and a third light-receiving element row, and the second light-receiving element row The charge accumulation timing is different for the first and third light-receiving element rows, and the control means is configured such that the first and third light-receiving element rows receive light for 0.5 pixels in the first half, and the second The charge accumulation timing is controlled so that the light-receiving element array receives light for 0.5 pixels in the latter half .

According to the present invention, it is possible to suppress deterioration of image quality due to misalignment of reading positions of a plurality of light receiving element arrays.

FIG. 1 is a configuration diagram of an image reading device. It is a block diagram of a CIS line sensor. It is a block diagram of a control part. 3 is a flowchart showing processing when reading an image. FIG. 3 is a diagram showing an example of light reception timing. It is a figure which shows an example of the image at the time of 600 dpi reading. It is a figure which shows an example of the image at the time of 300dpi reading. It is a figure which shows an example of the image at the time of 300dpi reading. FIG. 4 is a diagram showing an example of light reception timing during 300 dpi reading. It is a figure which shows an example of the image at the time of 300dpi reading. It is a figure which shows an example of the image at the time of 300dpi reading. FIG. 4 is a diagram showing an example of light reception timing when reading at 200 dpi. FIG. 3 is a diagram showing an arrangement of line sensors. It is a block diagram of a control part. 3 is a flowchart showing processing when reading an image. FIG. 3 is a diagram showing an example of gain adjustment.

Embodiments of the present invention will be described in detail below with reference to the drawings. Note that the following embodiments do not limit the claimed invention, and not all combinations of features described in the embodiments are essential to the solution of the invention.

(First embodiment)
FIG. 1 is a diagram showing an example of an image reading device 115 according to the first embodiment. The image reading device 115 is, for example, a scanner, a copying machine, a multifunction printer, or the like. The image reading device 115 is equipped with an automatic document feeder 100, and reads images on both sides of the document 102 using a front side reading section 121 and a back side reading section 122 while conveying the document 102, which is the object of image reading. I can do it.

The paper feed roller 103 is connected to the same driving source as the separation conveyance roller 104, rotates as the separation conveyance roller 104 rotates, and feeds the document. The paper feed roller 103 is normally retracted to its home position, which is an upper position in FIG. 1, so that it does not interfere with the document setting operation. When the paper feeding operation is started, the paper feeding roller 103 descends and comes into contact with the upper surface of the original 102 . The paper feed roller 103 is pivotally supported by an arm (not shown), and moves up and down as the arm swings.

A separation and conveyance driven roller 105 is arranged on the side opposite to the separation and conveyance roller 104 in a state where it is pressed against the separation and conveyance roller 104 side. The separation conveyance driven roller 105 is made of a rubber material or the like that has slightly less friction than the separation conveyance roller 104, and cooperates with the separation conveyance roller 104 to move the paper on the document tray 101 fed by the paper feed roller 103. The originals 102 are separated and fed one by one.

Further, the automatic document feeder 100 includes a registration roller 106, a registration driven roller 107, a lead roller 108, and a lead driven roller 109. The registration roller 106 and the registration driven roller 107 operate to align the leading edges of the documents fed at the separating section.

The lead roller 108 and the lead driven roller 109 transport the original 102 toward the scanning glass 116 . A platen roller 110 is arranged above the panning glass 116. The document 102 is conveyed by the platen roller 110 to a lead discharge roller 111 and a lead discharge driven roller 112 after passing over a scanning glass 116 that constitutes a first reading section.

At the end of the scanning glass 116 on the lead discharge roller 111 side, a jump base 117 for scooping up the document is provided in order to smoothly convey the document to the lead discharge roller 111.

The document 102 is conveyed so that its front surface is in contact with the scanning glass 116 , and at this time, a front surface reading section 121 disposed in the image reading device 115 reads the surface of the document via the scanning glass 116 . The front surface reading section 121 is constituted by a CIS (Contact Image Sensor) line sensor. Details will be described later.

The lead discharge roller 111 and the lead discharge driven roller 112 convey the document that has passed over the panning glass 116 to the panning glass 120 . A platen roller 119 is provided on one side of the scanning glass 120, and a back side reading section 122 is provided on the other side. Like the front side reading unit 121, the back side reading unit 122 is configured by a CIS (Contact Image Sensor) line sensor.

With this configuration, the back side image of the document 102 passing between the scanning glass 120 and the platen roller 119 is read by the back side reading section 122. Thereafter, the original 102 is conveyed to a paper ejection roller 113 and ejected to a paper ejection tray 114.

The image reading device 115 having the above configuration can read a document in two modes. The first mode is a fixed original reading mode in which the original 102 placed on the original platen glass 118 is read while the front side reading unit 121 is moved in the sub-scanning direction (arrow direction in the figure). The second mode is a panning mode in which the document is read at the panning glass 116 position while the automatic document feeder 100 transports the document 102 with the front side reading unit 121 stopped. In the second mode, by using the back side reading unit 122 in addition to the front side reading unit 121, it is possible to read the back side image in addition to the front side image of the document.

FIG. 2 is a diagram showing a configuration example of the CIS line sensor of the front side reading section 121 and the back side reading section 122 in FIG. 1. Since the front side reading unit 121 and the back side reading unit 122 have the same configuration, the configuration of the front side reading unit 121 will be explained here, and the explanation of the configuration of the back side reading unit 122 will be omitted.

The front surface reading section 121 is a reading unit that includes white LEDs 201a and 201b that are light emitting sections, a SELFOC (registered trademark) lens array 202, and a light receiving element 203 that is a light receiving section. The white LEDs 201a and 201b emit light to illuminate the surface of the original 102 when the original 102 passes through the pouring glass 116. The surface of the original 102 reflects the light emitted by the white LEDs 201a and 201b. The light reflected by the surface of the original 102 is diffused light. The diffused light passes through the SELFOC lens array 202 and is focused on the light receiving surface of the light receiving element 203. The light receiving element 203 is an image sensor, photoelectrically converts the received light, and outputs an analog electrical signal according to the amount of received light.

FIG. 13 is a diagram showing an arrangement of CIS line sensors of the front surface reading section 121 of FIG. 1. Since the front side reading unit 121 and the back side reading unit 122 have the same configuration, the configuration of the front side reading unit 121 will be explained here, and the explanation of the configuration of the back side reading unit 122 will be omitted.

FIG. 13 shows the arrangement of the light receiving elements 203 in FIG. 2. The line sensor is a light-receiving element array in which light-receiving elements 203 are arranged in a line in the main scanning direction in a number corresponding to the resolution. The plurality of line sensors are arranged offset in the sub-scanning direction. The surface reading unit 121 has a plurality of light receiving element rows in which a plurality of light receiving elements 203 are arranged in the main scanning direction in a sub-operation direction perpendicular to the main scanning direction, and generates read data by reading an image of the object. . The back side reading section 122 is also similar to the front side reading section 121.

In FIG. 13, a line of the light receiving element 203 reads an R (red) image, a line of the light receiving element 203 reads a G (green) image, and a line of the light receiving element 203 reads a B (blue) image. To this end, R, G, and B color filters are coated on the light-receiving surfaces of the three lines of light-receiving elements 203, respectively. As a result, the light diffused by the original 102 is separated into colors, and each light receiving element 203 receives the light that has passed through the color filter.

Since the three lines of light receiving elements 203 are arranged offset in the sub-scanning direction, the image reading device 115 does not read the R, G, and B color images from the same position on the document at the same timing. Therefore, it is necessary to align the image signals output from the light receiving elements 203 of each line. In the example of FIG. 13, RGB alignment can be performed by shifting G by one pixel and B by two pixels relative to R in the sub-scanning direction.

As shown in FIG. 13, the line sensor includes a plurality of lines of light-receiving elements 203 in the main scanning direction of the document 102, the number of which corresponds to the resolution, in the sub-scanning direction (conveyance direction). For example, the line sensor includes three lines of light-receiving elements 203 for 7500 pixels in the main scanning direction in the sub-scanning direction. The three lines of light-receiving elements 203 are arranged at 600 dpi and shifted by one pixel in the sub-scanning direction. A color filter is coated on the light receiving element 203 of each line. A color filter that transmits R light is coated on the light receiving element 203 of the first line. A color filter that transmits G light is applied to the light receiving element 203 of the second line. A color filter that transmits B light is applied to the light receiving element 203 of the third line.

FIG. 3 is a diagram showing an example of the configuration of the control section of the image reading device 115. The image reading device 115 includes a CPU (Central Processing Unit) 301, a nonvolatile memory 302, an operation section 303, an image processing section 306, a parallel/serial conversion section 310, and an image output controller 311. A front side reading section 121 and a back side reading section 122 are connected to the CPU 301 and the image processing section 306.

CPU 301 controls the operations of automatic document feeder 100 and image reading device 115 by executing programs stored in nonvolatile memory 302 . The operation unit 303 is a user interface for setting the double-sided reading mode of the document 102, setting the reading resolution, setting the destination of image data representing the read image, and the like. Contents input through the operation unit 303 are transmitted to the CPU 301.

The image processing unit 306 acquires the image data read by the front side reading unit 121 and the back side reading unit 122, performs predetermined image processing, and outputs the image data. The image processing section 306 includes A/D conversion sections 304 and 305, a line memory 307, a shading correction section 308, and a color shift correction section 209.

The A/D converters 304 and 305 convert the analog electrical signals read by the front side reading unit 121 and the back side reading unit 122 into image data that is a digital signal. Line memory 307 is a memory that stores image data converted by A/D converters 304 and 305.

The shading correction unit 308 performs shading correction processing on the read data of each color of R, G, and B stored in the line memory 307 to correct the effects of non-uniformity of light amount and sensitivity differences among the light receiving elements 203. . In the shading correction process, the correction process is performed using shading coefficients obtained by reading a white reference plate (not shown).

The color shift correction unit 309 performs color shift correction processing to align the reading positions on the read R, G, and B image data. In the color shift correction process, a process of correcting a color shift of an integer number of pixels and a process of correcting a color shift of a non-integer number of pixels are performed. Here, the correction is performed, for example, by filter processing with a window size of 1×3 (main scanning×sub-scanning). Here, if the image data for three pixels in the sub-scanning direction are PV1, PV2, and PV3, and the weighting coefficients are w1, w2, and w3, the pixel value PV2' after filter processing is calculated by equation (1).
PV2'=(PV1×w1+PV2×w2+PV3×w3)/(w1+w2+w3)...(1)

The parallel/serial converter 310 converts the read data of each color into serial data and sends it to the image output controller 311. The image output controller 311 outputs the read data converted into serial data as image data representing the read image of the document 102.

FIG. 4 is a flowchart illustrating a method of controlling the image reading device 115. A program that executes this flowchart is stored in nonvolatile memory 302 and executed by CPU 301.

In step S401, the CPU 301 acquires information set on the operation unit 303 and acquires reading conditions. For example, the CPU 301 acquires information such as resolution settings set on the operation unit 303.

In step S402, the CPU 301 determines the resolution information acquired in step S401. If the resolution information is 600 dpi (YES in S402), the CPU 301 proceeds to step S403, and if the resolution information is 300 dpi (NO in S402), the process proceeds to step S406.

In step S403, the CPU 301 instructs the automatic document feeder 100 to start document feeding at the conveyance speed setting for 600 dpi (normal mode).

In step S404, the CPU 301 instructs the R, G, and B color line sensors to start reading simultaneously at time Tn.

FIG. 5 shows a reset pulse that instructs to start reading when reading a document using the line sensor shown in FIG. ing. In the R, G, and B light receiving elements 203 in the figure, a high level indicates a period in which power is being stored, and a low level indicates a period in which power is not stored. When the R, G, and B light-receiving elements 203 detect the reset pulse at time Tn, they reset the stored charge and start storing power after a certain delay period. Therefore, as shown in FIG. 5, the R, G, and B light receiving elements 203 detect the reset pulse at time Tn, and a delay period occurs until the high-level state in which power is stored. The period from time Tn to Tn+1 shown in FIG. 5 corresponds to the charge of one sub-scanning pixel.

In step S405, the CPU 301 determines whether reading of the original 102 has been completed. If the reading of the original 102 is not completed, the CPU 301 returns to step S404 and instructs the reading of the next line. When the reading of the original 102 is completed, the CPU 301 ends the process of the flowchart in FIG. 4 .

Image data generated during 600 dpi reading will be explained using FIGS. 6(a) to 6(c). FIG. 6A is a diagram illustrating an example of sampling positions during 600 dpi reading when the original 102 is read at 600 dpi and a black line with a 4-pixel period is read. Arrows 601 indicate positions read by the line sensors of each color during a period from time Tn to Tn+1. As shown in FIG. 13, the line sensor is arranged such that G is shifted from R by 1 pixel at 600 dpi, and B is shifted from R by 2 pixels at 600 dpi in the sub-scanning direction. Therefore, as indicated by an arrow 601, the positions of images read at the same time are shifted.

FIG. 6B is a diagram showing the signal values of each color of R, G, and B before color shift correction, in which the read signal values are converted into digital signals by the A/D converters 304 and 305. In the signal value, the vertical axis represents the luminance value, and the horizontal axis represents the sub-scanning position. An arrow 601 in FIG. 6(b) indicates an image position read at the same time as the arrow 601 shown in FIG. 6(a). It can be seen that the signal value read from the black line portion of the original 102 is delayed by 600 dpi, one pixel at a time. In the part indicated by the arrow 601 in FIG. 6(b), the signal value of R is low, and the signal values of G and B are high, so that R≠(G=B). For this reason, in the original 102, the portions that were black lines become chromatic, and the image quality deteriorates.

FIG. 6C is a diagram showing the result of color shift correction for an integer number of pixels performed by the color shift correction unit 309 on the signal values in FIG. 6B. The signal value in FIG. 6(c) is an image obtained by correcting G by 1 pixel and B by 2 pixels based on R with respect to the signal value in FIG. 6(b), and the color of the R, G, and B signals is The deviation is corrected and it is possible to reproduce the original with black lines.

Next, using FIGS. 7(a) to (d) and FIGS. 8(a) to (d), when performing 300 dpi reading using the line sensor shown in FIG. An example of image data when the line sensors start reading at the same time will be described.

FIG. 7A is a diagram showing an example of sampling positions for 300 dpi reading when the original 102 is read at 600 dpi and a black line with a 4-pixel period is read. Arrows 701 in FIG. 7A indicate positions read by the line sensors of each color from time Tn to Tn+1. For example, in the line sensor, G is shifted from R by 1 pixel at 600 dpi, and B is shifted from R by 600 dpi by 2 pixels in the sub-scanning direction. When a line sensor with this configuration performs 300 dpi reading, the G line sensor reads a position shifted by 300 dpi and 0.5 pixel with respect to R and B.

FIG. 7B is a diagram showing signal values before color shift correction, which are obtained by converting the read signal values into digital signals by the A/D converters 304 and 305. An arrow 701 in FIG. 7(b) indicates an image position read at the same time as the arrow 701 shown in FIG. 7(a). The signal value read from the black line portion of the original 102 is a signal that is shifted by 300 dpi and 1 pixel for B compared to R, but is shifted by 300 dpi and 0.5 pixel for G compared to R.

FIG. 7C is a diagram showing the result of color shift correction for an integer number of pixels performed by the color shift correction unit 309 on FIG. 7B. The signal value in FIG. 7C is obtained by correcting B by 300 dpi by one pixel with respect to the signal value in FIG. 7B. In this state, since the G signal value is shifted by 300 dpi and 0.5 pixel, the part indicated by the arrow 701 becomes (R=B)≠G, and becomes a chromatic color.

FIG. 7D is a diagram showing a state in which the color shift correction unit 309 performs color shift correction of 0.5 pixels at 300 dpi on the G signal value in contrast to FIG. 7C. When formula (1) is used, 0.5 pixel color shift correction is performed using the weighting coefficient of the following formula.
PV2'=(PV1×0+PV2×64+PV3×64)/128

Due to the color shift correction processing, the signal values of R, G, and B in the part indicated by the arrow 701 become R=G=B, and the color shift of R, G, and B is corrected, but the signal values are not generated due to the smoothing effect of the correction processing. Blur occurs in the image data, and the image quality deteriorates.

FIG. 8A is a diagram showing an example of sampling positions for 300 dpi reading when the original 102 is read at 600 dpi and a black line with a 1.5 pixel period. Arrows 801 in FIG. 8A indicate positions read by line sensors of each color at the same time.

FIG. 8(b) is a diagram showing signal values after A/D conversion processing. FIG. 8C is a diagram showing signal values after color shift correction has been performed for an integer number of pixels. FIG. 8(d) is a diagram showing a state in which color shift correction of 0.5 pixel is performed on the G signal value with respect to the signal value of FIG. 8(c).

A black line of 600 dpi and a 1.5 pixel period has a periodicity of 200 dpi. Since the Nyquist frequency when reading at 300 dpi is 150 dpi, a black line with a period of 200 dpi cannot be resolved, and aliasing noise occurs. Since G is read at a position shifted by 0.5 pixel from R and B at 300 dpi, the aliasing noise generated in G is generated as a signal shifted by a half cycle of the aliasing noise. Therefore, as shown in FIG. 8D, even if color shift correction processing is performed for 0.5 pixels, R=G=B is not established, and chromatic pixels occur in the black line portion. For example, an image reading device of a document reader or a copying machine has a function such as ACS (Auto Color Select) that determines whether a read document is color or monochrome. When reading at 300 dpi with an image reading device having this function, the ACS function may determine that a monochrome original is a color original due to the generation of chromatic pixels due to aliasing noise.

The present embodiment provides an image reading device 115 that suppresses the above-described image quality deterioration and ACS misjudgment. Returning to the explanation of FIG. 4.

In step S402, if the resolution information is 300 dpi (NO in S402), the CPU 301 proceeds to step S406.

In step S406, the CPU 301 instructs the automatic document feeder 100 to start document feeding at a conveyance speed setting for 300 dpi (high speed mode). For example, the CPU 301 feeds the document at double speed for 600 dpi.

In step S407, the CPU 301 starts reading the R and B line sensors simultaneously at time Tn. In step S408, the CPU 301 starts reading the G line sensor at time Tn'.

FIG. 9 shows how the line sensor shown in FIG. 13 is used to detect the reset pulse for each R, G, and B that instructs the start of reading when reading a document at 300 dpi, and to detect the reset pulse for R, G, and B. 5 is a diagram showing the timing at which the light receiving element 203 stores power. FIG. For the R, G, and B light receiving elements 203, a high level indicates a period in which power is being stored, and a low level indicates a period in which power is not stored. When the R and B light-receiving elements 203 detect the reset pulse at time Tn, they reset the stored charge and start storing power after a certain delay period. On the other hand, when the G light receiving element 203 detects the reset pulse at time Tn', it resets the stored charge and starts storing power after a certain delay period. The period from time Tn to Tn+1 becomes the charge of one R and B sub-scanning pixel. Further, for G, the interval from Tn' to Tn+1' is the charge of one sub-scanning pixel of G. In this way, the reset pulse is controlled independently for each color, and control is performed so that R and B receive light for 0.5 pixels in the first half of 300 dpi, and G for 0.5 pixels in the latter half of 300 dpi.

Time Tn' is a time shifted by 300 dpi and 0.5 pixel reading time tm from time Tn, and Tn'=Tn+tm. Also, it is controlled so that R and B are received by 0.5 pixels in the first half of 300 dpi, and G by 0.5 pixels in the latter half of 300 dpi. It may also be controlled to receive light.

The CPU 301 is a control unit that independently controls the charge accumulation timing of the R, G, and B light receiving element arrays using a reset pulse. The CPU 301 controls the charge accumulation timings of the R, G, and B light receiving element arrays at different timings. The G light-receiving element array has a different charge accumulation timing from the R and B light-receiving element arrays. The CPU 301 controls the charge accumulation timing so that the R and B light-receiving element arrays receive light for 0.5 pixels in the first half, and the G light-receiving element array receives light for 0.5 pixels in the latter half.

In step S409, the CPU 301 determines whether reading of the original 102 has been completed. If the reading of the original 102 has not been completed, the CPU 301 returns to step S407 and reads the next line. When the reading of the original 102 is completed, the CPU 301 ends the process of the flowchart in FIG. 4 .

Next, image data generated during 300 dpi reading will be described using FIGS. 10(a) to 10(c) and FIGS. 11(a) to (c). FIG. 10A is a diagram showing an example of sampling positions for 300 dpi reading when the original 102 is read at 600 dpi and a black line with a period of 4 pixels is read. Arrows 1001 indicate the positions of the R and B light receiving elements 203 that are read from time Tn to Tn+1 shown in FIG. 9, and the positions of the G light receiving elements 203 that are read from time Tn' to Tn+1'. As shown in FIG. 10(a), R and G are read at the same time and at the same position. B reads a position shifted by 1 pixel at 300 dpi.

In this way, by setting the light received by the R and B light receiving elements 203 to be the first half of 0.5 pixels of 300 dpi, and the light receiving of the G light receiving element 203 being the second half of 0.5 pixels of 300 dpi, the R, G, and B It becomes possible to perform color shift correction by integer correction. This makes it possible to suppress image blurring caused by color shift correction for 0.5 pixels.

FIG. 10(b) is a diagram showing signal values after A/D conversion processing. FIG. 10C is a diagram showing signal values obtained by correcting color shift for an integer number of pixels, and it is possible to suppress deterioration due to color shift correction for 0.5 pixels.

The color shift correction unit 309 corrects the read position shift of the read data of the R, G, and B light receiving element arrays, as shown in FIG. 10(c). At this time, the color shift correction unit 309 performs reading position shift correction for an integer number of pixels.

FIG. 11A is a diagram showing an example of sampling positions for 300 dpi reading when the original 102 is read at 600 dpi and a black line with a 1.5 pixel period. Arrows 1101 indicate the positions of the R and B light receiving elements 203 that are read from time Tn to Tn+1 shown in FIG. 9, and the positions of the G light receiving elements 203 that are read from time Tn' to Tn+1'.

FIG. 11(b) is a diagram showing signal values after A/D conversion processing. FIG. 11(c) is a diagram showing signal values after color shift correction has been performed for an integer number of pixels. A black line with a 1.5 pixel period at 600 dpi causes aliasing noise when read at 300 dpi, but since the same position is read for R, G, and B, aliasing noise also occurs at almost the same position. Therefore, as shown in FIG. 11C, by performing color shift correction for an integer, it is possible to suppress the occurrence of chromatic pixels, and it is possible to suppress ACS misjudgments.

In this embodiment, control during reading at 300 dpi has been described, but the present invention is not limited to this. FIG. 12 shows how the line sensor shown in FIG. 13 is used to detect the reset pulse for each R, G, and B that instructs the start of reading when reading a document at 200 dpi, and to detect the reset pulse for R, G, and B. 5 is a diagram showing the timing at which the light receiving element 203 stores power. FIG. For the R, G, and B light receiving elements 203, a high level indicates a period in which power is being stored, and a low level indicates a period in which power is not stored. When the R light receiving element 203 detects the reset pulse at time Tn, it resets the stored charge and starts storing power after a certain delay period. When the G light receiving element 203 detects the reset pulse at time Tn', it resets the stored charge and starts storing power after a certain delay period. When the B light receiving element 203 detects the reset pulse at time Tn'', it resets the stored charge and starts storing power after a certain delay period. The interval from time Tn to Tn+1 becomes the charge of one R sub-scanning pixel. Further, G is the charge of one sub-scanning pixel of G in the interval from time Tn' to Tn+1'. For B, the period from time Tn'' to Tn+1'' is the charge of one sub-scanning pixel of B. In this way, by independently controlling the reset pulse for each color, it becomes possible to perform R, G, and B color shift correction by integer corrections, similar to 300 dpi.

As described above, according to the present embodiment, even with a white light source, it is possible to correct the deviation of image signals output from multiple line sensors by aligning integer pixels, and it is possible to suppress deterioration of image quality. It is possible to provide an image reading device capable of

When the line sensor reads at a reading speed of 600 dpi, the CPU 301 controls the charge accumulation timing of the R, G, and B light receiving element arrays to be the same timing. Further, when the line sensor reads at a reading speed of 300 dpi, the CPU 301 controls the charge accumulation timings of the R, G, and B light receiving element arrays to be different from each other.

(Second embodiment)
In the first embodiment, a method has been described in which the light reception timings of the R, G, and B line sensors are individually controlled, thereby making it possible to correct color shift in integer minutes and suppressing deterioration of image quality. However, in the method of the first embodiment, the light reception time of each line sensor when reading at 300 dpi is shorter than the light reception time at 600 dpi, and the amount of light is insufficient.

FIG. 16 is a diagram illustrating an example of gain adjustment for adjusting the white level. If the amount of light is insufficient, the brightness value obtained when reading the original 102 will be low, so the shading correction unit 308 corrects the effects of non-uniformity of the amount of light and sensitivity differences between the light receiving elements 203, and at the same time adjusts the white level. Perform gain adjustment.

When the shading correction unit 308 performs gain processing as shown in FIG. 16 on the read signal, the difference between the processed outputs out1' and out2' is greater than the difference between the outputs out1 and out2 with respect to the inputs in1 and in2. growing. For this reason, S/N (Signal/Noise) deteriorates and image quality deteriorates.

In the second embodiment, a method of performing smoothing processing for noise removal when the light intensity is insufficient during 300 dpi reading and the S/N deteriorates will be described. Control of document reading will be described below with reference to FIGS. 14 and 15.

FIG. 14 is a diagram illustrating a configuration example of a control unit of an image reading device 115 according to the second embodiment. FIG. 14 shows that a smoothing processing unit 1401 is added to FIG. 3. The smoothing processing unit 1401 is provided within the image processing unit 306. Hereinafter, the differences between the second embodiment and the first embodiment will be explained.

The smoothing processing unit 1401 performs noise removal on the data generated by the color shift correction unit 309 through filter processing. The smoothing processing unit 1401 uses a low-pass filter as a noise removal filter in order to suppress high frequency noise. For example, the smoothing processing unit 1401 performs noise removal using a bilateral filter. The bilateral filter is expressed by the following equation (2).

Here, w is the kernel size, σ ₁ is the control value of the Gaussian filter, and σ ₂ is the control value of the brightness difference. The smaller σ ₁ is, the smaller the smoothing effect is, and the larger σ ₁ is, the larger the smoothing effect is. Further, when σ ₂ is increased, the smoothing effect on edges becomes larger, and when σ ₂ is reduced, the smoothing effect on edges becomes smaller, but the noise removal effect also becomes weaker.

FIG. 15 is a flowchart illustrating a method of controlling the image reading device 115 according to the second embodiment. A program that executes this flowchart is stored in nonvolatile memory 302 and executed by CPU 301. FIG. 15 differs from FIG. 4 in that steps S1501 and S1502 are added. Hereinafter, the differences between FIG. 15 and FIG. 4 will be explained.

In step S402, if the resolution information is 600 dpi (YES in S402), the process advances to step S1501, and if the resolution information is 300 dpi (NO in S402), the process advances to step S1502.

In step S1501, the CPU 301 sets a weak smoothing processing coefficient for 600 dpi in the smoothing processing unit 1401, and ends the setting processing.

In step S1502, the CPU 301 sets a strong smoothing processing coefficient for 300 dpi in the smoothing processing unit 1401, and ends the setting processing.

The smoothing processing unit 1401 is a smoothing unit, and uses different coefficients for the read data corrected by the color shift correction unit 309 when the line sensor reads at a reading speed of 600 dpi and when the line sensor reads at a reading speed of 300 dpi. Perform smoothing with . The smoothing processing unit 1401 performs smoothing using, for example, a bilateral filter.

Here, the strong smoothing processing coefficient for 300 dpi is determined by, for example, making σ ₁ in equation (2) larger than the weak smoothing processing coefficient for 600 dpi, or making σ ₂ larger than the weak smoothing processing coefficient for 600 dpi. You can set it to . Further, both σ ₁ and σ ₂ may be set to be changed.

Furthermore, if it is desired to enhance the noise removal effect while preserving edges, a configuration may be adopted in which a plurality of bilateral filters are arranged and noise is removed by repeatedly processing the bilateral filters during 300 dpi reading.

As described above, according to this embodiment, even if the light intensity is insufficient for 300 dpi reading compared to 600 dpi reading, noise can be appropriately reduced by changing the smoothing process according to the reading resolution. This makes it possible to suppress deterioration of image quality.

According to the first and second embodiments, even when using a white light source, the image reading device 115 can correct deviations in image signals output from a plurality of line sensors by aligning integer pixels, thereby improving image quality. deterioration can be suppressed.

(Other embodiments)
The present invention provides a system or device with a program that implements one or more functions of the embodiments described above via a network or a storage medium, and one or more processors in a computer of the system or device reads and executes the program. This can also be achieved by processing. It can also be realized by a circuit (for example, ASIC) that realizes one or more functions.

Note that the above embodiments are merely examples of implementation of the present invention, and the technical scope of the present invention should not be interpreted to be limited by these embodiments. That is, the present invention can be implemented in various forms without departing from its technical idea or main features.

121 front side reading unit, 122 back side reading unit, 301 CPU, 302 nonvolatile memory, 303 operation unit, 308 shading correction unit, 309 color shift correction unit

Claims (9)

a reading means having a plurality of light receiving element rows in which a plurality of light receiving elements are arranged in a first direction in a second direction perpendicular to the first direction, and generating read data by reading an image of the object; ,
control means for independently controlling charge accumulation timing of the plurality of light receiving element arrays;
and a correction means for correcting a reading position shift of the read data of the plurality of light receiving element arrays generated by the reading means,
The plurality of light-receiving element rows include a first light-receiving element row, a second light-receiving element row, and a third light-receiving element row,
The second light-receiving element array has a different charge accumulation timing from the first and third light-receiving element arrays,
The control means controls the charge accumulation timing so that the first and third light-receiving element arrays receive light for 0.5 pixels in the first half, and the second light-receiving element array receives light for 0.5 pixels in the latter half. An image reading device characterized by: 2. The image reading apparatus according to claim 1, wherein the correction means corrects a reading position shift for an integer number of pixels. 3. The image reading apparatus according to claim 1, wherein the control means controls the charge accumulation timing using a reset pulse. 4. The image reading device according to claim 1, wherein the control means controls the charge accumulation timings of the plurality of light receiving element arrays at different timings. A red color filter is applied to the first light receiving element row,
A green color filter is applied to the second light receiving element row,
5. The image reading device according to claim 1 , wherein the third light-receiving element array is coated with a blue color filter. The control means includes:
When the reading means reads at the first reading speed, charge accumulation timings of the plurality of light receiving element arrays are controlled to be the same timing,
6. When the reading means reads at the second reading speed, the charge accumulation timings of the plurality of light-receiving element arrays are controlled at different timings . Image reading device. The apparatus further includes a smoothing means for smoothing the read data corrected by the correction means using different coefficients when the reading means reads at a first reading speed and when reading at a second reading speed. The image reading device according to any one of claims 1 to 6 , characterized in that: 8. The image reading device according to claim 7 , wherein the smoothing means performs smoothing using a bilateral filter. a reading step of having a plurality of light receiving element rows in which a plurality of light receiving elements are arranged in a first direction in a second direction perpendicular to the first direction, and generating read data by reading an image of the object; ,
a control step of independently controlling charge accumulation timing of the plurality of light receiving element arrays;
a correction step of correcting a reading position shift of the read data of the plurality of light receiving element arrays generated in the reading step ,
The plurality of light-receiving element rows include a first light-receiving element row, a second light-receiving element row, and a third light-receiving element row,
The second light-receiving element array has a different charge accumulation timing from the first and third light-receiving element arrays,
In the control step, the charge accumulation timing is controlled such that the first and third light receiving element arrays receive light for 0.5 pixels in the first half, and the second light receiving element array receives light for 0.5 pixels in the latter half. A method of controlling an image reading device, characterized in that:

Images