JP2021114759A - Image reading apparatus - Google Patents

Abstract

To provide an image reading apparatus suppressed in deterioration of detection accuracy of an oblique amount of an original.SOLUTION: An image reading apparatus comprises: an original tray 102 on which an original is mounted; a size detection part 1020 detecting a main scanning size of the original mounted on the original tray 102; an ADF feeding the original mounted on the original tray 102; a first reading unit 109A reading an image from the original to be fed, and outputting image data expressing the read image; an image memory 205 storing the image data output from the first reading unit 109A; a front surface registration calculation part 207 that sets a main scanning pixel width as an effective region used for detecting an oblique amount of the original on the basis of the main scanning size detected by the size detection part 1020, and calculates the oblique amount from the image data in the set main scanning pixel width; and a registration correction part 208 that performs an oblique correction of the image data stored in the image memory 205 on the basis of the oblique amount.SELECTED DRAWING: Figure 2

Description

The present invention relates to an image reader that reads an image of a conveyed document.

Conventionally, an image reading device including an automatic document transporting device (hereinafter, referred to as "ADF: Auto Document Feeder") is known. The image reading device reads the images of the sheet-shaped originals, which are conveyed one by one to the reading position by the ADF, line by line with the direction orthogonal to the conveying direction of the originals as the main scanning direction. By using the ADF, the image reading device can continuously read the images of a plurality of originals.

In such an image reading device, the side of the tip side of the document in the transport direction is tilted with respect to the main scanning direction due to variations in the nip pressure and rotation speed of the transport roller that transports the document, manufacturing error, and the like. It may be transported in a state. This is called "skew". If the original is scanned in a skewed state, the scanned image will also be tilted.

Patent Document 1 discloses an image reading device that rotates and corrects image data representing a reading result. This image reading device detects the tip edge of the document based on the image data obtained by scanning the document. Specifically, the image reading device detects the tip edge of the document based on the image data near the center position excluding the region corresponding to the corner portion of the document in the main scanning direction. Further, the image reading device detects the skew amount of the document based on the detected tip edge, and performs skew correction (rotation correction) on the image data according to the detected skew amount. By removing the predetermined area corresponding to the corner of the document, it is possible to prevent erroneous detection of the skew amount due to damage or folding at both ends of the document. In the following description, the area excluding the predetermined area corresponding to the corner portion of the document in the main scanning direction, that is, the area in the main scanning direction of the image data used when detecting the tip edge is referred to as an “edge detection area”. ..

JP-A-2017-92562

Damage or folds at both ends of the document can occur when the needle-bound document bundles are separated one by one for transport by the ADF. For example, in general, a small-sized sheet such as a business card is less likely to be needle-bound than an A4 size document. In the configuration disclosed in Patent Document 1, image data in a predetermined region is excluded when detecting the tip edge even for a small-sized sheet such as a business card that is unlikely to be needle-bound. If the image data in a predetermined area is removed when the original is a small-sized sheet such as a business card and the tip edge is detected, the pixel width in the main scanning direction required for detecting the skew amount is not secured, and the skew amount is not secured. Detection accuracy may decrease. A decrease in the detection accuracy of the skew amount causes distortion of the read image and hinders the output of a preferable product.

In view of the above problems, it is a main object of the present invention to provide an image reading device that suppresses a decrease in detection accuracy of the skew amount of a document.

The image reading device of the present invention includes a document tray on which a document is placed, a transport means for transporting the document placed on the document tray in a transport direction, a light receiving element that receives light, and the light receiving element. A reading unit that has a generating means for generating pixel data representing a light receiving result and reads an image of the document conveyed by the conveying means, and a document whose image is read by the reading unit are mainly orthogonal to the conveying direction. The acquisition means for acquiring information on the length in the scanning direction and the positions of one end and the other end of the image of the original document represented by the plurality of pixel data generated by the generation means in the main scanning direction. Invalidity which is a region in the main scanning direction of pixel data which is not used for determining the first determining means for determining and the amount of tilt corresponding to the tilt angle of the side on the tip end side of the document in the transport direction with respect to the main scanning direction. Based on the setting means that sets the area based on the information acquired by the acquisition means, and the pixel data corresponding to the side on the tip end side of the document included in the area excluding the invalid area in the main scanning direction. A second determining means for determining the tilt amount, and a correction means for correcting the tilt of the image represented by the pixel data so that the tilt amount determined by the second determining means is reduced. The invalid area is a region between a position separated from the one end portion by a set distance in the main scanning direction and a position of the one end portion, and a position separated from the other end portion by the set distance in the main scanning direction. The set distance is a region between the position of the other end and the position of the one end in the main scanning direction of the image of the document represented by the plurality of pixel data. The distance is shorter than the distance between the two, and in the setting means, the length of the invalid region in the main scanning direction when the length of the document in the main scanning direction is the first length is the length of the document in the main scanning direction. The set distance is set so that the length in the main scanning direction is longer than the length in the main scanning direction of the invalid region when the second length is shorter than the first length. And.

According to the present invention, it is possible to suppress a decrease in the detection accuracy of the skew amount of the document.

The block diagram of an image reader. Explanatory drawing of the controller. Explanatory drawing of storage timing of image data in image memory. (A) to (c) are explanatory views of processing of an edge extraction part. Explanatory drawing of the calculation method of registration information. (A) and (b) are explanatory views of registration correction. (A) to (c) are explanatory views of the main scanning pixel width. A flowchart showing an image reading process. A flowchart showing an image reading process. The block diagram of an image reader. Top sectional view of the ADF. Configuration diagram of the control system. A flowchart showing an image reading process. A flowchart showing the setting process of the edge detection area. A flowchart showing an image reading process. A flowchart showing the setting process of the edge detection area.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

1st Embodiment (configuration of image reading device)
FIG. 1 is a configuration diagram of an image reading device of the present embodiment. The image reading device 10 includes an ADF 100 that conveys a sheet-shaped document 101, and a reader 30 that reads an image (original image) of the document 101 conveyed by the ADF 100. The ADF 100 is rotatably attached to the housing of the reader 30 by a hinge (not shown). The reader 30 includes a first reading unit 109A and reads a document image on the first surface (front surface) of the document 101. The ADF 100 includes a second reading unit 109B, and reads a document image on the second surface (back surface) of the document 101.

The ADF 100 includes a document tray 102 capable of loading one or more documents 101, a transport path for transporting the documents 101, and an ejection tray 121 for ejecting the document 101 after image scanning. The reading position A of the document 101 by the first reading unit 109A and the reading position B of the document 101 by the second reading unit 109B are provided in the middle of the transport path. In the present embodiment, the reading position A of the first reading unit 109A is provided on the upstream side of the reading position B of the second reading unit 109B in the transport direction of the document 101.

The document tray 102 is provided with a side regulating plate 123 that can move in a direction (width direction) orthogonal to the transport direction of the document 101. The side regulating plate 123 abuts on the end of the document 101 in the width direction to regulate the position of the document 101 in the width direction. In the present embodiment, a configuration in which two side regulation plates 123 are provided to regulate both ends in the width direction of the document 101 will be described. However, one side regulation plate 123 is provided and one of the side regulation plates 123 in the width direction of the document 101. It may be configured to regulate only the end of the. When there is one side regulating plate 123, the other end in the width direction of the document 101 is regulated by the fixing regulating plate.

The two side regulation plates 123 are interlocked with each other by moving one of them by an interlocking mechanism (not shown) provided inside the document tray 102, so that the other moves. In the present embodiment, the transport center of the document 101 is the center in the width direction. The two side control plates 123 are configured to approach or separate from the center in the width direction. Therefore, the center of transport of the document 101 is the same as the center of the document 101 in the width direction. As the side regulation plate 123 moves, the resistance value of the volume resistor (not shown) provided inside the document tray 102 changes in tandem. The length in the width direction of the document 101 placed on the document tray 102 can be detected by the voltage value generated by the voltage drop due to the volume resistor. That is, the configuration including the side regulation plate 123 and the volume resistor functions as a size detection unit for detecting the size (length in the width direction) of the document 101. The length in the width direction is called "main scanning size".

In the ADF 100, the pickup roller 103, the separation rollers 104, 105, the front transfer roller 106, the lead roller 107, the lead roller 117, and the paper ejection as the transfer rotating body are sequentially arranged in the transfer path from the upstream side in the transfer direction of the document 101. A roller 120 is provided. A document detection sensor 113 for detecting the conveyed document 101 is provided on the downstream side of the lead roller 107. The reading position A of the first reading unit 109A is provided between the lead roller 107 and the lead roller 117. The reading position B of the second reading unit 109B is provided between the lead roller 117 and the paper ejection roller 120.

The pickup roller 103 is provided so as to be swingable, and when the document 101 is fed from the document tray 102, the pickup roller 103 comes into contact with the surface of the uppermost document 101 of the document loaded on the document tray 102. When paper feeding is not performed, the pickup roller 103 retracts upward so as not to interfere with the placement of the document 101 on the document tray 102. The pickup roller 103 takes in the document 101 from the document tray 102 and conveys it to the separation rollers 104 and 105. The pickup roller 103 is rotationally driven by a motor (not shown).

Since the pickup roller 103 feeds the document 101 only by the frictional force, a plurality of sheets may be fed at the same time depending on the friction coefficient of the document 101. Therefore, the separation roller 104 and the separation roller 105 separate the original documents 101 one by one. In the present embodiment, the separation roller 104 rotates in the direction of transporting the document 101, and the separation roller 105 is a non-rotating roller. By rotating only the separation roller 104, the documents 101 are separated one by one.

The separation roller 104 conveys the document 101 separated into one sheet to the pair of front transfer rollers 106. The front transfer roller 106 conveys the conveyed document 101 to the pair of lead rollers 107. The lead roller 107 conveys the conveyed document 101 to the reading position A of the first reading unit 109A. When the document 101 passes through the reading position A, the image on the first surface (front surface) is read by the first reading unit 109A.

If scanning is performed in a state where the posture of the document 101 is unstable, the distance between the first scanning unit 109A and the document 101 is not stable, and the shadow at the tip in the transport direction of the document cannot be detected stably. Therefore, at the reading position A, the surface guide plate 116 is arranged on the opposite side of the first reading unit 109A as an opposing member. A surface reading upstream roller 114 is provided on the upstream side of the surface guide plate 116, and a surface reading downstream roller 115 is provided on the downstream side. The first reading unit 109A side of the surface guide plate 116 is white. When the document 101 passes through the lower part of the surface guide plate 116, the image of the first surface (front surface) is read by the first reading unit 109A, so that the image is read in a stable posture.

The document 101 that has passed the reading position A of the first reading unit 109A is conveyed to the reading position B of the second reading unit 109B by the pair of lead rollers 117. The image on the second side (back side) of the document 101 is read by the second reading unit 109B when passing through the reading position B. At the reading position B, the back surface scanning glass 118 and the back surface guide plate 119 which is an opposing member are arranged. The back surface guide plate 119 is provided on the opposite side of the second reading unit 109B with the back surface scanning glass 118 interposed therebetween. The backside scanning glass 118 is a transparent member. The second reading unit 109B side of the back surface guide plate 119 is white. When the document 101 passes between the back surface scanning glass 118 and the back surface guide plate 119, the image on the second surface (back surface) is read by the second scanning unit 109B.

The document 101 that has passed the reading position B of the second reading unit 109B is ejected to the ejection tray 121 by the paper ejection roller 120. The second scanning unit 109B does not perform a scanning operation when it is not necessary to scan the document image on the second surface of the document 101.

When the document 101 is conveyed as described above, when the document detection sensor 113 detects the document 101, the start timing of the reading operation by the first reading unit 109A and the second reading unit 109B is determined based on the conveying speed of the document 101. NS.

The first reading unit 109A is provided in the housing of the reader 30. The housing of the reader 30 supports the platen glass 31 and the scanning glass 108. The ADF 100 opens and closes with respect to the platen glass 31 and the scanning glass 108 by rotating with respect to the reader 30. A white reference plate 122 is provided inside the housing between the platen glass 31 and the scanning glass 108. The white reference plate 122 serves as a reference reading member when acquiring shading data. The original 101 is placed on the platen glass 31 with the image forming surface facing the platen glass 31 side. When the image of the document 101 placed on the platen glass 31 is read, the first scanning unit 109A reads the image of the document 101 line by line while moving in the direction of arrow C.

The scanning glass 108 is provided at a position corresponding to the reading position A. The image of the document 101 conveyed by the ADF 100 is read by the first reading unit 109A as it passes between the scanning glass 108 and the surface guide plate 116. In this case, the first reading unit 109A is located directly below the scanning glass 108 and reads the image of the document 101 line by line.

The first reading unit 109A includes an LED (Light Emitting Diode) 110 which is a light source that emits light, an optical component group 112, and a reading sensor 111. The LED 110 irradiates the first surface of the document 101 passing through the reading position A with light. The light reflected by the first surface of the document 101 is guided to the reading sensor 111 by the optical component group 112, and is received by the reading sensor 111. The reading sensor 111 is configured by arranging a plurality of photoelectric conversion elements in the above-mentioned width direction. The direction in which the photoelectric conversion elements are arranged (width direction) is the main scanning direction. The plurality of photoelectric conversion elements are light receiving elements that receive reflected light, and generate an electric signal obtained by photoelectrically converting the received reflected light. The reading sensor 111 generates digital pixel data by performing predetermined processing on the electric signals generated by the plurality of photoelectric conversion elements. The pixel data corresponds to data representing the light receiving result of the photoelectric conversion element (pixel). That is, the image data representing the original image on the first surface of the original 101 is composed of a plurality of pixel data.

The second reading unit 109B has the same configuration as the first reading unit 109A and operates in the same manner. That is, the second reading unit 109B includes an LED that is a light source that emits light, an optical component group, and a reading sensor. The second scanning unit 109B reads the document image from the second surface of the document 101 passing through the scanning position B via the back surface scanning glass 118. The LED irradiates the second surface of the document 101 with light. The light reflected by the second surface of the document 101 is guided to the reading sensor by the optical component group and received by the reading sensor. The reading sensor generates pixel data based on the received reflected light.

The image reading device 10 having such a configuration can read the original image in two modes, a “scanning mode” and a “fixed reading mode”. In the scanning mode, the original image is read from the original 101 conveyed by the ADF 100. In the fixed reading mode, the original image is read from the original 101 placed on the platen glass 31. In this embodiment, since the skew correction of the document 101 in the scanning mode will be described, the description of the fixed reading mode will be omitted.

In the scanning mode, as described above, the document 101 is conveyed from the document tray 102 onto the scanning glass 108 at the reading position A of the first reading unit 109A. The first reading unit 109A moves directly under the scanning glass 108 when the scanning mode is started. The first scanning unit 109A reads the original image on the first surface line by line from the original 101 passing over the scanning glass 108.

When reading only the original image on one side (first side) of the original 101, the original 101 that has passed over the scanning glass 108 is discharged to the discharge tray 121 by the lead roller 117 and the paper ejection roller 120. At this time, the document 101 passes through the reading position B of the second reading unit 109B, but the second reading unit 109B does not perform the reading operation.

When reading the original image on both sides of the original 101, the original 101 is read by the second reading unit 109B while passing through the reading position B of the second reading unit 109B. The second scanning unit 109B reads the document image on the second surface line by line from the document 101 passing through the transport path.

(controller)
FIG. 2 is an explanatory diagram of a controller that controls the operation of the image reading device 10 having the above configuration. The controller 200 is composed of at least one ASIC (Application Specific Integrated Circuit), and realizes each function described below.
The controller 200 of this embodiment is provided in the image reading device 10. The controller 200 is connected to the transfer motor 201 and the operation unit 202 in addition to the document tray 102, the first reading unit 109A, the second reading unit 109B, and the document detection sensor 113. The transfer motor 201 is a drive source that rotationally drives each of the above rollers that convey the document 101. The operation unit 202 includes an input interface that is operated by the user and receives various instructions and input data from the user, and a display unit for notifying the user. The user can use the operation unit 202 to set the resolution in the main scanning direction when reading the image of the document 101, for example.

The controller 200 includes a CPU (Central Processing Unit) 203 and a non-volatile memory 209. The CPU 203 controls the operation of each part of the image reading device 10 by executing a predetermined computer program stored in the non-volatile memory 209. Instructions and input data are input to the CPU 203 from the operation unit 202. The CPU 203 starts the operation control of the image reading device 10 in response to the instruction acquired from the operation unit 202. In addition, the controller 200 includes a front surface shading unit 204A, a back surface shading unit 204B, an image memory 205, an edge extraction unit 206, a surface registration calculation unit 207, a registration correction unit 208, and an image inversion unit 210.

When the CPU 203 acquires an image reading instruction from the operation unit 202, the CPU 203 drives the transport motor 201 to start transporting the document 101, and also starts driving the first reading unit 109A and the second reading unit 109B. When the first scanning unit 109A and the second scanning unit 109B acquire the drive instruction from the CPU 203, the first scanning unit 109A and the second scanning unit 109B perform a scanning process of the image of the document 101. The first reading unit 109A and the second reading unit 109B transmit 8-bit (0 to 255) pixel data per pixel, which is a reading result, to the controller 200. The pixel data shows a higher value as the intensity of the reflected light increases. The level of this numerical value is hereinafter referred to as "luminance value". Further, when the CPU 203 acquires an image reading instruction from the operation unit 202, the CPU 203 acquires information indicating the main scanning size from the size detection unit 1020 of the document tray 102.

The controller 200 starts acquiring image data from the first reading unit 109A at a predetermined timing from the timing when the document detection sensor 113 detects the tip of the document 101 in the transport direction until the tip reaches the reading position A. .. Further, the controller 200 acquires image data from the second reading unit 109B at a predetermined timing from the timing when the document detection sensor 113 detects the tip of the document 101 in the transport direction until the tip reaches the reading position B. Start.

The reading sensor 111 of the present embodiment has 7488 photoelectric conversion elements (pixels) that receive light of three colors of red (R), green (G), and blue (B), respectively, in the main scanning direction.

The pixel data output from the first reading unit 109A is input to the surface shading unit 204A. The pixel data output from the second reading unit 109B is input to the back surface shading unit 204B. The front surface shading unit 204A and the back surface shading unit 204B perform shading correction based on the input pixel data. The shading correction corrects the non-uniformity of the light amount of the LED 110 in the main scanning direction and the influence of the sensitivity unevenness of each pixel of the reading sensor 111.

The pixel data on the first surface and the pixel data on the second surface after shading correction are stored in the image memory 205 at a predetermined timing. The image data representing the image on the second surface passes through the image inversion unit 210 before being stored in the image memory 205. The second reading unit 109B of the present embodiment has the same configuration as the first reading unit 109A. The second reading unit 109B is attached to the ADF100 in a form in which the direction of the main scanning direction is the same as that of the first reading unit 109A and the second reading unit 109B is turned upside down. Therefore, the image data of the second reading unit 109B is an image inverted in the main scanning direction with respect to the image data of the first reading unit 109A. The image inversion unit 210 aligns the direction with the image data on the first surface by inverting the image data on the second surface in the main scanning direction before storing the image in the image memory 205.

FIG. 3 is an explanatory diagram of the storage timing of the pixel data in the image memory 205. When a predetermined time (t1) elapses from the start of transporting the document 101 (t0), the tip of the document 101 reaches the detection position of the document detection sensor 113. When the tip of the document 101 reaches the detection position of the document detection sensor 113, the detection signal indicating the detection result is turned on (transition from Low to Hi in FIG. 3). This detection signal is input to the CPU 203. The CPU 203 starts storing the pixel data in the image memory 205 at a predetermined timing (t2) from when the detection signal of the document detection sensor 113 is turned on until the document 101 reaches the reading position A.

(Edge extraction)
In the present embodiment, the edge extraction unit 206 acquires the pixel data at the same timing (t2 in FIG. 3) when the pixel data is stored in the image memory 205. FIG. 4 is an explanatory diagram of the processing of the edge extraction unit 206.

FIG. 4A shows a part of image data representing an image of a manuscript. The edge extraction unit 206 acquires pixel data from the timing before the document 101 reaches the reading position A. Therefore, the image data includes an image of the original document 101 and an image on the upstream side of the shadow of the tip of the original document 101 (an image of the surface guide plate 116).

The edge extraction unit 206 performs binarization processing on the pixel data constituting the image data by setting a region of 9 pixels of 3 pixels each in the main scanning direction and the sub scanning direction as one block. In the present embodiment, the pixel position in the main scanning direction is n (1 ≦ n ≦ 7486), and the pixel position in the sub scanning direction is m (1 ≦ m ≦ 11998). The brightness value of each pixel is px (x = 0 to 8).

The edge extraction unit 206 calculates the difference between the maximum value pmax and the minimum value pmin of the brightness values of 9 pixels in one block (FIG. 4B).
In the white region of the surface guide plate 116 in which all nine pixels are white pixels as shown at point A in FIG. 4A, all nine pixels are white pixels. Therefore, the difference between the maximum value pmax and the minimum value pmin is small. At the boundary between the surface guide plate 116 and the shadow (gray) at the tip of the document 101 as shown by point B in FIG. 4A, white pixels and gray pixels are mixed in the 9 pixels. Therefore, the difference between the maximum value pmax and the minimum value pmin becomes large.

When the difference between the maximum value pmax and the minimum value pmin is larger than the predetermined threshold value pth, the pixel becomes a pixel (edge pixel) of the edge portion of the document 101. Specifically, the central pixel (pixel of coordinates (n, m)) of one block when the following equation 1 is satisfied is determined to be an edge pixel. The edge extraction unit 206 sequentially performs such a determination process for each pixel included in the image data.
pmax-pmin ≧ pth (Equation 1)

FIG. 4 (c) is image data (binarized data) obtained by binarizing the image data of FIG. 4 (a) with a threshold value pth (14 in the figure). The pixels shown in white are edge pixels satisfying the above equation 1. In this way, the shadow at the tip of the document 101 is determined as an edge.

(Registration information)
The surface registration calculation unit 207 acquires binarized data from the edge extraction unit 206. The surface registration calculation unit 207 calculates the registration information of the original 101 from the binarized data. FIG. 5 is an explanatory diagram of a method of calculating registration information.

Since the pixel data is input to the edge extraction unit 206 from the timing before the document 101 reaches the reading position A, the surface registration calculation unit 207 acquires the binarized data in the range shown by the dotted line in FIG. do. In the present embodiment, the surface registration calculation unit 207 has, for example, the x-coordinate from 0 to 7487 and the y-coordinate among the pixel data in the range from the coordinates (0, 0) to the coordinates (7487, 11999). Acquires binarized data of pixel data from 0 to 3999.

The surface registration calculation unit 207 of the present embodiment acquires the document tip angle θ1, the angle direction (sign), the upper left coordinates (x1, y1) of the image on the first surface (front surface) based on the binarized data. The width W in the main scanning direction of the original image in the binarized data is calculated. The surface registration calculation unit 207 transmits the calculated information to the CPU 203. Specifically, the surface registration calculation unit 207 is based on the shadow represented by the acquired binarized data (the shadow of the side on the front end side of the document and the shadow on the side edge adjacent to the front end side). Then, the width W is calculated. More specifically, the surface registration calculation unit 207 calculates the length from the leftmost shadow to the rightmost shadow in the acquired binarized data as the width W in the main scanning direction. Further, the surface registration calculation unit 207 performs, for example, linear approximation to the pixel data (pixel data corresponding to the shadow of the side on the tip side of the document 101) in the main scanning pixel width described later, thereby performing the document 101. The side on the tip side of is detected. The surface registration calculation unit 207 calculates the document tip angle θ1 and the angle direction (sign) based on the detection result of the side on the tip side of the document 101. Further, the surface registration calculation unit 207 calculates the upper left coordinates (x1, y1) from a plurality of perpendicular lines with respect to a linearly approximated straight line, for example. The surface registration calculation unit 207 sets the intersection of the perpendicular line passing through the position of the pixel corresponding to the shadow of the side edge of the document 101 at the position where the y coordinate is a predetermined value and the linearly approximated straight line on the upper left. Calculated as coordinates (x1, y1). The predetermined value is set at a position where the pixel corresponding to the shadow of the side edge of the document 101 is located and is not affected by damage, bending, or bending at both ends of the document. For example, the predetermined value may be set at the position on the rear end side (y coordinate is between 3000 and 3999) in the binarized data acquired by the surface registration calculation unit 207. In the present embodiment, the sign of the angle is a positive direction (+ direction) in the diagonal direction upward to the right and a negative direction (− direction) in the downward downward direction with respect to the main scanning direction. The document tip angle θ1, the angle direction (sign), the upper left coordinates (x1, y1), and the width W are the registration information of the present embodiment.

(Registration correction)
The CPU 203 transmits the document tip angle θ1, the angle direction, the upper left coordinates (x1, y1), and the width W acquired from the surface registration calculation unit 207 to the registration correction unit 208. The registration correction unit 208 uses the document tip angle θ1, the angle direction, the upper left coordinates (x1, y1), and the width W to rotate the image data so that the document tip angle θ1 is reduced, thereby rotating the image. Correct (reduce) the tilt. That is, the registration correction unit 208 uses the document tip angle θ1, the direction of the angle, the upper left coordinates (x1, y1), and the width W to obtain the image data of the first surface (surface) stored in the image memory 205. Read while correcting the registration. Specifically, in the registration correction, for example, the image data is read from the image memory 205 from the upper left coordinates (x1, y1) along the side of the tip side of the document image (along the direction of the document tip angle θ1). It is done by. The registration correction may be performed by, for example, a general affine transformation or the like.

FIG. 6 is an explanatory diagram of registration correction of the present embodiment. FIG. 6A exemplifies a case where the image data of FIG. 5 is subjected to rotation correction based on the document tip angle θ1 and the sign of the angle. The document tip angle θ1 represents the amount of skew, and the skew correction is performed by rotation correction. FIG. 6B exemplifies a case where the image data after the rotation correction is translated in the main scanning direction and the sub-scanning direction based on the upper left coordinates (x1, y1). By performing the registration correction, the image data of the first surface (front surface) is skewered so that the upper left coordinate becomes the position (0, 0).

(Main scanning pixel width)
The surface registration calculation unit 207 calculates the pixel width in the main scanning direction, which is an effective region (edge detection region) used for calculating the document tip angle θ1 when calculating the document tip angle θ1 from the image on the first surface (front surface). Select (main scanning pixel width).

When at least one of both ends of the front end side of the document 101 in the main scanning direction is damaged or broken, it is difficult to accurately calculate the document tip angle θ1 from the image data including such a region. It should be noted that damage or breakage at both ends of the document may occur when the needle-bound document bundles are separated one by one for transportation by the ADF. Therefore, when calculating the document tip angle θ1, invalid areas are set at both ends in the main scanning direction according to the size (main scanning size) of the document 101 detected by the size detection unit 1020 of the document tray 102. Will be done. The invalid area is a region between a position separated by a predetermined set distance from one end in the main scanning direction of the original image and the other end, and a position separated from the other end by a predetermined set distance and the other end. The area between. The surface registration calculation unit 207 determines one end and the other end using image data. The invalid area is an area in the main scanning direction of pixel data that is not used for determining the amount of inclination corresponding to the inclination angle. When the invalid area is a fixed value, the main scanning pixel width when calculating the document tip angle θ1 from the image data of a small-sized document such as a business card is shorter than that of a normal-sized document. This causes a decrease in accuracy of the document tip angle θ1. By setting the invalid area according to the main scanning size, a sufficient main scanning pixel width is secured even for a small-sized document, and the accuracy of the document tip angle θ1 is prevented from deteriorating. Generally, a small-sized sheet such as a business card is less likely to be needle-bound than an A4-sized document. In the present embodiment, the invalid area when the main scanning size is equal to or larger than the predetermined size is set to be larger than the invalid area when the main scanning size is smaller than the predetermined size.

FIG. 7 is an explanatory diagram of the main scanning pixel width. In FIG. 7, the main scanning size detected on the document tray 102 is the main scanning size A, the invalid areas set at both ends in the main scanning direction are invalid areas B, and the pixels obtained by removing the invalid area B from the main scanning size A. Is the effective pixel width C. FIG. 7A exemplifies the main scanning pixel width (edge detection region) of an A4 size document. FIG. 7B exemplifies the main scanning pixel width (edge detection area) of an A6 size document. FIG. 7C exemplifies the main scanning pixel width (edge detection area) of a business card-sized document.

The invalid area B is variable according to the document size (main scanning size). In the present embodiment, when the main scanning size A is the predetermined number of pixels or more, here 1457 pixels or more, the invalid area B is 472 pixels, and when the main scanning size A is smaller than 1457 pixels, the invalid area B is 307 pixels. do. When the original is separated from the stapled state, the corners are often broken or damaged. In the present embodiment, the range in which such breakage or breakage may occur is 472 pixels, and the invalid area B is 472 pixels. Further, it is considered that a document having a main scanning size smaller than 1457 pixels is less frequently stapled, and the range in which breakage or breakage may occur is smaller than that of a document having 1457 pixels or more. Therefore, the invalid area B of the document whose main scanning size is smaller than 1457 pixels is set to 307 pixels. The invalid area B of the document whose main scanning size is smaller than 1457 pixels may be set to 0 pixels. That is, the invalid area when the main scan size is smaller than 1457 pixels is smaller than the invalid area when the main scan size is 1457 pixels or more.

The effective pixel width C is calculated by the following equation (2).
Effective pixel width C = Main scan size A − Invalid area B × 2 (Equation 2)

As shown in FIG. 7A, the A4 size document has a main scanning size of 4961 pixels (210 [mm]). Therefore, the invalid area B is 472 pixels. The effective pixel width C is 4017 pixels as shown below.
Effective pixel width C = 4961 pixels-472 pixels x 2 = 4017 pixels

The surface registration calculation unit 207 selects the main scanning pixel width used for calculating the document tip angle θ1 from a plurality of types of pixel width candidates prepared in advance, based on the effective pixel width C. In the present embodiment, the main scanning pixel width is selected from three types of pixel width candidates of 512 pixels, 1024 pixels, and 2048 pixels. Three types of pixel width candidates are prepared according to the document size that can be conveyed by the ADF 100. The surface registration calculation unit 207 compares the effective pixel width C with a plurality of types of pixel width candidates prepared in advance, and sets the largest pixel width candidate having an effective pixel width C or less as the main scanning pixel width. Since the effective pixel width C of the A4 size document is 4017 pixels, the main scanning pixel width used for calculating the document tip angle θ1 is 2048 pixels.

As shown in FIG. 7B, the A6 size document has a main scanning size of 2480 pixels (105 [mm]). Therefore, the invalid area B is 472 pixels. The effective pixel width C is 1536 pixels as shown below. Since the effective pixel width C is 1536 pixels, the main scanning pixel width used for calculating the document tip angle θ1 is 1024 pixels.
Effective pixel width C = 2480 pixels-472 pixels x 2 = 1536 pixels

As shown in FIG. 7C, a business card-sized document has a main scanning size of 1133 pixels (48 [mm]). Therefore, the invalid area B is 307 pixels. The effective pixel width C is 519 pixels as shown below. Since the effective pixel width C is 519 pixels, the main scanning pixel width used for calculating the document tip angle θ1 is 512 pixels.
Effective pixel width C = 1133 pixels-307 pixels x 2 = 519 pixels

The surface registration calculation unit 207 performs, for example, linear approximation to the pixel data (pixel data corresponding to the shadow of the side on the tip side of the document 101) in the determined main scanning pixel width. Specifically, the surface registration calculation unit 207 determines the position of the pixel in the main scanning direction used for linear approximation as follows. An example is taken when the main scanning pixel width is 512 pixels. In this case, the position of the central pixel in the main scanning direction of the width W coincides with the position of the 256th (or 257th) pixel in the main scanning direction from the leftmost pixel of the determined main scanning pixel width. The position of the pixels in the main scanning direction used for linear approximation is determined. The surface registration calculation unit 207 linearly approximates the pixel data (pixel data corresponding to the shadow of the side on the tip side of the document 101) at the determined position in the main scanning direction to the tip side of the document 101. Detects the side of. The surface registration calculation unit 207 calculates the document tip angle θ1 and the angle direction (sign) based on the detection result of the side on the tip side of the document 101. The surface registration calculation unit 207 calculates the upper left coordinates (x1, y1) as follows. For example, among a plurality of perpendicular lines with respect to a linearly approximated straight line, a perpendicular line passing through the position of a pixel corresponding to the shadow of the side edge of the document 101 at a position where the y coordinate is a predetermined value, and the linearly approximated straight line. The intersection with is calculated as the upper left coordinates (x1, y1).

In the present embodiment, the largest pixel width candidate having an effective pixel width C or less is set as the main scanning pixel width, but this is not the case. For example, the effective pixel width C may be set to the main scanning pixel width.

(Slanting correction processing)
8 and 9 are flowcharts showing an image reading process including the skew process of the present embodiment. This process is performed by the controller 200 in response to an image reading instruction being input from the operation unit 202 to the controller 200.

The user regulates the document 101 placed on the document tray 102 in the main scanning direction by the two side regulation plates 123. As a result, the size detection unit 1020 of the document tray 102 detects the main scan size A of the document 101. The CPU 203 acquires information representing the main scanning size A of the document 101 from the size detection unit 1020 (S801).

The CPU 203 sets the invalid area B based on the main scanning size A by the surface registration calculation unit 207. If the main scanning size A is 1457 pixels or more (S802: Y), the surface registration calculation unit 207 sets the invalid region B to 472 pixels (S803). If the main scanning size A is smaller than 1457 pixels (S802: N), the surface registration calculation unit 207 sets the invalid region B to 307 pixels (S804).

The CPU 203 in which the invalid area B is set calculates the effective pixel width C by the surface registration calculation unit 207 (S805). For example, when the main scanning size A is 4961 pixels (210 [mm]), the invalid area B is set to 472 pixels, and the effective pixel width C is 4017 pixels. When the main scanning size A is 2480 pixels (105 [mm]), the invalid area B is set to 472 pixels, and the effective pixel width C is 1536 pixels. When the main scanning size A is 1133 pixels (48 [mm]), the invalid area B is set to 307 pixels, and the effective pixel width C is 519 pixels.

The surface registration calculation unit 207 sets the main scanning pixel width used for calculating the document tip angle θ1 based on the effective pixel width C (S8060). The surface registration calculation unit 207 compares the effective pixel width C with a plurality of types of pixel width candidates (2048 pixels, 1024 pixels, 512 pixels), and selects the largest pixel width candidate having an effective pixel width C or less as the main scanning pixel width. Set to. FIG. 9 is a flowchart showing the processing of S8060 executed by the CPU 203.

If the effective pixel width C is 2048 pixels or more (S806: Y), the surface registration calculation unit 207 sets the main scanning pixel width to 2048 pixels (S807).
If the effective pixel width C is smaller than 2048 pixels (S806: N) and 1024 pixels or more (S808: Y), the surface registration calculation unit 207 sets the main scanning pixel width to 1024 pixels (S809).
If the effective pixel width C is smaller than 1024 pixels (S808: N) and 512 pixels or more (S810: Y), the surface registration calculation unit 207 sets the main scanning pixel width to 512 pixels (S811).
If the effective pixel width C is smaller than 512 pixels, which is the smallest pixel width candidate (S810: N), the surface registration calculation unit 207 does not set the main scanning pixel width. This is because the main scanning pixel width for accurately calculating the document tip angle θ1 cannot be secured. In this case, the document tip angle θ1 is set to 0 ° (S812). Further, the CPU 203 displays information indicating that the registration correction has not been performed on the display unit of the operation unit 202 to notify the user.

After setting the main scanning pixel width, the CPU 203 drives and controls the transport motor 201 to start transporting the documents 101 placed on the document tray 102 one by one (S813). After the document transfer is started, the CPU 203 waits until the document detection sensor 113 detects the document 101 (814: N). When the document detection sensor 113 detects the document 101 (S814: Y), the CPU 203 starts scanning the image of the document 101 by the first scanning unit 109A (S815).

When a predetermined time has elapsed from the start of the reading process, in this case (t2-t1) (S816), the CPU 203 starts storing the image data acquired from the first reading unit 109A in the image memory 205 ( S817). At this time, the image data is also transmitted to the edge extraction unit 206. The edge extraction unit 206 starts generating binarized data based on the acquired image data. Further, the surface registration calculation unit 207 performs registration calculation to calculate the original tip angle θ1, the angle code, the upper left coordinates (X1, Y1), and the width W. At this time, the surface registration calculation unit 207 calculates the document tip angle θ1 based on the pixels having the main scanning pixel width set in the process of S8060. The surface registration calculation unit 207 transmits the registration calculation result to the CPU 203 (S818).

The CPU 203 reads the image data from the image memory 205 according to the result of the registration calculation by the registration correction unit 208 and outputs the image data (S819: N). As a result, skew correction is performed on the image data. When the output of the image data from the image memory 205 by the registration correction unit 208 is completed, the image reading of one original document 101 is completed (S819: Y). When the image scanning of one document 101 is completed, the CPU 203 determines whether or not the next document is placed on the document tray 102 (S820). When the next manuscript is placed (S820: Y), the CPU 203 repeatedly executes the processes after S801 until the next manuscript is exhausted. When the next document is not placed (S820: N), the CPU 203 ends the image reading process.

The image reading device 10 of the present embodiment as described above performs skew correction by rotating and position-correcting the image data. Since the main scanning pixel width for calculating the skew amount is set according to the size of the document, highly accurate skew correction is possible even for a small size document such as a business card.

In the present embodiment described above, the invalid region B has two types, 472 pixels and 307 pixels. This is an example of the invalid area B, and the invalid area B may be set from more options. For example, options for the invalid area B may be prepared for each size of the document that can be conveyed by the ADF 100. As for the main scanning pixel width, three or more options of 2048 pixels, 1024 pixels, and 512 pixels may be prepared. For example, options for the main scanning pixel width may be prepared for each size of the document that can be conveyed by the ADF 100.

Although the embodiments of the present invention have been described above, the present invention is not limited to the above-described embodiments. Although the skew correction based on the reading result of the first reading unit 109A has been described above, the skew correction can be similarly performed on the reading result of the second reading unit 109B. The effects described in the embodiments of the present invention merely list the most preferable effects arising from the present invention, and the effects according to the present invention are not limited to those described in the embodiments of the present invention.

In the present embodiment, the main scanning size A is detected based on the detection result of the size detection unit 1020 of the document tray 102, but this is not the case. For example, the calculated width W may be used as the main scanning size A.

-Second embodiment (configuration of image forming apparatus)
In the following description, the description of the portion where the configuration of the image reading device is the same as that of the first embodiment will be omitted.
FIG. 10 is a configuration diagram of the image reading device of the present embodiment. The image reading device 1000 includes an ADF 100 that conveys a sheet-shaped document 101 to be read, and a reader 30 that reads an image of the document 101 conveyed by the ADF 100. The ADF 100 is rotatably attached to the housing of the reader 30 by a hinge (not shown). The housing of the reader 30 supports the platen glass 31 and the scanning glass 108. The ADF 100 can be opened and closed with respect to the platen glass 31 and the scanning glass 108 by rotating. The ADF 100 includes a document tray 102 capable of loading one or more documents 101, a transport path for transporting the documents 101, and an ejection tray 121 for ejecting the scanned document 101.

The document tray 102 is provided with a side regulating plate 123 that can move in the main scanning direction (width direction) orthogonal to the transport direction of the document 101. The side regulating plate 123 abuts on the end of the document 101 in the width direction to regulate the position of the document 101 in the width direction. In the present embodiment, a configuration in which two side regulation plates 123 are provided in order to regulate both ends in the width direction of the document 101 will be described. However, one side regulation plate 123 is provided in the width direction of the document 101. It may be configured to regulate only one end. When there is one side regulating plate 123, the other end in the width direction of the document 101 is regulated by the fixed regulating plate.

The two side regulation plates 123 are interlocked with each other by moving one of them by an interlocking mechanism (not shown) provided inside the document tray 102, so that the other moves. In the present embodiment, the transport center of the document 101 is the center in the width direction. The two side control plates 123 are configured to approach or separate from the center in the width direction. Therefore, the transport center of the document 101 is at the same position as the center in the width direction regardless of the size of the document 101.

The ADF 100 has a pickup roller 103, a separation roller 2, a separation pad 21, a pull-out roller 3, a transfer roller 4, a lead roller 5, a lead roller 7, and a paper ejection roller 120 in this order from the upstream side in the transport direction of the document 101. To be equipped.

The document tray 102 is provided with a document presence / absence detection sensor 14 for detecting the presence / absence of the document 101. The pickup roller 103 is provided so as to be swingable, and when the document 101 is fed from the document tray 102, the pickup roller 103 falls on the surface of the uppermost document 101 and rotates. When paper feeding is not performed, the pickup roller 103 retracts upward so as not to interfere with the placement of the document 101 on the document tray 102. The pickup roller 103 drops on the surface of the document 101 and rotates to take in the uppermost document 101 of the document tray 102 and convey it to the separation roller 2. The separation roller 2 separates the document 101 into one sheet by a separation nip portion formed between the separation roller 2 and the separation pad 21. Such separation of the original document 101 is a known separation technique.

The document 101 separated into one sheet by the separation roller 2 and the separation pad 21 is conveyed to the transfer roller 4 by the extraction roller 3. The lead roller 5 conveys the document 101 conveyed by the transfer roller 4 to the reading position (upper part of the scanning glass 108) of the first reading unit 402.

The scanning glass 108 is provided at a reading position of the first reading unit 402 when reading an image of a document conveyed by the ADF 100. A white surface guide plate 6 which is an opposing member is provided at a position (opposite side) facing the first reading unit 402 with the scanning glass 108 interposed therebetween. A surface reading upstream roller 51 is provided on the upstream side of the surface guide plate 6, and a surface reading downstream roller 52 is provided on the downstream side. The document 101 is conveyed between the scanning glass 108 and the surface guide plate 6 by the surface reading upstream roller 51. The first scanning unit 402 reads an image of the first surface (front surface) from the document 101 passing between the scanning glass 108 and the surface guide plate 6 via the scanning glass 108.

The document 101 that has passed the reading position of the first reading unit 402 is conveyed to the reading position of the second reading unit 502 by the surface reading downstream roller 52. A lead roller 7 is provided between the reading position of the first reading unit 402 and the reading position of the second reading unit 502. A white back surface guide plate 8 which is an opposing member is provided at a position (opposite side) facing the second reading unit 502 across the back surface scanning glass 118. A back surface reading upstream roller 53 is provided on the upstream side of the back surface guide plate 8, and a back surface reading downstream roller 54 is provided on the downstream side. The document 101 is conveyed between the back surface scanning glass 118 and the back surface guide plate 8 by the back surface reading upstream roller 53. The second scanning unit 502 reads the image of the second surface (back surface) from the document 101 passing between the back surface scanning glass 118 and the back surface guide plate 8 through the back surface scanning glass 118.

The document 101 that has passed the reading position of the second reading unit 502 is conveyed by the back surface reading downstream roller 54 to the paper ejection roller 120 provided on the downstream side. The paper ejection roller 120 ejects the document 101 to the ejection tray 121. When reading the original image on one side of the original 101, the first scanning unit 402 reads the original image on the first side of the original 101, and the second scanning unit 502 does not operate. When reading the original image on both sides of the original 101, the first scanning unit 402 reads the original image on the first side of the original 101, and the second scanning unit 502 reads the original image on the second side of the original 101.
The process of reading an image while transporting the document 101 using the ADF 100 as described above is called "scanning".

The original 101 is placed on the platen glass 31 with the image forming surface facing the platen glass 31 side. In this case, the first reading unit 402 reads the image of the document placed on the platen glass 31 line by line while moving at a constant speed in the direction of arrow C. The process of reading the image of the original on the platen glass 31 is called "fixed reading".

The first reading unit 402 includes LEDs 403a and 403b, reflection mirrors 404a to 404c, and a reading sensor 408, which are light sources that emit light. The LEDs 403a and 403b irradiate the document 101 passing over the scanning glass 108 with light. The reflected light of the irradiated light by the document 101 is reflected by the reflection mirrors 404a to 404c and received by the reading sensor 408. The reading sensor 408 is configured by arranging a plurality of photoelectric conversion elements, which are light receiving elements, in a direction orthogonal to the arrow C direction. The reading sensor 408 generates image data by performing a predetermined process on an electric signal obtained by photoelectrically converting the reflected light received by a plurality of photoelectric conversion elements. This image data represents an image of the first surface of the document 101. Image data is generated each time the image is read line by line. The direction in which the photoelectric conversion elements are arranged is the main scanning direction. The arrow C direction is the sub-scanning direction.

The second scanning unit 502 has the same configuration as the first scanning unit 402, and reads the document image on the second surface of the document 101 in the same operation. That is, the second reading unit 502 includes LEDs 503a and 503b, reflection mirrors 504a to 504c, and a reading sensor 508, which are light sources that emit light. The LEDs 503a and 503b irradiate the document 101 passing between the backside scanning glass 118 and the backside guide plate 8 with light. The reflected light of the irradiated light by the document 101 is reflected by the reflection mirrors 504a to 504c and received by the reading sensor 508. The reading sensor 508 generates image data by performing a predetermined process on an electric signal obtained by photoelectrically converting the reflected light received by a plurality of photoelectric conversion elements which are light receiving elements. Image data is generated each time the image is read line by line. This image data represents an image of the second surface of the document 101.

The white reference plate 122 is a member used during the shading process and serves as a white reference. The shading process using the white reference plate 122 is performed before starting the reading of the image in both the fixed reading and the scanning.

The separation sensor 15, the extraction sensor 16, the document width detection sensor 13, the lead sensor 17, and the paper ejection sensor 18 are provided in the transport path in this order from the upstream side in the transport direction of the document 101. The start timing of the reading operation by the first reading unit 402 and the second reading unit 502 is determined according to the detection result of the read sensor 17. A plurality of document width detection sensors 13 are arranged side by side at different positions in the main scanning direction orthogonal to the transport direction. According to the detection result of the document width detection sensor 13, the length (document width) of the document 101 being conveyed on the transport path in the main scanning direction is detected.

FIG. 11 is an upper cross-sectional view of the ADF 100. In FIG. 11, the A4 size document 101 is placed on the document tray 102. The document tray 102 is provided with document length detection sensors 810 and 811 for detecting the length (document length) of the placed document 101 in the transport direction (sub-scanning direction). By detecting the presence or absence of the document 101 placed on the document tray 102 by each of the document length detection sensors 810 and 811, the approximate size of the document length of the document 101 is detected. Further, the document 101 is regulated in the width direction by the two side regulation plates 123. The distance between the two side regulation plates 123 is the document width Ls of the document 101.

As described above, a plurality of document width detection sensors 13 (document width detection sensors 13a, 13b, 13c, 13d) are arranged at different positions in the main scanning direction. The document width of the document 101 being conveyed can be measured by the detection state of the document 101 by the document width detection sensors 13a, 13b, 13c, and 13d. For example, when the document width detection sensors 13a and 13b detect the document 101 and the document width detection sensors 13c and 13d do not detect the document 101, the document width of the document 101 is determined as follows. That is, the document width of the document 101 is at least twice the distance from the center of transport of the document 101 to the position of the document width detection sensor 13b and less than twice the distance from the center of transport of the document 101 to the position of the document width detection sensor 13c. Judged.

(Control system)
FIG. 12 is a configuration diagram of a control system that controls the operation of the image reading device 1000. The control system of the present embodiment includes an image reading controller 310 provided in the image reading device 1000 and an image processing controller 300. For example, when the image reading device 1000 is combined with the image forming device to form a copying machine or a multifunction device, the image processing controller 300 is provided in the image forming device. The image processing controller 300 may be configured to be included in the image reading device 1000.

The operation of the image reading controller 310 is controlled by the CPU 801 and the ROM (Read Only Memory) 802, and the RAM (Random Access Memory) 803. The CPU 801 controls the operation of each part of the image reading device 1000 by executing a computer program stored in the ROM 802. At that time, the RAM 803 provides a work area, and various data are temporarily stored.

A separation motor 805, a extraction motor 806, and a reading motor 807 are connected to the CPU 801 in order to realize a document transport function. The separation motor 805 rotationally drives the pickup roller 103 and the separation roller 2 under the control of the CPU 801. The extraction motor 806 rotates and drives the extraction roller 3 and the transfer roller 4 under the control of the CPU 801. The reading motor 807 rotationally drives the lead roller 5, the front surface reading upstream roller 51, the front surface reading downstream roller 52, the lead roller 7, the back surface reading upstream roller 53, the back surface reading downstream roller 54, and the paper ejection roller 120 under the control of the CPU 801. ..

A document presence / absence detection sensor 14, a separation sensor 15, a pull-out sensor 16, a read sensor 17, and a paper ejection sensor 18 are connected to the CPU 801. In addition, the CPU 801 is connected to a tray width detection sensor 12 that detects the document width Ls of the document 101 placed on the document tray 102, and a document width detection sensor 13. The tray width detection sensor 12 and the document width detection sensor 13 are size detection units in the width direction (main scanning direction) for detecting the document width. The tray width detection sensor 12 detects the document width Ls of the document 101 placed on the document tray 102 by detecting the distance between the two side regulation plates 123.

An optical motor 804 that moves the first reading unit 402 in the direction of arrow C is connected to the CPU 801. The optical motor 804, the separation motor 805, the extraction motor 806, and the reading motor 807 are pulse motors. The CPU 801 manages the rotation speed of each motor by controlling the operation of these motors while counting the number of drive pulses.

For example, the CPU 801 counts the number of drive pulses of the separation motor 805 from the time when the extraction sensor 16 detects (on) the document 101 to the time when the document 101 is not detected (off) during document transfer. The amount of transport of the document 101 is measured by the number of driving pulses of the separation motor 805 and the amount of advance (gear ratio) per pulse of the gear that transmits the driving force of the separation motor 805 to the separation roller 2. Therefore, the CPU 801 can measure the document length of the document 101 being conveyed by the count value of the number of drive pulses of the separation motor 805. Further, the CPU 801 counts the number of drive pulses of the separation motor 805 from the state in which the extraction sensor 16 does not detect the document 101 (off) to the state in which the document 101 is detected (on) during document transfer. .. As a result, the CPU 801 can measure the interval (between papers) of the documents 101 that are continuously conveyed.

The CPU 801 can determine the document length of the document 101 placed on the document tray 102 from the detection results of the document length detection sensors 810 and 811. For example, the distance from the tip position of the document 101 placed on the document tray 102 to the document length detection sensor 810 is 220 [mm], and the distance from the tip position of the document 101 to the document length detection sensor 811 is 330 [. mm]. When the document length detection sensors 810 and 811 do not detect the document (off), the CPU 801 determines that the document length of the document 101 is less than 220 [mm]. When the document length detection sensor 810 detects the document (on) and the document length detection sensor 811 does not detect the document (off), the CPU 801 sets the document length of the document 101 to 220 [mm] or more and less than 330 [mm]. Judge. When both the document length detection sensors 810 and 811 detect (ON) the document, the CPU 801 determines that the document length of the document 101 is 330 [mm] or more.

As described above, the CPU 801 can measure the document length of the document 101 being conveyed and the document length of the document 101 placed on the document tray 102. Further, the CPU 801 can measure the document width by the width detecting unit. The CPU 801 can determine the original size of the original to be read based on the measured original length and the measured original width.

A second reading unit 502 is connected to the CPU 801 in order to realize an image reading function. The second reading unit 502 includes a communication control unit 505 in addition to the LEDs 503a and 503b (hereinafter, simply referred to as "LED503") and the reading sensor 508. The CPU 801 controls the light emission of the LED 503 via the communication control unit 505, and causes the reading sensor 508 to output image data. The first reading unit 402 includes a communication control unit 405 in addition to the LEDs 403a and 403b (hereinafter, simply referred to as "LED403") and the reading sensor 408. The CPU 801 controls the light emission of the LED 403 via the communication control unit 405, and causes the reading sensor 408 to output image data.

An image processing unit 808 and an oblique detection unit 809 are connected to the CPU 801 in order to perform processing on the read image. The image data output from the reading sensor 508 is input to the image processing unit 808 via the image communication line 302. The image data output from the reading sensor 408 is input to the image processing unit 808 via the image communication line 303. The image processing unit 808 performs image processing such as skew correction, shading processing, and various filter processing for correcting the inclination of the acquired image data according to the instruction of the CPU 801. That is, the image processing unit 808 has the functions of the front surface shading unit 204A, the back surface shading unit 204B, the image inversion unit 210, the image memory 205, and the registration correction unit 208 in the first embodiment. The image processing unit 808 transmits the processed image data to the image processing controller 300 via the image communication line 305.

The skew detection unit 809 performs skew detection processing for detecting the skew amount of the document 101 (the amount of tilt of the read image with respect to the main scanning direction) based on the image data. That is, the skew detection unit 809 has the functions of the edge extraction unit 206 and the surface registration calculation unit 207 in the first embodiment.

The CPU 801 is connected to the image processing controller 300 by the communication line 301. The CPU 801 processes the vertical synchronization signal, which is the reference of the tip of the image represented by the image data, and the horizontal synchronization signal, which indicates the beginning of the tip of the pixel of one line, via the communication line 301 in accordance with the reading timing of the original image. It is transmitted to the controller 300.

The image processing controller 300 includes a CPU 901, a ROM 902, a RAM 903, an image processing unit 905, and an image memory 906. The operation unit 202 is connected to the image processing controller 300. The CPU 901 controls the overall operation of the image reading device 1000 by executing a computer program stored in the ROM 902. At that time, the RAM 903 provides a work area, and various data are temporarily stored. The CPU 901 can communicate with the image reading controller 310 (CPU801) by the communication line 301.

The image processing unit 905 acquires image data from the image processing unit 808 of the image reading controller 310 via the image communication line 305. The image processing unit 905 performs predetermined image processing such as color determination on the acquired image data, and stores the processed image data in the image memory 906.

The operation unit 202 is configured by combining an input interface that is operated by the user to receive various instructions and input data, and an output interface that provides various information to the user on the screen. The CPU 901 receives instructions and input data from the operation unit 202. The operation unit 202 displays screens such as an input screen, a setting screen, and a status display screen according to the instruction of the CPU 901.

(Image reading process)
The reading process of the original image by the scanning performed by the image reading device 1000 as described above will be described. FIG. 13 is a flowchart showing the image reading process. This process is started when the CPU 901 receives an instruction to start reading an image from the operation unit 202. When the CPU 901 receives the instruction to start reading the image, the CPU 901 instructs the CPU 801 to read the image.

When the CPU 801 receives an instruction from the CPU 901, the optical motor 804 moves the first reading unit 402 to the shading position (S101). The shading position is a position for the first reading unit 402 to read the white reference plate 122. The CPU 801 causes the first reading unit 402 to read the white reference plate 122, and generates a shading coefficient based on the reading result (S102). The CPU 801 uses a shading coefficient to perform a known shading correction when reading a document image.

The CPU 801 that generated the shading coefficient moves the first reading unit 402 directly under the scanning glass 108 (reading position) by the optical motor 804 in order to read the original image from the original 101 conveyed by the ADF 100 (S103). .. The CPU 801 detects the document width Ls of the document 101 placed on the document tray 102 according to the detection result of the tray width detection sensor 12 (S104). After that, the CPU 801 drives the separation motor 805, the extraction motor 806, and the reading motor 807 to start the transfer of the document 101 (S105).

When the CPU 801 starts transporting the document 101, the CPU 801 waits until the separation sensor 15 detects (turns on) the document 101 (S106). If the separation sensor 15 does not detect the document 101 within a predetermined time, the CPU 801 may determine that a jam has occurred and stop the transport of the document 101. When the separation sensor 15 detects the document 101, the CPU 801 waits until the extraction sensor 16 detects (turns on) the document 101 (S107). When the extraction sensor 16 detects the document 101, the CPU 801 stops the separation motor 805 and stops the paper feeding (S108). The CPU 801 stops the separation motor 805 to prevent the next document from being continuously fed. Only the separation motor 805 is stopped, and the extraction motor 806 and the reading motor 807 continue to operate. Therefore, the transport of the document 101 that has already been fed is not stopped.

The CPU 801 conveys the document 101 by a predetermined distance, here 50 [mm], from the position where the extraction sensor 16 detects the document 101 (S109). The CPU 801 controls the transport distance of the document 101 by the number of drive pulses of the extraction motor 806. When the CPU 801 has a built-in timer, the transport distance of the document 101 may be controlled by the transport time and the transport speed of the document 101. The document 101 reaches the detection position of the document width detection sensor 13 by being conveyed by 50 [mm]. The predetermined distance at which the document 101 is conveyed is set according to the distance between the position of the document 101 detected by the extraction sensor 16 and the position of the document width detection sensor 13. In the present embodiment, the distance between the extraction sensor 16 and the document width detection sensor 13 is about 30 [mm].

While the document 101 is conveyed by 50 [mm], the CPU 801 detects the document width of the document 101 being conveyed based on the detection result of the document width detection sensor 13 (S110). The document width detection sensors 13a to 13d of the present embodiment are each composed of a flag type sensor, and the state (on, off) of the photo interrupter is switched when the conveyed document 101 turns off the flag. In addition to the flag type sensor, the document width detection sensor 13 may be composed of a photo interrupter whose state (on, off) is switched by blocking the optical path by the conveyed document 101. The CPU 801 stores each on / off state of the document width detection sensors 13a to 13d in the RAM 803 as width information.

For example, when the document 101 is placed horizontally in A4 size, the document width detection sensors 13a to 13d all detect the document 101 and turn on. When the document 101 is placed vertically in A4 size (dotted line in FIG. 11), the document width detection sensors 13a and 13b detect the document 101 and turn it on, and the document width detection sensors 13c and 13d do not detect the document 101. Turns off. In this way, the on / off states of the document width detection sensors 13a to 13d are switched according to the document width of the document 101.

The side regulation plate 123 operates so as to bring the document 101 closer to the center in the main scanning direction. The document width is detected by detecting the presence or absence of the document passing by the document width detection sensors 13a, 13b, 13c, and 13d on the premise that the document 101 is abutted against the side regulation plate 123 on the back side when viewed from the user. NS. Further, the document width detection sensors 13a to 13d may be arranged on both sides instead of one side in the main scanning direction.

The CPU 801 that detects the document width of the document 101 being conveyed sets an edge detection area, which is an area in the main scanning direction of the image data used when the skew detection unit 809 detects the tip edge (S111). FIG. 14 is a flowchart showing the setting process of the edge detection area.

The CPU 801 refers to the width information stored in the RAM 803 by the process of S110, and determines whether or not the document width detection sensors 13a, 13b, 13c, and 13d are all in the ON state in which the document 101 is detected (S201). When the document width detection sensors 13a, 13b, 13c, and 13d are all on (S201: Y), the CPU 801 sets the edge detection area to 5000 pixels (S202). When the document width detection sensors 13a to 13d are all on, it is estimated that the document width of the document 101 is large. Therefore, the CPU 801 sets an edge detection area for a large size. Specifically, for example, the CPU 801 sets an area of 2500 pixels to the left and 2500 pixels to the right from the central position in the main scanning direction of the width W as an edge detection area.

When all of the document width detection sensors 13a, 13b, 13c, and 13d are not on (S201: Y), the CPU 801 has the document width detection sensors 13a and 13b on and the document width detection sensors 13c and 13d off. It is determined whether or not there is (S203). When the document width detection sensors 13a and 13b are on and the document width detection sensors 13c and 13d are off (S203: Y), the CPU 801 sets the edge detection area to 2500 pixels (S204). When the document width detection sensors 13a and 13b are on and the document width detection sensors 13c and 13d are off, it is estimated that the document width of the document 101 is medium. Therefore, the CPU 801 sets an edge detection area for a medium size. Specifically, for example, the CPU 801 sets a region of 1250 pixels to the left and 1250 pixels to the right from the central position in the main scanning direction of the width W as an edge detection region.

When the document width detection sensors 13a and 13b are on and the document width detection sensors 13c and 13d are not off (S203: N), the CPU 801 sets the edge detection area to 500 pixels (S205). When the document width detection sensors 13a and 13b are not on and the document width detection sensors 13c and 13d are not off, it is estimated that the document width of the document 101 is small. Therefore, the CPU 801 sets an edge detection area for a small size. Specifically, for example, the CPU 801 sets a region of 250 pixels to the left and 250 pixels to the right from the central position in the main scanning direction of the width W as an edge detection region.

As described above, the CPU 801 sets the edge detection area to be larger as the document width is larger. The edge detection regions of 5000 pixels, 2500 pixels, and 500 pixels are examples, and the number of pixels may be different as long as the magnitude relationship is not reversed. Further, in the present embodiment, the edge detection region is classified into three stages, but it does not have to be three stages. Further, the edge detection region may be set to match the document width detected on the transport path and the document width detected on the document tray 102.

The CPU 801 that has set the edge detection area waits until the read sensor 17 detects (turns on) the document 101 (S112). The CPU 801 starts reading the image of the first surface of the document 101 by the first reading unit 402 according to the timing when the read sensor 17 is turned on (S113). The CPU 801 may count the number of drive pulses of the reading motor 807 and provide a waiting time until the start of reading the image between the time when the read sensor 17 is turned on and the start of reading the image.

With respect to the image data output from the first reading unit 402, the CPU 801 performs the tip edge detection process and the skew of the document based on the image data in the edge detection area set in the process of S111 by the skew detection unit 809. The row amount detection process is started (S114). When the skew detection unit 809 detects the skew amount, it transmits an interrupt signal to the CPU 801. When the CPU 801 acquires an interrupt signal from the skew detection unit 809, the CPU 801 completes the edge detection process by the skew detection unit 809 (S115).

When the CPU 801 finishes reading the document 101 for one page (S116), the image processing unit 808 performs the above image processing on the image data for one page. At this time, the image processing unit 808 performs skew correction (registration correction) based on the skew amount detected by the skew detection unit 809. As a result, the inclination generated in the read image due to the skew of the document 101 is corrected. When the CPU 801 receives an image transmission request from the CPU 901 via the communication line 301, the image processing unit 808 transmits the image processed image data to the image processing controller 300 via the image communication line 305.

The CPU 801 that has finished scanning determines the presence / absence of the next document in the document tray 102 based on the detection result of the document presence / absence detection sensor 14 (S117). When the next document is on the document tray 102 (S117: Y), the CPU 801 repeats the processes after S105 until there are no documents in the document tray 102. When the next document is not on the document tray 102 (S117: N), the CPU 801 waits until the document 101 is ejected (S118). The ejection of the document 101 is determined based on the detection result of the paper ejection sensor 18.

When the document 101 is ejected, the CPU 801 stops driving the separation motor 805, the extraction motor 806, and the reading motor 807 (S119). After that, the CPU 801 drives the optical motor 804 to move the first reading unit 402 to a predetermined standby position (S120). The standby position of the first reading unit 402 is, for example, a shading position. As a result, when the next image reading process is instructed, the total processing time is shortened. With the above, the image reading process is completed.

As described above, the image reading device 1000 of the present embodiment sets the edge detection region based on the size of the document 101 in the main scanning direction. As a result, skew detection is realized in an appropriate region according to the size of the document 101 in the main scanning direction, and the skew amount can be detected with high accuracy. That is, it is possible to suppress a decrease in the accuracy of reading the conveyed document and detecting the skew amount of the document. Therefore, it is possible to perform skew correction at the time of scanning with high accuracy not only for standard size manuscripts 101 such as A4 size and LTR size but also for manuscripts 101 of various sizes such as business cards, slips, and checks. It will be possible.

A description of a portion in which the configuration of the image reading device of the third embodiment is the same as that of the second embodiment will be omitted. In the second embodiment, the edge detection region is set based on the document width detected by the document width detection sensors 13a, 13b, 13c, and 13d during the transport of the document 101. In the present embodiment, the edge detection area is set based on the detection result of the tray width detection sensor 12.

(Image reading process)
FIG. 15 is a flowchart showing an image reading process by scanning. This process is started when the CPU 901 receives an instruction to start reading an image from the operation unit 202. When the CPU 901 receives the instruction to start reading the image, the CPU 901 instructs the CPU 801 to read the image.

The process of feeding the original 101 after generating the shading coefficient and stopping the feeding by detecting the original 101 by the extraction sensor 16 is the same as the processes of S101 to S108 of FIG. 13 (S301 to S308). ). The CPU 801 sets the edge detection area when the paper feeding is stopped (S309). FIG. 16 is a flowchart showing the setting process of the edge detection area.

The CPU 801 determines whether or not the document width Ls detected in the process of S304 is larger than the first predetermined width, here 210 [mm] (S401). When the document width Ls is larger than 210 [mm] (S401: Y), the CPU 801 sets the edge detection region to 5000 pixels (S402). When the document width Ls is larger than 210 [mm], the CPU 801 sets an edge detection area for a large size.

When the document width Ls is not larger than 210 [mm] (S401: Y), the CPU 801 determines whether the document width Ls is larger than the first predetermined width, that is, 105 [mm]. (S403). When the document width Ls is larger than 105 [mm] (S403: Y), the CPU 801 sets the edge detection area to 2500 pixels (S404). When the original width Ls is larger than 105 [mm] and smaller than 210 [mm], it is estimated that the original width Ls of the original 101 is medium. Therefore, the CPU 801 sets an edge detection area for a medium size.

When the document width Ls is not larger than 105 [mm] (S403: N), the CPU 801 sets the edge detection area to 500 pixels (S405). When the document width Ls is smaller than 105 [mm], the CPU 801 sets an edge detection area for a small size.

As described above, the CPU 801 sets the edge detection area to be larger as the document width is larger. The edge detection regions of 5000 pixels, 2500 pixels, and 500 pixels are examples, and the number of pixels may be different as long as the magnitude relationship is not reversed. Further, in the present embodiment, the edge detection region is classified into three stages, but it does not have to be three stages.

The CPU 801 for which the edge detection area is set reads the original image until there are no more originals placed on the original tray 102, corrects the skew of the read image, and reads the image, as in the processes of S112 to S120 of FIG. The process is terminated (S310-318).

As described above, the image reading device 1000 of the present embodiment sets the edge detection region based on the size of the document 101 placed on the document tray 102 in the main scanning direction. Since the edge detection area is set by the document width Ls that is more accurate than the document width detected from the document 101 being conveyed, skew detection in an appropriate area according to the size of the document 101 in the main scanning direction is realized. , The amount of skew can be detected with higher accuracy. That is, it is possible to suppress a decrease in the accuracy of reading the conveyed document 101 and detecting the skew amount of the document. Therefore, it is possible to perform skew correction at the time of scanning with high accuracy not only for standard size manuscripts 101 such as A4 size and LTR size but also for manuscripts 101 of various sizes such as business cards, slips, and checks. It will be possible.

When the originals 101 are unevenly loaded on the original tray 102, or when originals 101 of different sizes are loaded, the originals 101 are likely to be greatly skewed. In such a case, the image reading device 1000 of the second and third embodiments controls the transfer of the document 101 so as to increase the interval (between the papers) of the documents 101 to be continuously conveyed. As a result, the image reading device 1000 becomes easy to use, preventing the loss of the read image and minimizing the decrease in productivity.

Fourth Embodiment In the following description, the description of the portion where the configuration of the image reading device is the same as that of the first embodiment will be omitted. In the first, second, and third embodiments, the registration information was calculated based on the side on the front end side of the document. In the present embodiment, registration information is calculated based on the side on the side end side adjacent to the side on the front end side of the document. Hereinafter, the configuration of the controller 200 in the first embodiment will be described as an example, but the method for calculating the registration information in the present embodiment may be applied to the control system in the second embodiment.

(Registration information)
The surface registration calculation unit 207 acquired the inclination angle θ2, the angle direction (signature), and the upper left coordinates (x1, y1) with respect to the sub-scanning direction of the image of the first surface (surface) based on the binarized data. The width H in the sub-scanning direction of the original image in the binarized data is calculated. The surface registration calculation unit 207 transmits the calculated information to the CPU 203. Specifically, the surface registration calculation unit 207 is based on the shadow represented by the acquired binarized data (the shadow of the side on the front end side of the document and the shadow on the side edge adjacent to the front end side). Then, the width H is calculated. More specifically, the surface registration calculation unit 207 calculates the length from the shadow at the front end to the shadow at the rear end in the acquired binarized data as the width H in the sub-scanning direction. Further, the surface registration calculation unit 207 performs, for example, linear approximation to the pixel data (pixel data corresponding to the shadow of the side edge of the document 101) in the sub-scanning pixel width described later, thereby performing the document 101. Detects the side edge of. The surface registration calculation unit 207 calculates the angle θ2 and the direction (sign) of the angle based on the side of the side edge of the detected document 101. The surface registration calculation unit 207 calculates the upper left coordinates (x1, y1) as follows. For example, among a plurality of perpendicular lines with respect to a linearly approximated straight line, a perpendicular line passing through the position of a pixel corresponding to the shadow of the side of the tip of the original document 101 at a position where the x coordinate is a predetermined value, and the linearly approximated straight line. The intersection of is calculated as the upper left coordinate (x1, y1). The predetermined value is set at a position where the pixel corresponding to the shadow of the edge of the tip of the document 101 is located and is not affected by damage, bending, or bending at both ends of the document. For example, the predetermined value may be set at the position of the tip (x coordinate is between 3000 and 4500) in the binarized data acquired by the surface registration calculation unit 207.

The registration correction is performed, for example, by reading the image data from the image memory 205 from the upper left coordinates (x1, y1) along the side on the tip end side of the original image (along the direction of the angle θ2). The registration correction may be performed by, for example, a general affine transformation or the like.

The surface registration calculation unit 207 has a pixel width (sub-scanning pixel) in the sub-scanning direction, which is an effective region (edge detection region) used for calculating the angle θ2 when calculating the angle θ2 from the image of the first surface (surface). Width) is selected.
As the sub-scanning pixel width, the method described in FIG. 7 may be applied by substituting the main scanning direction with the sub-scanning direction. The sub-scanning size A (corresponding to the main scanning size A in the first embodiment) is set after the document is detected by a sensor provided in the transport path such as the document detection sensor 113 after detecting the tip of the document. It may be calculated based on the time until the edge is detected and the transport speed of the document. Further, the sub-scanning size A (corresponding to the main scanning size A in the first embodiment) is based on, for example, the size of the document determined based on the detection results of the document length detection sensors 810 and 811 in the second embodiment. May be decided.

Claims (14)

The manuscript tray on which the manuscript is placed and the manuscript tray
A transport means for transporting the document placed on the document tray in the transport direction, and
A reading unit that has a light receiving element that receives light and a generation means that generates pixel data representing the light receiving result of the light receiving element, and reads an image of the document conveyed by the conveying means.
An acquisition means for acquiring information on the length of a document whose image is read by the scanning unit in the main scanning direction orthogonal to the transport direction, and
A first determining means for determining the position of one end and the position of the other end in the main scanning direction of the image of the original document represented by the plurality of pixel data generated by the generating means.
The acquisition means acquires an invalid region, which is a region in the main scanning direction of pixel data that is not used for determining the tilt amount corresponding to the tilt angle of the front end side side of the document in the transport direction with respect to the main scanning direction. Setting means for setting based on the above-mentioned information and
A second determining means for determining the amount of inclination based on pixel data corresponding to the edge on the tip end side of the document included in the region excluding the invalid region in the main scanning direction.
A correction means for correcting the inclination of the image represented by the pixel data is provided so that the amount of inclination determined by the second determination means is reduced.
The invalid region includes a region between a position separated from the one end portion by a set distance and a position of the one end portion in the main scanning direction and a position separated from the other end portion by the set distance in the main scanning direction. The area between the position of the other end and the position of the other end.
The set distance is a distance shorter than the distance between the position of the center and the position of one end of the image of the document represented by the plurality of pixel data in the main scanning direction.
In the setting means, when the length of the document in the main scanning direction is the first length, the length of the invalid region in the main scanning direction is the length of the document in the main scanning direction. The set distance is set so as to be longer than the length of the invalid region in the main scanning direction when the second length is shorter than the length of 1.
Image reader. At the reading position where the scanning unit reads the image of the document, the scanning unit further has an opposing member provided on the side opposite to the scanning unit with respect to the transport path through which the document is conveyed.
The reading unit has a light source that emits light.
The second determining means includes a detecting means for detecting a shadow cast on the facing member by the light emitted from the light source and the transmitted document.
The second determining means is characterized in that the tilt amount is determined based on the pixel data corresponding to the shadow included in the region other than the invalid region in the main scanning direction.
The image reading device according to claim 1. The facing member is characterized in that the side of the reading unit is white.
The image reading device according to claim 2. The pixel data includes a luminance value representing the intensity of light received by the light receiving element.
The detecting means is characterized in that the shadow is detected based on the luminance value.
The image reading device according to claim 2. Further, it has a size detecting means for detecting the length of the document placed on the document tray in the main scanning direction.
The acquisition means is characterized in that the detection result by the size detection means is acquired as the information.
The image reading device according to claim 1. A plurality of sensors are arranged side by side at different positions in the main scanning direction on the transport path through which the document is conveyed, and further have a plurality of sensors for detecting the document.
The acquisition means is characterized in that the detection results of the plurality of sensors are acquired as the information.
The image reading device according to claim 1. The acquisition means is characterized in that the length between the one end portion and the other end portion determined by the first determination means is acquired as the information.
The image reading device according to claim 1. The manuscript tray on which the manuscript is placed and the manuscript tray
A transport means for transporting the document placed on the document tray in the transport direction, and
A light receiving element that receives light, a reading unit that has a means for generating pixel data representing the light receiving result of the light receiving element, and reads an image of the original document conveyed by the conveying means.
An acquisition means for acquiring information on the length of the document read by the scanning unit in the main scanning direction orthogonal to the transport direction, and
A first determining means for determining the position of one end and the position of the other end in the main scanning direction of the image of the original document represented by the plurality of pixel data generated by the generating means.
The acquisition means acquires an invalid region, which is a region in the main scanning direction of pixel data that is not used for determining the tilt amount corresponding to the tilt angle of the front end side side of the document in the transport direction with respect to the main scanning direction. Setting means for setting based on the above-mentioned information and
A second determining means for determining the amount of inclination based on pixel data corresponding to the edge on the tip end side of the document included in the region excluding the invalid region in the main scanning direction.
A correction means for correcting the inclination of the image represented by the pixel data is provided so that the amount of inclination determined by the second determination means is reduced.
The invalid region includes a region between a position separated from the one end portion by a set distance and a position of the one end portion in the main scanning direction and a position separated from the other end portion by the set distance in the main scanning direction. The area between the position of the other end and the position of the other end.
The set distance is a distance shorter than the distance between the position of the center and the position of one end of the image of the document represented by the plurality of pixel data in the main scanning direction.
The setting means sets the invalid area when the length of the document in the main scanning direction is the first length, and the length of the document in the main scanning direction is larger than the first length. When the second length is short, the invalid area is not set.
Image reader. At the reading position where the scanning unit reads the image of the document, the scanning unit further has an opposing member provided on the side opposite to the scanning unit with respect to the transport path through which the document is conveyed.
The reading unit has a light source that emits light.
The second determining means includes a detecting means for detecting a shadow cast on the facing member by the light emitted from the light source and the transmitted document.
The second determining means is characterized in that the tilt amount is determined based on the pixel data corresponding to the shadow included in the region other than the invalid region in the main scanning direction.
The image reading device according to claim 8. The facing member is characterized in that the side of the reading unit is white.
The image reading device according to claim 9. The pixel data includes a luminance value representing the intensity of light received by the light receiving element.
The detecting means is characterized in that the shadow is detected based on the luminance value.
The image reading device according to claim 9. Further, it has a size detecting means for detecting the length of the document placed on the document tray in the main scanning direction.
The acquisition means is characterized in that the detection result of the size detection means is acquired as the information.
The image reading device according to claim 8. A plurality of sensors are arranged side by side at different positions in the main scanning direction on the transport path through which the document is conveyed, and further have a sensor for detecting the document.
The acquisition means is characterized in that the detection result of the sensor is acquired as the information.
The image reading device according to claim 8. The acquisition means is characterized in that the length between the one end portion and the other end portion determined by the first determination means is acquired as the information.
The image reading device according to claim 8.

Images