JP2019134278A - Image reading apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image reading apparatus capable of shortening a skew detection time in accordance with an actual skew amount of a document. An image reading apparatus 1000 detects a skewed amount of a document according to a surface reading sensor 208 that reads one side of a conveyed document and a detection region set in advance for the read image. It includes a skew detection unit (α) 271 and a skew detection unit (β) 272, and a skew correction unit 270 that performs skew correction based on the detection result of the skew amount. The skew detection unit (α) 271 is set with a first detection area length (set to 12°), and the skew detection unit (β) 272 is set with a second detection area length shorter than the first detection area length. Set the detection area length (3° setting). [Selection diagram] Fig. 14

Description

  The present invention relates to a technology of an apparatus for reading an image.

An image forming apparatus such as a copying machine or a facsimile apparatus includes an image reading device such as a scanner that optically reads an image of a document to obtain image data. These image reading apparatuses not only read a document placed on a platen glass but also read a plurality of documents with a single operation so as to be referred to as an automatic document feeder (hereinafter referred to as ADF). ) In many cases.
Further, in an image reading apparatus equipped with an ADF, two image reading units are provided to improve productivity, and the front and back (one side and the other side) of a document are read by one conveyance (one-pass duplex reading). Also known is a configuration that can be called

In an image reading apparatus, when an original is conveyed one by one with an ADF and an original image is read with an image reading unit, the read image is caused by the assembly accuracy of the image reading unit or the ADF, a manufacturing error of a roller used for conveying the original, and the like. May be tilted. Further, even when the document is placed on the ADF paper feed tray in a tilted state, the read image may be tilted.
Regarding the inclination generated in the read image in the image reading apparatus, after the original is read, the skew amount of the original is calculated from the read image, and the inclination of the read image is corrected by image processing based on the calculated skew amount. Technology is known.

For example, in Patent Document 1, an edge is extracted from a read image, and an edge of the original is determined from the extracted edge, thereby specifying an original image area and calculating an inclination of the original.
In Patent Document 2, the image reading apparatus is provided with an image processing circuit for specifying the front end of the original from the read image and calculating the inclination of the original from the specified front end of the document for the front side and the back side.

JP2013-123119A JP 2010-118911 A

In order to improve the productivity of the image reading apparatus, that is, to shorten the time from reading the document to outputting it, the time required to calculate the skew amount (hereinafter referred to as skew detection) from the read image ( Hereinafter, it is desirable to shorten the skew detection time).
In order to detect skew by extracting an edge from a read image as in Patent Document 1, first, it is determined whether or not all pixels of image data to be subjected to skew detection are edges. To extract. Therefore, the edge extraction requires time corresponding to the data amount of the target image data. Therefore, if the image area to be detected for skew (hereinafter referred to as skew detection area) is narrowed, the time required to complete edge extraction can be shortened, and as a result, the skew detection time can be shortened.

FIG. 3 is a diagram for explaining an example of the skew detection region. In FIG. 3A, the skew detection area is an image area including the leading edge of the document. Therefore, in the case of FIG. 3A, edge extraction is performed for the skew detection region, the leading edge of the document is determined from the extracted edge, and the skew amount of the document can be calculated.
On the other hand, there is a case where the amount of skew cannot be calculated due to the skew detection region and the size and inclination of the document whose skew is detected. FIG. 3B illustrates an example in which the skew amount detection region cannot be calculated in the same skew detection region as that in FIG. As in this case, when the left or right (or both) apexes of the front end side of the document are out of the skew detection area in the document transport direction, the front end of the document can be specified from the extracted edge. Therefore, the skew amount cannot be calculated.

  From the above, in order not to calculate the skew amount of the document, it is necessary to appropriately set the skew detection region so that the maximum skew amount that the document can take can be detected. However, in this case, even if the skew amount of the document is small, there is a problem that it takes a uniform skew detection time corresponding to the maximum skew amount that the document can take.

  The main object of the present invention is to provide an image reading apparatus capable of reducing the skew detection time in accordance with the skew amount of the actual document.

  The image reading apparatus of the present invention includes a first reading unit that reads an image on one side of a conveyed document, and an image of the document that is output from the first reading unit, corresponding to a first detection area. Based on the image data of the document, the first detection unit for detecting the skew amount of the document, and the image data of the document image output from the first reading unit corresponding to the second detection area. A second detection unit that detects a skew amount of the document, and an image data of an image of the document based on the skew amount detected by the first detection unit or the second detection unit. A correction unit that performs skew correction, and a control unit that controls the first detection region and the second detection region, and the length of the first detection region in the document transport direction is: Of the second detection area in the document transport direction. Wherein the longer than is.

  According to the present invention, it is possible to shorten the skew detection time according to the actual document skew amount.

1 is a schematic longitudinal sectional view illustrating an example of a configuration of an image reading apparatus including an ADF. (A), (b) is a block diagram which shows an example of a structure of the control part of an image reading apparatus. (A), (b) is a figure for demonstrating an example of a skew detection area | region. The figure which shows an example of transmission / reception (path | route) of the image data in a reader | leader and ADF. (A), (b) is a figure for demonstrating an example of the image input selection function of an image input selection part. (A)-(d) is a figure for demonstrating an example of the method of detecting skew feeding of a document from the read image. (A)-(d) is a figure for demonstrating a skew feeding correction method. (A), (b) is a figure which shows an example of the timing from image reading to image output. (A1)-(a3) is a figure which shows an example of the skew detection area | region according to the maximum value of the skew amount to detect. (B1)-(b3) is a figure which shows an example of the skew detection area | region according to the maximum value of skew amount to detect. (A)-(d) is the figure which looked at the document bundle | stacking mounted in ADF and a document tray from the top. The table | surface which shows an example of the maximum value of the skew feeding amount which arises in the read image of the small sized original in a mixed width mixed job. (A), (b) is a figure showing an example of timing from image reading to image output in a single-sided reading mixed width mixed job. The flowchart which shows an example of the control flow in a non-reading job. The flowchart which shows an example of the control flow in a non-reading job different from FIG. 12 is a flowchart illustrating an example of acquisition control of setting values θΔ and (xΔ, yΔ) for each machine of the image reading apparatus according to the second embodiment. (A)-(d) is a figure for demonstrating the calculation method of setting value (theta) (DELTA). (A)-(d) is a figure for demonstrating the calculation method of setting value (x (DELTA), y (DELTA)). The flowchart which shows an example of the control flow in a non-reading job. The flowchart which shows an example of the control flow in a non-reading job different from FIG. (A1)-(a4) is a figure for demonstrating the positional relationship on a conveyance path about a read sensor and the original in conveyance. (B1)-(b4) is a figure for demonstrating the positional relationship on a conveyance path about a read sensor and the original in conveyance. (B5) and (b6) are diagrams for explaining the positional relationship of the lead sensor and the original document being conveyed on the conveyance path.

  Hereinafter, a case where the driving device according to the present invention is applied to an image reading device will be described as an example with reference to the drawings.

[First Embodiment]
[Configuration example of image reading apparatus and image forming apparatus of this embodiment]
FIG. 1 is a schematic longitudinal sectional view showing an example of the configuration of an image reading apparatus including an ADF.
An image reading apparatus 1000 according to this embodiment includes an image reading unit (hereinafter referred to as a reader) 200 that reads an image of a document, and an automatic document feeding unit (hereinafter referred to as ADF) 100. The image reading apparatus 1000 is connected to a controller 300 (not shown).
Hereinafter, the operation of the ADF 100 will be described with reference to FIG.

  FIG. 2 is a block diagram illustrating an example of the configuration of the control unit of the image reading apparatus 1000. FIG. 2A is a control block diagram of the reader 200 and the ADF 100, and FIG. 2B is a control block diagram of the controller 300.

[Configuration of Reader 200 and ADF 100]
First, the control block diagram of FIG. The CPU (A) 251 is a central processing unit that comprehensively controls each unit of the reader 200 and the ADF 100. The ROM (A) 252 is a storage device that stores control contents to be executed by the CPU (A) 251 as a program. The RAM (A) 253 is a storage device used as a work area necessary for the CPU (A) 251 to perform control.

  The CPU (A) 251 includes an optical motor 226 for moving the front surface LED 203, the front surface reading sensor 208, the back surface LED 103, the back surface reading sensor 108, and the front surface reading unit 202 in the sub scanning direction in order to realize an image reading function. Are connected to each other. Further, the CPU (A) 251 is connected to a glass motor 112 for moving the back-flow-reading glass 101, an image memory (A) 260, an image processing unit (A) 261, and an image transfer unit (A) 255.

The front side reading sensor 208 and the back side reading sensor 108 are sensors that scan an image of a document and read it for each line.
The image memory (A) 260 is a storage device that temporarily stores image data read by the front surface reading sensor 208 and the back surface reading sensor 108.
The image processing unit (A) 261 corrects the read image stored in the image memory (A) 260 by image processing, which will be described later, as necessary.
The image transfer unit (A) 255 transfers the image data subjected to image processing by the image processing unit (A) 261 and a skew feeding correction unit 270 described later to an image memory (B) 406 of the controller 300 described later.

The CPU (A) 251 includes a transport system motor 111 that drives various transport rollers and transport system sensors (sensors 15 and 16 shown in FIG. , 17, 18, 11) are connected to each other.
The CPU (A) 251 is connected to a document detection sensor 14 that determines the presence or absence of documents on the document tray 30.

A backup unit (A) 256 is connected to the CPU (A) 251.
The backup unit (A) 256 is a storage device for storing a part of work data used for control of the reader 200 and the ADF 100, setting values when there is a setting for each machine, and the like.
Further, the CPU (A) 251 includes an image input selection unit 273, a skew detection unit (α) 271, and an image input selection unit 273, a skew detection unit (α) 271, and the like. A skew detection unit (β) 272 and a skew correction unit 270 are connected.
Details of the image input selection function by the image input selection unit 273, the skew detection function by the skew detection unit (α) 271 and the skew detection unit (β) 272, and the skew correction function by the skew correction unit 270 will be described later.

[Configuration of Controller 300]
Next, the control block diagram of FIG. The controller 300 is a device that controls the entire image reading system including the reader 200 and the ADF 100.
The CPU (B) 301 is a central processing unit that comprehensively controls each unit of the controller 300.
A ROM (B) 302 is a storage device that stores control contents to be executed by the CPU (B) 301 as a program.
A RAM (B) 303 is a storage device used as a work area necessary for the CPU (B) 301 to perform control.
The image transfer unit (B) 308 receives the image from the image transfer unit (A) 255 and stores it in the image memory (B) 306. The operation unit 304 is for performing an operation instruction from the user to the entire image reading system, displaying a message to the user, and displaying a read image. The operation unit 304 communicates with the CPU (B) 301 to perform a desired display. Or input.

  The CPU (B) 301 exchanges control commands related to image reading control and exchanges control data via the communication line 401 with the CPU (A) 251. For example, the CPU (B) 301 receives a user image reading start instruction from the operation unit 304, and transmits an image reading start request to the CPU (A) 251. Further, for example, the CPU (B) 301 receives a user document size setting instruction from the operation unit 304 and transmits the document size (main scanning width and sub-scanning length) to the CPU (A) 251.

[Operation of Reader 200 and ADF 100]
Operations of the reader 200 and the ADF 100 will be described with reference to FIG.
During the operation of the image reading apparatus 1000, the CPU (A) 251 monitors the output of the document detection sensor 14 provided on the document tray 30, and whenever there is a change in the output, the document is sent via the communication line 401. The CPU (B) 301 is notified of the output of the detection sensor 14. The CPU (B) 301 determines whether or not the document bundle S is stacked on the document tray 30 based on the notified output of the document detection sensor 14.

  The reader 200 reads a document in response to an image reading start request from the CPU (B) 301. When an image reading start instruction is input from the user via the operation unit 304, the CPU (B) 301 transmits an image reading start request to the CPU (A) 251 via the communication line 401. The image reading start request is either a “fixed reading start request” or a “streaming reading start request”.

The “fixed reading start request” is for reading a document by scanning the surface reading unit 202 at a constant speed in the sub-scanning direction indicated by the arrow in FIG. 1 for the document placed on the document table glass 209. ”Is requested.
On the other hand, the “flow-reading start request” is a “flow-reading” for reading a document while transporting the document by the ADF 100 with the front surface reading unit 202 stopped at a predetermined position for the document bundle S stacked on the document tray 30. Is requested to start.

If a document bundle S is not stacked on the document tray 30 when a document reading start instruction is input from the user, the CPU (B) 301 indicates to the CPU (A) 251 that “ Send “Read Fixed Reading Request”.
On the other hand, when the document reading start instruction is input from the user, if the document bundle S is stacked on the document tray 30, the CPU (B) 301 sends a “flow scanning start request” to the CPU (A) 251. Send.
The reader 200 starts the fixed reading or the flow reading depending on whether the image reading start request from the CPU (B) 301 is a “fixed reading start request” or a “flow scanning start request”.
Hereinafter, the operation of the ADF 100 will be described with reference to FIG.

[Configuration of ADF100]
[Single-side scanning]
Here, a single-sided document reading operation (single-sided scanning reading) using the ADF 100 will be described.
The ADF 100 includes a document tray 30 on which a document bundle S composed of one or more document sheets is stacked, and a separation pad that projects the document bundle S from the document tray 30 and restricts downstream advancement before the document conveyance starts. 21, a separation roller 2, and a paper feed roller 1.
The document tray 30 is provided with a document detection sensor 14 so that the presence or absence of a document on the document tray 30 can be determined.
The paper feed roller 1 rotates by dropping on the original surface of the original bundle S stacked on the original tray 30. As a result, the uppermost document in the document bundle is fed. The original fed by the paper feed roller 1 is separated into one sheet by the action of the separation roller 2 and the separation pad 21. This separation is realized by a known separation technique.

The document separated by the separation roller 2 and the separation pad 21 is conveyed to the conveyance roller 4 by the drawing roller 3. On the downstream side of the transport roller 4, a paper feed path is provided that transports the original that has passed through the transport roller 4 to the surface of the reading glass 201.
The document sent to the paper feed path is conveyed to the surface flow reading position by the surface reading and conveying roller 5, the surface reading upstream roller 51, and the surface reading downstream roller 52. The surface of the document passing between the surface flow reading glass 201 and the surface glass facing member 6 is irradiated by the surface LEDs 203a and 203b. The reflected light is bent by the plurality of mirrors 204a, 204b, and 204c, and the surface reading sensor 208 reads the surface image of the document line by line (surface reading).
The document conveyed by the front surface reading downstream roller 52 is conveyed by the back surface reading and conveying roller 7, the back surface reading upstream roller 53 and the back surface reading downstream roller 54, and then conveyed to the paper discharge tray 13 by the paper discharge roller 12.
When there are a plurality of documents on the document tray 30, the feeding, separation, transport processing, and surface flow from the above-described document bundle S until the reading of the surface of the final document and the discharge to the discharge tray 13 are completed. The single-sided reading process and the paper discharge process at the reading position are repeated.

[Double-sided original scanning]
Here, a double-sided document reading operation (double-sided scanning reading) using the ADF 100 will be described.
Double-sided scanning is obtained by adding a reading operation on the back side of the document to reading the surface of the document in single-sided scanning scanning. Since document feeding, separation, conveyance, and front side reading are the same as those in the above-described single-sided reading, description thereof will be omitted.
The document conveyed by the front surface reading downstream roller 52 is conveyed to the back surface reading position by the back surface reading and conveying roller 7, the back surface reading upstream roller 53, and the back surface reading downstream roller 54. Before the original reaches the reverse flow reading position, the reverse flow reading glass 101 configured to be movable is moved to the position shown in FIG.

The back side of the document passing between the back side scanning glass 101 and the back side glass facing member 8 is irradiated by the back side LEDs 103a and 103b. The reflected light is bent by the plurality of mirrors 104a, 104b, and 104c, and the back side image of the original is read by the back side reading sensor 108 (back side reading). The document conveyed by the back surface reading downstream roller 54 is conveyed to the paper discharge tray 13 by the paper discharge roller 12.
When there are a plurality of documents on the document tray 30, the above-described feeding, separation, and conveyance processing from the document bundle S is completed until the front side reading of the final document, the back side reading, and the discharge to the discharge tray 13 are completed. The front and back reading processes at the front surface reading position and the back surface reading position and the paper discharge process are repeated.

[Transport control]
The rollers used for conveyance rotate using a conveyance system motor 111 (one or a plurality of motors not shown in FIG. 1) as a drive source. The rollers used for conveyance are a paper feed roller 1, a separation roller 2, a drawing roller 3, a conveyance roller 4, a reading upstream roller 5, a surface reading upstream roller 51, a surface reading downstream roller 52, and a back surface reading conveying roller 7. . Further, a back surface reading upstream roller 53, a back surface reading downstream roller 54, and a paper discharge roller 12.

Further, in processing of document feeding, separation, conveyance, reading at the front surface scanning reading position and back surface scanning reading position, and paper discharge processing, sensors 15, 16, 17, 18, 11 detects the presence or absence of the document or the position on the conveyance path, and performs processing according to the detection result.
In particular, the leading edge of the document being conveyed is detected using the lead sensor 18, and reading at the surface flow reading position is started based on the document leading edge detection timing of the lead sensor 18 and the distance from the lead sensor 18 to the surface flow reading position. To control. Similarly, based on the document leading edge detection timing of the lead sensor 18 and the distance from the lead sensor 18 to the backside scanning reading position, the start of reading at the backside scanning reading position is controlled.

[Configuration of Image Reading Unit (Reader) 200]
The operation of the reader 200 will be described with reference to FIG.
The reader 200 moves a surface reading unit 202 configured to be movable by an optical system motor 226 (not shown in FIG. 1) in the sub-scanning direction indicated by an arrow in FIG. 1 for a document placed on the platen glass 209. Scan at a constant speed. Thereby, the image information recorded on the original is optically read line by line (fixed reading).

Further, with respect to the documents stacked on the document tray 30, the surface reading unit 202 is moved so as to come to the center position of the surface glass facing member 6 of the ADF 100, and the surface of the document fed and conveyed by the above-described method is optically changed. Read (single-sided flow reading).
When reading both sides of a document loaded on the document tray 30, the front side (one side) reads an image in the same manner as single-sided scanning, and the back side (the other side) is the back side mounted on the ADF 100 as described above. An image is read by the reading unit 102 (double-sided scanning reading).

[Shading control]
The front shading white plate 210 and the back shading white plate 110 are white plates for creating white level reference data by shading.
Before reading the original, the front shading white plate 210 and the back shading white plate 110 are read by the front reading unit 202 and the back reading unit 102, respectively, and the read results are subjected to image processing to create reference data for each front and back.

  The back surface shading white plate 110 is attached to a back surface reading glass 101 configured to be movable by a glass motor 112 (not shown in FIG. 1). The back side shading white plate 110 moves the back side reading glass 101 so that the back side shading white plate 110 is positioned on the opposite surface of the back side reading unit 102, whereby the back side shading white plate 110 can be read. After reading the back side shading white plate 110, the back side reading glass 101 is moved to the position shown in FIG.

[Scanned image processing]
FIG. 4 is a diagram illustrating an example of image data transmission / reception (path) in the reader 200 and the ADF 100. Image processing of the read image in the reader 200 and the ADF 100 will be described with reference to FIG.

[Select image input]
As shown in FIG. 4, the image data read by the front surface reading sensor 208 and the back surface reading sensor 108 is stored in the image memory (A) 260 as described above, but is also input to the image input selection unit 273 at the same time. The
The image input selection unit 273 selects the front side reading sensor 208 and the back side reading sensor 108, the skew detection unit (α) 271, and the skew detection unit (β) 272 based on information preset by the CPU (A) 251. Selectively connect.

FIG. 5 is a diagram for explaining an example of the image input selection function of the image input selection unit 273. The image input selection function of the image input selection unit 273 will be described with reference to FIG.
FIG. 5A shows a case where an input from the front surface reading sensor 208 is output to the skew detection unit (α) 271 and an input from the back surface reading sensor 108 is output to the skew detection unit (β) 272. ing.
The image input selection unit 273 includes an image input selection unit (α) 273a that selects output to the skew detection unit (α) 271 and an image input selection unit that selects output to the skew detection unit (β) 272 ( β) 273b.

  The image input selection unit (α) 273a and the image input selection unit (β) 273b have the same configuration, and both receive input from one or both of the front surface reading sensor 208 and the back surface reading sensor 108, Either one is selected and output. The selection is performed based on a prior setting from the CPU (A) 251.

  In FIG. 5A, the input from the front side reading sensor 208 is selected in the image input selection unit (α) 273a, and the input from the back side reading sensor 108 is selected in the image input selection unit (β) 273b. . In this case, skew detection is performed by the skew detection unit (α) 271 for the read image on the front surface, and skew detection is performed by the skew detection unit (β) 272 for the read image on the back surface.

  In FIG. 5B, the input from the surface reading sensor 208 is selected in the image input selection unit (α) 273a, and the input from the surface reading sensor 208 is also selected in the image input selection unit (β) 273b. ing. In this case, the skew detection is performed on both the skew detection unit (α) 271 and the skew detection unit (β) 272 for the read image on the front surface.

[Declination detection]
Here, as illustrated in FIG. 4, the skew detection unit (α) 271 and the skew detection unit (β) 272 receive image data input from the image input selection unit 273.
The skew detection unit (α) 271 and the skew detection unit (β) 272 read based on image data of a read image input from the image input selection unit 273 and information set in advance by the CPU (A) 251. Document skew detection is performed from the image. As preset information, there is a skew detection area. Since the skew detection unit (α) 271 and the skew detection unit (β) 272 have the same configuration, the skew detection unit (α) 271 will be described below as an example.

FIG. 6 is a diagram for explaining an example of a method for detecting skew of a document from a read image. A skew detection method will be described with reference to FIG.
FIG. 6A shows a document, and FIG. 6B shows a read image obtained by scanning the surface of the document in FIG.
When the document image is read by the front surface reading unit 202, an area larger than the document size is read in advance as shown in FIG. Specifically, even if skew has occurred up to the upper limit of the amount allowed for the image reading apparatus 1000, the reading area is expanded so that the entire document can be read.
The main scanning width of the reading area is fixed at the maximum width that can be read by the front surface reading sensor 208 and the back surface reading sensor 108. The sub-scanning length of the reading area is determined by the document size, the reading start timing, and the reading end timing.

21, 22, and 23 are diagrams for explaining the positional relationship between the lead sensor 18 and the original document being conveyed on the conveyance path. A method for determining the document reading start timing based on the leading edge of the document detected by the read sensor 18 will be described with reference to the drawings.
FIGS. 21 (a1), (a2), (a3), and (a4) represent cases where no skew has occurred in the document being conveyed. FIGS. 21A1 and 21A3 are cross-sectional views of the ADF 100, and FIGS. 21A2 and 21A4 are views in which a curved conveyance path of the ADF 100 is developed on a plane. The chain lines between (a1) and (a2) and (a3) and (a4) in FIG. 21 indicate the correspondence of positions in the two diagrams, respectively.

As shown in FIGS. 21A2 and 21A4, the lead sensor 18 is arranged at the center position in the main scanning direction in the transport path. When the front end of the document reaches the position of the lead sensor 18 while the document is being conveyed, the output of the lead sensor 18 changes from OFF to ON. Thereby, the CPU (A) 251 determines that “the leading edge of the document has reached the read sensor 18”.
Let L be the distance between the sub-scanning position of the lead sensor 18 and the surface flow reading position. In FIG. 21 (a2), after the leading edge of the document reaches the lead sensor 18, the document is transported by a distance L, and the position of the document when no skew occurs in the transport is as shown in FIG. 21 (a4). Become.
For example, as shown in FIGS. 21A2 and 21A4, if the document is not skewed, the timing at which the document is conveyed by the distance L after the leading edge of the document reaches the lead sensor 18 (at FIG. 21A4). It is assumed that the reading of the surface is started at the time indicated). Even in this case, the entire front end of the document (from the top left corner of the document to the top right corner of the document) can be read.

22 (b1), (b2), (b3), (b4), and FIGS. 23 (b5) and (b6) represent cases where skew is occurring in the document being conveyed.
22 (b1), (b3), and FIG. 23 (b5) are cross-sectional views of the ADF 100, and FIGS. 22 (b2), (b4), and FIG. 23 (b6) are developed views of the curved conveyance path in a plane. is there.
Note that the chain lines between FIGS. 22 (b1) and (b2), (b3) and (b4), and FIGS. 23 (b5) and (b6) indicate the correspondence between positions in the two diagrams.

In FIG. 22 (b2), when the output of the lead sensor 18 changes from OFF to ON and the CPU (A) 251 determines that “the leading edge of the document has reached the lead sensor 18”, the document is skewed. Therefore, the upper left vertex of the document is actually positioned downstream of the sub-scanning position of the read sensor 18.
In FIG. 22 (b2), after the leading edge of the document reaches the lead sensor 18, the document is transported by a distance L, and the position of the document when no new skew occurs in the transport is as shown in FIG. 22 (b4). become. At this time, the center position in the main scanning direction in the transport path of the leading edge of the document, that is, the position detected by the lead sensor 18 in the leading edge of the document (hereinafter referred to as a leading edge detection point) has reached the surface flow reading position. However, the upper left apex of the document is located downstream from the surface sink reading position.
In such a case, when reading of the front surface is started at this timing (the time point shown in FIG. 22 (b4)), the original portion located downstream from the front surface reading position cannot be read and the read image is displayed. Missing will occur.

On the other hand, after the leading edge of the document reaches the lead sensor 18 in FIG. 22 (b2), the document is transported by a distance (L-20.0 [mm]), and the document is not newly skewed during the transport. The position is as shown in FIG. At this time, the tip detection point is located 20.0 [mm] upstream of the surface flow reading position.
Further, the upper left vertex of the document is located upstream of the surface-flow reading position, and if the reading of the surface is started at this timing, the read image is not lost. In the present embodiment, in order to prevent the document image leading edge from being lost, the front surface reading is started at the timing when the document is conveyed by a distance (L-20.0 [mm]) after the leading edge of the document reaches the lead sensor 18. . That is, it is assumed that image reading starts from 20.0 [mm] before the leading edge detection point reaches the surface reading start position.

As with the leading edge, the output of the lead sensor 18 changes from ON to OFF to prevent the trailing edge of the document image from being lost, and the CPU (A) 251 determines that “the trailing edge of the document has reached the lead sensor 18”. Then, the front side scanning is finished at the timing when the document is conveyed by the distance (L + 20.0 [mm]). That is, the image reading is terminated after 20.0 [mm] of the position detected by the lead sensor 18 in the trailing edge of the document (hereinafter referred to as a trailing edge detection point) reaches the front surface reading start position.
Also for the back side scanning, similarly to the front side scanning, the reading start timing and the reading end timing are determined so that the read image is not lost. Image reading is started from 20.0 [mm] before the leading edge detection point reaches the back surface reading start position, and image reading is terminated after 20.0 [mm] when the trailing edge detection point reaches the back surface reading start position.

The skew detection unit (α) 271 performs skew detection on the read image thus obtained. When a document is read, a shadow is generated outside the document edge due to the thickness of the document itself.
In the present embodiment, the skew detection unit (α) 271 detects this shadow, determines the leading edge of the document, and calculates the skew amount. Specifically, after edge extraction is performed on the image in the skew detection area of the read image, successive edges are extracted to determine the shadow of the document leading edge and the document leading edge, and the skew amount is calculated from the document leading edge. To do.

As will be described later, the skew detection area is set so as to include the entire leading edge of the document. FIG. 6C shows the edge extraction result in the skew detection region in the read image of FIG. 6B. FIG. 6D shows a result of determining the leading edge of the document from the edge extraction result shown in FIG. 6C and a method of calculating the skew amount.
As shown in FIGS. 6C and 6D, among the extracted edges, consecutive edges are determined as the shadow of the document, and the inside of the document shadow is the document, that is, one inside the shadow of the document. Are determined as the document edge. After determining the leading edge of the document, as shown in FIG. 6D, the skew amount θ is calculated from the angle formed by the reading area and the edge of the leading edge of the document in the main scanning direction. The coordinates of the upper left vertex and upper right vertex of the document are also calculated. In particular, the coordinates of the upper left vertex (hereinafter referred to as a reference end point) are used as a reference for rotation processing during skew correction.

Note that there is a case where the amount of skew cannot be calculated depending on the setting of the skew detection area and the size and inclination of the document for skew detection. The case where the skew amount cannot be calculated is a case where the entire document front end portion (from the left end to the right end) is not included in the skew detection area in the present embodiment.
For example, as shown in FIG. 3B, it is assumed that the inclination of the document increases with respect to the set skew detection area. In this case, either the left or right (or both) vertices on the leading edge side of the document in the transport direction deviate from the skew detection area, and the leading edge of the document cannot be specified from the extracted edge, and the skew amount is calculated. Can not do it.
After the skew detection, the skew detection unit (α) 271 detects the skew detection result (whether or not the skew amount can be calculated, and when the skew amount can be calculated, the skew amount and the reference) Keep the coordinates of the endpoints inside. The CPU (A) 251 reads and uses the skew detection result held inside the skew detection unit (α) 271.

[Image processing]
As shown in FIG. 4, the image processing unit 261 receives input of image data read from the image memory 260. The image processing unit 261 performs image processing for detecting and correcting a dust image such as a streak image on the image data on the input image data based on information set in advance from the CPU (A) 251.

[Skew correction and image transfer to controller 300]
The image data processed by the image processing unit 261 is input to the skew feeding correction unit 270 as shown in FIG. The skew correction unit 270 corrects the skew of the image by image processing (hereinafter referred to as skew correction) based on information set in advance from the CPU (A) 251 with respect to the input image data. The image is output to the image transfer unit (A) 255.

FIG. 7 is a diagram for explaining the skew correction method. A skew correction method will be described with reference to FIG.
FIG. 7A shows a read image of a document, that is, an original image before correction. In the case shown in FIG. 7A, the document image is inclined to the right by a skew amount θ. By the skew detection by the skew detection unit (α) 271 and the skew detection unit (β) 272, the skew amount θ of the document and the coordinates of the reference end point are calculated.
The skew correction unit 270 performs processing for each pixel of the input image data based on the main scanning width and sub-scanning length of the document, the skew amount (angle), and the coordinates of the reference end point set in advance by the CPU (A) 251. Pixels are rearranged by controlling the output availability and the output order.

  FIG. 7B shows an image obtained as a result of performing skew correction on the image of FIG. FIG. 7C is an enlarged view of the upper left part of FIG. 7A, and each region divided into squares represents one pixel. FIG. 7 (d) is an enlarged view of the upper left part of FIG. 7 (b). A pixel rearrangement method will be described with reference to FIGS. 7C and 7D.

  As described in the above-described skew detection method, a shadow is generated at the end portion of the document when the document is read. The area inside the shadow of the original is the original area, and in particular, one pixel inside the shadow of the original is the end of the original. The shadow of the document and the background outside the document are areas outside the document. The pixels are rearranged by sequentially selecting and outputting the pixels in the document area.

  The pixel output (rearrangement) starts from the reference end point, and is performed in order from the image leading end side (upper side in FIG. 7) for each line in the main scanning direction (left-right direction in FIG. 7). In one line in the main scanning direction, the scanning is performed in order from the left end side of the image (left side in FIG. 7). In the case of FIG. 7C, the process is performed in the order of 1-2, 1-3,. mn (m, n is an integer) indicates the nth pixel from the left end in the mth line from the top in the corrected image.

The pixel is output based on the skew amount θ, the reference end point, and whether or not the pixel is an area outside the document. In the case of FIG. 7C, after outputting the pixel 1-1 based on the information of the reference end point, the right pixel of the pixel 1-1 is confirmed. However, the pixel to the right of the pixel 1-1 is an area outside the document because it is a shadow of the document. Therefore, the lower right pixel of the pixel 1-1 is confirmed next.
Since the pixel 1-2 at the lower right of the pixel 1-1 is not an area outside the document and has not been output, the pixel is output as the pixel 1-2. After the output of the pixel 1-2, since the pixel on the right is not an area outside the document and has not already been output, the pixel is output as the pixel 1-3.

  The skew correction unit 270 repeats the above processing on the input image data, and sequentially outputs the pixels of the entire document. The output pixels are received as input by the image transfer unit (A) 255 and transferred to the controller 300 in order. The controller 300 obtains a corrected image shown in FIG. 7D by arranging the sent pixels in order for each line. Further, since the pixels in the area outside the document are not output in the rearrangement process, the image after correction is the one in which the document area is cut out as shown in FIG.

[Switch skew detection area]
[Declination detection time]
FIG. 8 is a diagram illustrating an example of timing from image reading to image output. The skew detection time will be described with reference to FIG.
FIG. 8A shows an example in which the skew detection is performed by the skew detection unit (α) 271 in the single-sided reading. In this case, as described above, after the surface image reading by the surface reading sensor 208 ((1) in the figure) is started, the skew detection unit (α) 271 selects the image data read by the surface reading sensor 208 as an image input. Through the unit 273, it is sequentially received as input ((2-1) in the figure), and skew detection is performed.
In the skew detection, as described above, edge extraction is first performed on an image in the skew detection region of the read image ((2-2) in the figure). Edge extraction is performed in parallel with image data input. After the edge extraction of the image in the skew detection area is completed, the leading edge of the document is specified from the extracted edge ((2-3) in the figure), and the skew amount and the coordinates of the reference end point are calculated (2- in the figure). 4) The skew detection is completed.
After the skew detection is completed, image processing in the image processing unit 261 ((3) in the figure), skew correction and image output ((4) in the figure) in the skew correction unit 270 and the image transfer unit (A) 255 are performed. Is called.

  As shown in FIG. 8A, in this embodiment, the skew detection time, that is, the time from the start of the skew detection to the completion of the skew detection, the edge extraction is performed by receiving the input of the image data. The ratio of such time ((2-2) in the figure) is high. Therefore, in order to shorten the skew detection time, it is effective to reduce the time for performing edge extraction upon receiving image data input.

FIG. 8B shows the case where the skew detection is performed by the skew detection unit (α) 271 in the single-sided reading as in FIG. 8A, and the skew detection region compared to the case of FIG. 8A. In this example, the sub-scan length (number of lines) is set to be short and skew detection is detected.
When the sub-scanning length of the skew detection area is shortened in this way, the time for performing edge extraction upon receiving image data input ((2-2) in the figure) is shortened, so that the timing of skew detection completion is reduced. Get faster. That is, the skew detection time is shortened. As the skew detection completion timing is advanced, the image output start timing is also advanced. As a result, the image output completion timing is also advanced, so that the time from the start of reading a document to the completion of output can be shortened. That is, the productivity of the image reading apparatus is improved.

[Set skew detection area]
When the skew detection unit (α) 271 and the skew detection unit (β) 272 detect skew, the CPU (A) 251 sets the skew detection region in advance. The skew detection unit (α) 271 and the skew detection unit (β) 272 perform skew detection by the above-described method based on the set skew detection region.
Further, as described above, it is desirable to narrow the sub-scanning length to the minimum necessary for the skew detection area indicated by the oblique lines in FIGS. 6C and 6D in order to reduce the skew detection time. On the other hand, as described above, there is a case where the skew amount cannot be calculated depending on the setting of the skew detection region and the size and inclination of the document to be detected. Therefore, the skew detection area needs to be set according to the skew amount to be detected. Specifically, the sub-scanning direction start position and the sub-scanning length of the skew detection area are set according to the maximum value of the skew amount to be detected.
The main scanning width of the skew detection area is fixed at the maximum width that can be read by the front surface reading sensor 208 and the back surface reading sensor 108.

  9 and 10 are diagrams illustrating an example of the skew detection area corresponding to the maximum value of the skew amount to be detected. The sub-scanning direction start position and the sub-scanning length of the skew detection area corresponding to the maximum value of the skew amount to be detected will be described with reference to FIGS. 9 and 10 are examples of front side reading, the same applies to the case of back side reading.

FIG. 9A1 shows a detection area (detection area length) in which the amount of skew can be calculated when an A3-size original (main scanning width: 297 [mm]) is skewed by 3 °. 9 (a2) is an enlarged view of the tip side of FIG. 9 (a1), and FIG. 9 (a3) is a diagram in which the curved conveyance path of the ADF 100 corresponding to FIG. 9 (a1) is developed on a plane. is there.
As shown in FIGS. 9 (a1) and 9 (a2), when it is desired to detect the skew of an A3 size vertically fed document up to 3 [°], image reading is started and 10.0 [mm] in the sub-scanning direction is started. When it is read, skew detection is started. In other words, 10.0 [mm] before the tip detection point detected by the lead sensor 18 reaches the surface flow reading position is set as the sub-scanning direction start position of the skew detection region.
Also, regarding the sub-scanning length of the skew detection area, since both the left and right vertices on the leading edge side of the document are within the skew detection area, that is, it is possible to detect skew of 3 [°] of A3 size vertical feed. Needs to be longer than 297 × sin 3 [°] ≈15.5 [mm].

In this embodiment, a margin is added in consideration of an error during document conveyance, and the sub-scanning length of the skew detection area in this case is 20.0 [mm]. That is, the skew detection region is defined as a point where the leading edge detection point detected by the lead sensor 18 reaches the surface flow reading position and is further read by 10 mm.
In the present embodiment, the A3-size vertically fed document is a document of the maximum size in the main scanning direction that can be read. Accordingly, when the skew detection area is set as shown in FIGS. 9A1 and 9A2, skew detection up to 3 [°] can be performed for all documents. Hereinafter, the skew detection area setting shown in FIGS. 9A1 and 9A2 is referred to as “3 [°] setting”.

  FIG. 10B1 shows a detection area (detection area length) in which a skew amount can be calculated when an A4 size longitudinal feed (main scanning width: 210 [mm]) document is skewed by 12 [°]. Yes. 10 (b2) is an enlarged view of the tip side of FIG. 10 (b1), and FIG. 10 (b3) is a diagram in which the curved conveyance path of the ADF 100 corresponding to FIG. 10 (b1) is developed on a plane. is there.

As shown in FIGS. 10 (b1) and 10 (b2), when it is desired to detect the skew of the A4 size vertically fed document up to 12 [°], the skew detection is started simultaneously with the start of the image reading. That is, 20.0 [mm] before the leading edge detection point detected by the lead sensor 18 reaches the surface flow reading position is set as the sub scanning direction start position of the skew detection region.
Further, regarding the sub-scanning length of the skew detection area, since both the left and right vertices on the leading edge side of the document are within the skew detection area, that is, it is possible to detect a skew of 12 [°] of A4 size vertical feed. Needs to be longer than 210 × sin 12 [°] ≈43.7 [mm].

In this embodiment, a margin is added in consideration of an error during document conveyance, and the sub-scanning length of the skew detection region in this case is set to 50.0 [mm]. That is, the skew detection region is defined by the point where the tip detection point detected by the lead sensor 18 reaches the surface flow reading position and is further read by 30.0 [mm].
When the detection area is set so that the skew of 12 [°] of A4 size vertical feed can be detected as shown in FIGS. 10 (b1) and 10 (b2), the original of B4 size vertical feed is also 12 [°]. ] Can be detected. Hereinafter, the setting of the skew detection area shown in FIGS. 10B1 and 10B2 is referred to as “12 [°] setting”.

[Amount of skew when mixed with different width]
FIG. 11 is a view of a bundle of documents placed on the ADF 100 and the document tray 30 as viewed from above. The document bundle is placed on the document tray 30 with the document surface facing upward (referred to as face-up). The skew amount when mixed with different widths will be described with reference to FIG.

FIGS. 11A and 11B show a case where all the originals to be read in one job are the same size (referred to as non-mixed loading).
The document tray 30 is provided with a document guide 120 movable in the vertical direction in FIG. When the user places the document bundle on the document tray 30, the document guide 120 is moved in accordance with the main scanning width of the document bundle. As a result, the placed document bundle is regulated and arranged in the main scanning direction by the document guide 120. As a result, it is possible to suppress the skew of the original when feeding.

  In FIGS. 11A and 11B, the originals read in one job are all the same size, but the main scanning width is the same even if the originals have different sizes (referred to as same width mixed loading). ), The document bundle can be adjusted in the main scanning direction by the document guide 120 in the same manner as in FIGS.

On the other hand, FIGS. 11C and 11D show a case where a document bundle read in one job is composed of two-size documents having different main scanning widths (referred to as mixed loading of different widths).
In FIGS. 11C and 11D, the document with the smaller main scanning width (“small size document” in the figure) is positioned at the top of the document bundle.
In the case of mixed loading of different widths, the document guide 120 must be moved in accordance with the maximum main scanning width of the documents, that is, the document having the larger main scanning width (“large size document” in the figure). For this reason, the small size original is not regulated by the original guide 120.

  When the user operates the operation unit 304 to input a mixed job of different widths, a message is displayed on the operation unit 304 to abut the small-size document against the document guide 120 on the back side (upper side in FIG. 11). When the user hits the small-size document against the document guide 120 according to the message, the state shown in FIG. In this case, since a small-size document is regulated by the document guide 120 on the upper side in FIG. 11 in the main scanning direction, it is inferior to the cases of FIGS. 11 (a) and 11 (b). Skew is suppressed.

On the other hand, if the user does not hit the small-size document or if the bump is insufficient, the small-size document is placed in an inclined state as shown in FIG. As a result, skew may occur in the read image of a small-size document. Further, the skew amount in this case may be larger than that in the case of a document regulated in the main scanning direction.
In the present embodiment, when a document regulated in the main scanning direction by the document guide 120 is fed and read, that is, when the document is not mixed or mixedly mixed, the skew amount generated in the read image of the document is maximum. 3 [°].

FIG. 12 is a table showing an example of the maximum skew amount generated in the read image of the small-size document in the mixed job with different widths.
In the mixed loading of different widths guaranteed as the image reading apparatus 1000, the skew amount generated in the read image of the conveyed small-size document is 8.9 [°] at the maximum for a B4-size vertically fed document, as shown in FIG. The maximum is 12 [°] for an A4 size vertically fed document. Here, it is assumed that A3 size vertical feed and B4 size vertical feed are mixed, and B4 size vertical feed and A4 size vertical feed are mixed.

[Switch skew detection area]
As described above, it is desirable to minimize the sub-scanning length of the skew detection area in order to reduce the skew detection time. Therefore, in a job other than mixed width loading (non-mixed job or same width mixed job), it is sufficient that all originals can be detected up to 3 [°]. Therefore, the skew detection area is set to 3 [°]. In this case, the skew detection area is set as the second detection area.
On the other hand, in a mixed-width job, it is necessary to set a skew detection area so that it can detect up to 8.9 [°] for a B4 size vertically fed document and 12 [°] for an A4 size vertically fed document. That is, the skew detection area needs to be set to 12 [°]. Note that the skew detection area in this case is defined as a first detection area.
Further, even when different widths are mixed, when the user strikes a small-size document, skew feeding is suppressed and the skew feeding amount may be within 3 [°]. If it falls within 3 [°] as in this case, it is preferable to set the skew detection area to 3 [°] because the skew detection time is shortened.

Therefore, in the single-sided reading mixed width mixed job, as shown in FIG. 5B, the image input selection unit 273 is set, and the skew detection unit (α) 271 and the skew detection unit ( β) 272 so that the skew detection is performed. In addition, the skew detection unit (α) 271 sets 12 [°] necessary for mixed loading of different widths, that is, the skew detection area is set as shown in FIGS. 10 (b1) and 10 (b2). I do.
In the skew detection unit (β) 272, when the skew amount falls below 3 [°], 3 [°] is set in order to shorten the skew detection time, that is, as shown in FIGS. 9 (a1) and 9 (a2). In this manner, the skew detection area is set to detect skew.

FIG. 13 is a diagram illustrating an example of timing from image reading to image output in a single-sided reading mixed width job.
FIG. 13A shows a case where the skew amount of the document is 3 [°] or less, and FIG. 13B shows a case where the skew amount of the document exceeds 3 [°] and is 12 [°] or less.
In the case of FIG. 13A, since the skew amount can be calculated by the skew detection of the skew detection unit (β) 272, the skew detection is completed at that time.

  On the other hand, in the case of FIG. 13B, the skew amount cannot be calculated by the skew detection of the skew detector (β) 272, and the skew amount is calculated by the skew detection of the skew detector (α) 271. Therefore, the skew detection is completed at that time. The CPU (A) 251 starts image processing in the image processing unit 261 when either of the skew detection unit (α) 271 and the skew detection unit (β) 272 is detected. .

As shown in FIGS. 13A and 13B, the skew detection is detected by both the skew detection unit (α) 271 and the skew detection unit (β) 272 as shown in FIG. 13A. When the skew amount of the original is 3 [°] or less, the skew detection time is shortened.
Note that for the double-sided job, the image input selection unit 273 is set as shown in FIG. The skew detection unit (α) 271 performs skew detection on the front side read image, and the skew detection unit (β) 272 performs skew detection on the back side read image.
This is because the front surface image and the back surface image have different inclinations such as the assembly error of the front surface reading unit 202 and the back surface reading unit 102, the assembly error of the ADF 100 with respect to the reader 200, and the manufacturing error of various transport rollers in the ADF 100 Is to deal with this.

  For jobs other than double-sided mixed width loading, the skew detection area (α) 271 and skew detection section (β) 272 both set the skew detection area to 3 [°], and the scanned images on the front and back sides are skewed. Perform detection. In a mixed-width job on both sides, the skew detection unit (α) 271 and the skew detection unit (β) 272 both set the skew detection region to 12 [°], and the scanned images on the front and back sides are skewed. Perform detection.

[Control of skew detection area switching]
14 and 15 are flowcharts illustrating an example of a control flow in the non-stop scanning job of the image reading apparatus 1000.
Each process shown in FIGS. 14 and 15 is mainly executed by the CPU (A) 251. 14 and 15, control of the reader 200 and the ADF 100 from when a request for start of flow reading is received from the controller 300 to when the original image is transmitted to the reading controller 300 will be described.

The flowchart of FIG. 14 will be described.
When the CPU (A) 251 receives a flow reading start request from the CPU (B) 301, the CPU (A) 251 first starts conveying a document (S101). The original is conveyed by the above-described method. However, before the original reaches the reading start position, the image input selection unit 273, the skew detection unit (α) 271 and the skew detection unit (β) 272 are set. There is a need. These settings differ depending on whether the job is a single-sided reading job or a double-sided reading job, a mixed-width job, or another job.

The CPU (A) 251 determines whether it is a single-sided reading job (S102).
If the job is a single-sided reading job (S102: Yes), the CPU (A) 251 determines whether the job is a mixed-width job (S103).
When it is a mixed-width job (S103: Yes), the CPU (A) 251 performs setting of the image input selection unit 273 as shown in FIG. 5B (S104). That is, the skew detection is performed by both the skew detection unit (α) 271 and the skew detection unit (β) 272 for the read image on the front surface.

The CPU (A) 251 sets the skew detection area of the skew detection unit (α) 271 to 12 [°] (S105).
The CPU (A) 251 sets the skew detection area of the skew detection unit (β) 272 to 3 [°] (S106).
The CPU (A) 251 performs a document surface reading process on the document whose conveyance has been started in the process of step S101 (S107). In the present embodiment, as described above, the image reading is started from 20.0 [mm] before the leading edge detection point detected by the lead sensor 18 reaches the reading start position (surface flow reading position in the process of step S107). . The read image is stored in the image memory (A) 260 and input to the image input selection unit 273 as described above.

The CPU (A) 251 starts skew reading when the skew detection time at the setting of 3 [°] has elapsed after the start of the surface image reading (S108: Yes). β) The skew detection result of 272 is read out, and it is confirmed whether or not the skew amount has been calculated (S109).
CPU (A) 251 repeats the process of step S108 if the skew detection time in the 3 [°] setting has not elapsed (S108: No).

When the skew detection unit (β) 272 can calculate the skew amount (S109: Yes), the CPU (A) 251 determines the skew amount and the reference end point coordinates held by the skew detection unit (β) 272. Is read (S110).
The CPU (A) 251 applies a skew to the read image stored in the image memory (A) 260 based on the skew amount read in the image processing in the image processing unit 261 and the processing in step S110 and the coordinates of the reference end point. The skew correction in the line correction unit 270 is performed (S111).
The CPU (A) 251 sequentially transfers the corrected images to the controller 300 by the image transfer unit (A) 255 as described above (S112).

When the transfer of one image is completed (S113: Yes), the CPU (A) 251 completes the reading process for one document. When the transfer is not completed (S113: No), the CPU (A) 251 repeats the process of step S113. If there is a document on the document tray 30 (S114-YES), feeding of the next document is started (S101).
When there is no document on the document tray 30 (S114: No), the CPU (A) 251 completes the reading process of all the documents, and thus the CPU (A) 251 ends the job.

  In a mixed-width job, if a small-size original is sufficiently abutted against the original guide 120, the main guide direction can be controlled by the original guide 120 even for a small-size original. The amount may fall within 3 [°]. When the skew amount falls within 3 [°], skew detection and skew correction can be performed in the processing from step S108 to step S111.

It is assumed that the CPU (A) 251 cannot calculate the skew amount in the skew detection performed by setting 3 [°] (S109: No).
In this case, when the skew detection time in the 12 [°] setting has elapsed (S120: Yes), the CPU (A) 251 detects the skew of the skew detection unit (α) 271 that has performed the skew detection with the 12 [°] setting. The row detection result is read, and it is confirmed whether or not the skew amount has been calculated (S121).

When the skew detection amount (α) 271 has been calculated by the skew detection unit (α) 271 (S121: Yes), the CPU (A) 251 stores the skew amount and the reference end point coordinates held by the skew detection unit (α) 271. Is read (S122).
The CPU (A) 251 performs, on the read image stored in the image memory (A) 260, the image processing in the image processing unit 261 and the skew amount read in the processing of step S122 and the coordinates of the reference end point. The skew correction in the skew correction unit 270 is performed (S123).
In the subsequent processing, the CPU (A) 251 performs the same processing as when the skew amount has been calculated by the skew detection unit (β) 272 (S109: Yes) (S112).

  As described above, when the small-size original is not sufficiently abutted against the original guide 120 in the mixed-width mixed job, the effect of the restriction in the main scanning direction by the original guide 120 is small, and the small-size original is the maximum. 12 [°] skew. When the skew amount of the small-size document exceeds 3 [°] and is 12 [°] or less, skew detection and skew correction can be performed in the processing from step S120 to step S123.

The CPU (A) 251 can calculate the skew amount when the skew detection unit (α) 271 cannot calculate the skew amount (S121: No), that is, whether the setting is 3 [°] or 12 [°]. If not, it is determined that a skew exceeding 12 [°] has occurred. In other words, skew that exceeds the amount of skew that normally occurs has occurred.
Such a document that is largely skewed and conveyed is highly likely to be damaged by a paper jam. Therefore, in this case, the CPU (A) 251 stops conveying the document in order to prevent the document from being damaged (S130). Thereafter, the CPU (A) 251 notifies the controller 300 of the occurrence of a paper jam (JAM) (S131) and ends the job.

If the job is a job other than mixed loading (S103: No), the CPU (A) 251 sets the image input selection unit 273 as shown in FIG. 5A (S140). That is, skew detection is performed by the skew detection unit (α) 271 for the read image on the front surface.
In this case, since it is a single-sided reading job, the back side image is not read, and the back side read image is not input to the image input selection unit 273.

The CPU (A) 251 sets the skew detection area of the skew detection unit (α) 271 to 3 [°] (S141).
The CPU (A) 251 performs a document surface reading process on the document whose conveyance has been started in step S101 (S142). The reading process is the same as the process in step S107.

The CPU (A) 251 starts skew reading when the skew detection time in the 3 [°] setting has elapsed after the start of the surface image reading (S143: Yes). α) The skew detection result of 271 is read to check whether or not the skew amount has been calculated (S144).
CPU (A) 251 repeats the process of step S143 if the skew detection time in the setting of 3 [°] has not elapsed (S143: No).

When the skew detection amount (α) 271 has been calculated by the skew detection portion (α) 271 (S144: Yes), the CPU (A) 251 stores the skew amount and the reference end point coordinates held by the skew detection portion (α) 271. Is read out (S145).
The CPU (A) 251 performs, on the read image stored in the image memory (A) 260, based on the image processing in the image processing unit 261 and the skew amount read in the processing of step S145 and the coordinates of the reference end point. The skew correction in the skew correction unit 270 is performed (S146).

As described above, in a job other than the mixed loading of different widths, the effect of regulating the main scanning direction by the document guide 120 of the document bundle can be expected. When the skew amount falls within 3 [°], skew detection and skew correction can be performed in the processing from step S143 to step S146.
In the subsequent processing, the CPU (A) 251 performs the processing in step S112 in the same manner as described above.

  When the skew amount cannot be calculated by the skew detection unit (α) 271 (S144: No), that is, the CPU (A) 251 cannot calculate the skew amount by setting 3 [°]. It is determined that skewing exceeding °] has occurred. In this case, the process after step S130 is performed similarly to the case of step S121: No.

On the other hand, the flowchart of FIG. 15 in the case of a duplex reading job (S102: No) will be described.
The CPU (A) 251 sets the image input selection unit 273 as shown in FIG. 5A (S201). That is, skew detection is performed by the skew detection unit (α) 271 for the read image on the front surface, and skew detection is performed by the skew detection unit (β) 272 for the read image on the back surface.

  The CPU (A) 251 determines whether the job is a mixed-width job (S202). If the job is a mixed-width job (S202: Yes), the CPU (A) 251 sets the skew detection area of the skew detection unit (α) 271 to 12 [°] (S203). The CPU (A) 251 also sets 12 [°] for the skew detection region of the skew detection unit (β) 272 (S204).

  The CPU (A) 251 performs both-side document reading processing on the document whose conveyance has been started in step S101 (S205). In the present embodiment, as described above, the leading edge detection point detected by the lead sensor 18 reaches the reading start position (each of the front surface reading position and the back surface reading position in the process of step S205) 20.0 [mm]. Image reading starts from the front. The read image is stored in the image memory (A) 260 and input to the image input selection unit 273 as described above.

When the skew detection time in the 12 [°] setting has elapsed after the start of reading the surface image (S206: Yes), the CPU (A) 251 performs the skew detection unit (α ) The skew detection result of 271 is read and it is confirmed whether or not the skew amount has been calculated (S207).
CPU (A) 251 repeats the process of step S206 if the skew detection time in the 12 [°] setting has not elapsed (S206: No).

When the skew detection amount (α) 271 has been calculated by the skew detection unit (α) 271 (S207: Yes), the CPU (A) 251 stores the skew amount and the reference end point coordinates held by the skew detection unit (α) 271. Is read (S208).
The CPU (A) 251 performs the image processing in the image processing unit 261 on the surface-read image stored in the image memory (A) 260 and the skew amount read in the processing of step S208 and the coordinates of the reference end point. Then, skew correction is performed in the skew correction unit 270 (S209).

The CPU (A) 251 sequentially transfers the corrected images to the controller 300 by the image transfer unit (A) 255 as described above (S210).
When the transfer of one image is completed (S211: Yes), the CPU (A) 251 completes the document surface reading process. When the transfer is not completed (S211: No), the CPU (A) 251 repeats the process of step S211.

When the skew detection time at the 12 [°] setting has elapsed after the start of reading the back surface image (S212: Yes), the CPU (A) 251 detects the skew by detecting the skew at the 12 [°] setting (β ) The skew detection result of 272 is read and it is confirmed whether or not the skew amount has been calculated (S213).
If the skew detection time in the 12 [°] setting has not elapsed (S212: No), the CPU (A) 251 repeats the process of step S211.
When the skew detection unit (β) 272 has calculated the skew amount (S213: Yes), the CPU (A) 251 determines the skew amount and the reference end point coordinates held by the skew detection unit (β) 272. Is read (S214).

The CPU (A) 251 performs the image processing in the image processing unit 261 on the back side read image stored in the image memory (A) 260 and the skew amount read in the processing of step S213 and the coordinates of the reference end point. Then, the skew correction in the skew correction unit 270 is performed (S215).
The CPU (A) 251 sequentially transfers the corrected images to the controller 300 by the image transfer unit (A) 255 as described above (S216).
When the transfer of one image is completed (S217: Yes), the CPU (A) 251 completes the reading process of the back side of the document. When the transfer has not been completed (S217: No), the CPU (A) 251 repeats the process of step S217.
As described above, when the process proceeds to the process of step S217, the front and back reading process for one original is completed. Thereafter, the process of step S114 is executed as in the case of the single-sided job.

The CPU (A) 251 determines that a skew exceeding 12 [°] has occurred when the skew detection unit (α) 271 cannot calculate the skew amount of the surface-read image (S207: No). To do.
Further, when the skew detection amount of the front side read image can be calculated by the skew detection unit (α) 271 but the skew amount of the back side read image cannot be calculated by the skew detection unit (β) 272 (No in S212). , 12 [°] is determined to have occurred. That is, when the skew amount cannot be calculated with the setting of 12 [°], it is determined that skew exceeding 12 [°] has occurred. In this case, the CPU (A) 251 performs the processing from step S130 onward as in the case of step S121: No.

  In the case of a job other than mixed width (S202: No), the CPU (A) 251 has three skew detection areas of the front and back read images by the skew detection unit (α) 271 and the skew detection unit (β) 272. In addition to setting [°], this is the same as in the case of a mixed-width job (S202: Yes) (S220 to 232).

  As described above, in the image reading apparatus according to this embodiment, it is possible to shorten the skew detection time when the skew amount of the document is 3 [°] or less in the single-sided mixed width job.

[Second Embodiment]
Hereinafter, a second embodiment will be described. The same functional configuration as that of the first embodiment is denoted by the same reference numeral, and the description thereof is omitted.

[Set value for each aircraft]
In the image reading apparatus 1000, an assembly error between the front surface reading unit 202 and the back surface reading unit 102, an assembly error of the ADF 100 with respect to the reader 200, a manufacturing error of various conveyance rollers in the ADF 100, and the like may occur. In this case, different inclinations may occur between the front image and the back image.

The inclination due to the manufacturing accuracy is a phenomenon unique to the machine body, and the same inclination occurs every time image reading is performed. Therefore, the inclination of the back image with respect to the front image when reading both sides of the document is calculated in advance and stored in the backup unit (A) 256 as a setting value for each machine.
As a result, if the skew amount of the front image is known, the skew amount of the back image can be calculated, so only the front skew detection can be performed without detecting the front and back skew when reading both sides of the document. Just do it.

[Set value calculation for each aircraft]
The image reading apparatus 1000 according to this embodiment includes setting values θΔ and (xΔ, yΔ) for each machine in the backup unit (A) 256.
The set value θΔ is a skew amount (relative skew amount) of the back image with respect to the front image at the time of duplex reading. The set values (xΔ, yΔ) are the difference in coordinates between the point corresponding to the surface reference end point coordinate in the back surface image at the time of duplex reading and the surface image.
The setting values θΔ and (xΔ, yΔ) are acquired when the image reading apparatus 1000 is manufactured. Prior to the acquisition of the set values θΔ and (xΔ, yΔ), the operator places an adjustment chart on the document tray 30 in parallel, and regulates in the main scanning direction by the document guide 120.

  FIG. 16 is a flowchart illustrating an example of acquisition control of setting values θΔ and (xΔ, yΔ) for each machine of the image reading apparatus 1000. Note that each process shown in FIG. 16 is mainly executed by the CPU (A) 251.

CPU (A) 251 The conveyance of the document is started with respect to the document (adjustment chart) placed in parallel on the document tray 30 (S301). The original is conveyed by the above-described method. However, before the original reaches the reading start position, the image input selection unit 273, the skew detection unit (α) 271 and the skew detection unit (β) 272 are set. There is a need.
The CPU (A) 251 sets the image input selection unit 273 as shown in FIG. 5A (S302). That is, skew detection is performed by the skew detection unit (α) 271 for the read image on the front surface, and skew detection is performed by the skew detection unit (β) 272 for the read image on the back surface.

The CPU (A) 251 sets the skew detection area of the skew detection unit (α) 271 to 3 [°] (S303).
The CPU (A) 251 also sets 3 [°] for the skew detection area of the skew detection unit (β) 272 (S304).
If the skew amount generated in the read images on the front and back sides of the conveyed document is up to 3 [°] by the processing in steps S302 and S303 and the processing in steps S302 and S304, the skew detection unit (α) 271. The skew detection unit (β) 272 can detect the skew.

The CPU (A) 251 performs a double-sided reading process on the document whose conveyance has been started in step S301 (S305).
In the present embodiment, as described above, the image is detected from 20 [mm] before the leading edge of the document detected by the read sensor 18 reaches the reading start position (each of the front side scanning reading position and the back side scanning reading position in the process of step S305). Start reading.
The read image is stored in the image memory (A) 260 and input to the image input selection unit 273 as described above.

When the skew detection time in the 3 [°] setting has elapsed after the start of the surface image reading (S306: Yes), the CPU (A) 251 detects the skew detection by the skew detection by the 3 [°] setting (α ) The skew detection result of 271 is read, and it is confirmed whether or not the skew amount has been calculated (S307).
CPU (A) 251 repeats the process of step S306 if the skew detection time in the setting of 3 [°] has not elapsed (S306: No).
When the skew detection amount (α) 271 has been calculated by the skew detection unit (α) 271 (S307: Yes), the CPU (A) 251 determines the skew amount θ1 held by the skew detection unit (α) 271 and the top right corner of the document. The coordinates (x1, y1) are read out (S308).

When the skew detection time in the 3 [°] setting has elapsed after the start of reading the back surface image (S309: Yes), the CPU (A) 251 performs the skew detection unit (β ) The skew detection result of 272 is read and it is confirmed whether or not the skew amount has been calculated (S310).
CPU (A) 251 repeats the process of step S309 if the skew detection time in the setting of 3 [°] has not elapsed (S309: No).

When the skew detection amount (β) 272 has been calculated by the skew detection unit (β) 272 (S310: Yes), the CPU (A) 251 determines the skew amount θ2 held by the skew detection unit (β) 272 and the reference end point. The coordinates (x2, y2) are read (S311).
The CPU (A) 251 calculates the set value θΔ based on the skew amount θ1 read in the process of step S308 and the skew amount θ2 read in the process of step S311 (S312). Hereinafter, calculation of the set value θΔ will be described.

FIG. 17 is a diagram for explaining a method of calculating the set value θΔ.
ΘΔ is obtained with respect to the skew amount θ1 = −2 [°] of the front side read image shown in FIG. 17A and the skew amount θ2 = 3 [°] of the back side read image shown in FIG. .
When there is no difference between the front and back of the skew amount with respect to the front image of FIG. 17A, the back image is as shown in FIG. 17C, and θ2 = −θ1. A difference (shown in FIG. 17D) of the skew amount in FIG. 17B with respect to FIG. 17C becomes the set value θΔ. Accordingly, the set value θΔ is obtained by θΔ = θ2 − (− θ1) = θ2 + θ1. Therefore, at this time, θΔ = 1 [°].
As shown in FIG. 17, the skew amount is positive for counterclockwise and negative for clockwise with respect to the upper left vertex of the document, which is the reference end point.

  Further, the set values (xΔ, yΔ) are calculated based on the reference end point coordinates (x1, y1) read in the process of step S308 and the coordinates (x2, y2) of the upper right corner of the original read in the process of step S311 ( S313). Hereinafter, calculation of the set values (xΔ, yΔ) will be described.

FIG. 18 is a diagram for explaining a method of calculating the set values (xΔ, yΔ).
With respect to the coordinates (x1, y1) of the upper right vertex of the document of the front side scanned image shown in FIG. 18A and the coordinates (x2, y2) of the reference end point of the back side scanned image shown in FIG. )
When there is no difference between the front and back of the skew amount with respect to the front image of FIG. 18 (a), the back image is as shown in FIG. 18 (c). In this case, the reference end point corresponds to the upper right vertex of the document on the front surface, and (x3, y3) = (x1, ymax−y1).
In FIG. 18, (x0, y0) is the origin, that is, (x0, y0) = (0, 0), and ymax is the maximum width (number of pixels) that can be read by the front surface reading sensor 208 and the back surface reading sensor 108. It is.

  The difference between the coordinates of the reference end point in FIG. 18B relative to FIG. 18C (shown in FIG. 18D) is the set value (xΔ, yΔ). Therefore, the set values (xΔ, yΔ) are obtained by xΔ = x2−x3 = x2−x1, yΔ = y2−y3 = y2− (ymax−y1).

  Returning to the description of FIG. 16, the CPU (A) 251 stores the calculated set values θΔ and (xΔ, yΔ) in the backup unit (A) 256 (S313), and ends the process.

The CPU (A) 251 determines that a skew exceeding 3 [°] has occurred when the skew detection unit (α) 271 cannot calculate the skew of the surface-read image (S307: No). To do.
Further, the CPU (A) 251 can calculate the skew amount of the front side read image by the skew detection unit (α) 271, but can calculate the skew amount of the back side read image by the skew detection unit (β) 272. If not (S310: No), it is determined that a skew exceeding 3 [°] has occurred.
In this case, the CPU (A) 251 stops conveying the document in order to prevent the document from being damaged (S320).
The CPU (A) 251 notifies the controller 300 of the occurrence of a paper jam (JAM) (S321) and ends the process.
When a JAM is generated in this way when setting values are acquired, the operator receives a JAM generation notification through the operation unit 304. Thereafter, JAM elimination processing is performed in accordance with the instruction displayed on the operation unit 304, and then the adjustment chart is again placed in parallel on the document tray 30, and the document guide 120 is used to regulate in the main scanning direction. Get the value again.

[Calculating skew amount using set values]
By using the set values θΔ and (xΔ, yΔ), if only the front skew detection is performed without performing the skew detection for each of the front and back read images even when reading both sides, the surface skew amount and the set value θΔ are obtained. Based on this, the skew amount on the back surface can be calculated. Further, the reference end point coordinates on the back surface can also be calculated based on the coordinates of the upper right vertex of the document on the front surface and the set values (xΔ, yΔ).

  For example, when the skew amount on the front side is θ1 and the coordinates of the upper right vertex of the document are (x1, y1), the skew amount θ2 on the back side and the reference end point coordinates (x2, y2) are θ2 = θΔ−θ1, x2 = xΔ + x1. , Y2 = yΔ + (ymax−y1).

[Switch skew detection area]
19 and 20 are flowcharts illustrating an example of a control flow in a non-stop scanning job of the image reading apparatus 1000. Each process shown in FIGS. 19 and 20 is mainly executed by the CPU (A) 251. 19 and 20, control of the reader 200 and the ADF 100 from when a request for start of flow reading is received from the controller 300 to when the original image is transmitted to the reading controller 300 will be described.
Each process shown in FIG. 19 is the same as the process shown in FIG. That is, the control at the time of single-sided reading in this embodiment is the same as in the case of the first embodiment.
Hereinafter, the case of a duplex reading job (S402: No) will be described with reference to FIG.

The CPU (A) 251 determines whether the job is a mixed-width job (S501).
If the job is a mixed-width job (S501: Yes), the CPU (A) 251 sets the image input selection unit 273 as shown in FIG. 5B (S502). That is, the skew detection is performed by the skew detection unit (α) 271 and the skew detection unit (β) 272 for the read image on the front surface.

The CPU (A) 251 sets the skew detection area of the skew detection unit (α) 271 to 12 [°] (S503).
The CPU (A) 251 sets the skew detection area of the skew detection unit (β) 272 to 3 [°] (S504).
The CPU (A) 251 performs reading processing on both sides of the document for which the conveyance is started in step S401 (S505).

  In the present embodiment, as described above, the tip detection point detected by the lead sensor 18 reaches the reading start position (in the processing of step S505, each of the front surface reading position and the back surface reading position) 20.0 [mm]. Image reading starts from the front. The read image is stored in the image memory (A) 260 and input to the image input selection unit 273 as described above.

The CPU (A) 251 starts skew reading when the skew detection time in the 3 [°] setting has elapsed after the start of the surface image reading (S506: Yes), and performs the skew detection unit (β ) The skew detection result of 272 is read, and it is confirmed whether the skew amount has been calculated (S507).
If the skew detection time in the 3 [°] setting has not elapsed (S506: No), the CPU (A) 251 repeats the process of step S506.

When the skew detection unit (β) 272 has calculated the skew amount (S507: Yes), the CPU (A) 251 determines the skew amount and the reference end point coordinates held by the skew detection unit (β) 272. , And the coordinates of the upper right vertex are read (S508).
The CPU (A) 251 performs the image processing in the image processing unit 261 on the surface-read image stored in the image memory (A) 260 and the skew amount and the coordinates of the reference end point read out in step S508. Then, skew correction is performed in the skew correction unit 270 (S509).

The CPU (A) 251 sequentially transfers the corrected images to the controller 300 by the image transfer unit (A) 255 as described above (S510).
When the transfer of one image is completed (S511: Yes), the document surface reading process is completed. When the transfer is not completed (S511: No), the CPU (A) 251 repeats the process of step S511.

The CPU (A) 251 reads the set values θΔ and (xΔ, yΔ) stored in the backup unit (A) 256 (S512).
The CPU (A) 251 determines the skew amount of the back side read image based on the skew amount and the coordinates of the upper right vertex read in the process of step S508 and the set values θΔ and (xΔ, yΔ) read in the process of step S512. Then, reference end point coordinates are calculated (S513).

The CPU (A) 251 performs the skew amount and the reference end point coordinates of the back side read image calculated in the image processing of the image processing unit 261 and the process of step S513 for the back side read image stored in the image memory (A) 260. Based on the above, skew correction is performed in the skew correction unit 270 (S514).
The CPU (A) 251 sequentially transfers the corrected images to the controller 300 by the image transfer unit (A) 255 as described above (S515).
When the transfer of one image is completed (S516: Yes), the reading process of the back side of the document is completed. When the transfer has not been completed (S516: No), the CPU (A) 251 repeats the process of step S516.
As described above, when the process proceeds to step S516, the front and back reading process for one original is completed. Thereafter, as in the case of the single-sided job, the process of step S413 is executed.

  The CPU (A) 251 determines that the skew detection time in the 12 [°] setting has elapsed when the skew amount cannot be calculated in the skew detection performed by the 3 [°] setting (S507: No). S520: Yes). In this case, the CPU (A) 251 reads out the skew detection result of the skew detection unit (α) 271 that has detected skew according to the 12 [°] setting, and checks whether the skew amount has been calculated. (S521).

When the skew detection amount (α) 271 has been calculated by the skew detection portion (α) 271 (S521: Yes), the CPU (A) 251 has the skew amount and the reference end point coordinates held by the skew detection portion (α) 271. , And the coordinates of the upper right vertex are read (S522).
The CPU (A) 251 applies the skew amount and the reference end point coordinates read out in the image processing of the image processing unit 261 and the processing in step S522 to the surface reading image stored in the image memory (A) 260. Then, the skew correction in the skew correction unit 270 is performed (S523).

The CPU (A) 251 sequentially transfers the corrected images to the controller 300 by the image transfer unit (A) 255 as described above (S524).
When the transfer of one image is completed (S525: Yes), the document surface reading process is completed. When the transfer has not been completed (S525: No), the CPU (A) 251 repeats the process of step S525.

The CPU (A) 251 reads the set values θΔ and (xΔ, yΔ) stored in the backup unit (A) 256 (S526).
The CPU (A) 251 determines the skew amount of the back side scanned image based on the skew amount read in the process of step S522 and the coordinates of the upper right vertex and the set values θΔ and (xΔ, yΔ) read in the process of step S526. Then, the reference end point coordinates are calculated by the above-described method (S527).

The CPU (A) 251 performs the skew amount and the reference end point coordinates of the back side read image calculated in the image processing of the image processing unit 261 and the process of step S527 for the back side read image stored in the image memory (A) 260. Based on the above, the skew correction in the skew correction unit 270 is performed (S528).
In the subsequent processing, the CPU (A) 251 performs the same processing as when the skew amount has been calculated by the skew detection unit (β) 272 (S507: Yes) (S515).

  The CPU (A) 251 can calculate the skew amount when the skew detection unit (α) 271 cannot calculate the skew amount (S521: No), that is, whether the setting is 3 [°] or 12 [°]. If not, it is determined that skewing exceeding 12 [°] has occurred. In this case, the CPU (A) 251 performs the processing after step S430 as in the case of step S421: No.

Note that the processing of the CPU (A) 251 in the case of a job other than the mixed loading of different widths (S501: No) is the same as that in the first embodiment, and therefore the description thereof is omitted.
In the present embodiment, in a double-sided job other than mixed-width mixed loading, the front side read image is detected by the skew detection unit (α) 271 and the back side read image is detected by the skew detection unit (β) 272. However, it is not limited to this. For example, the skew detection unit (α) 271 performs skew detection on the read image on the front surface, and the skew amount on the back surface is set with the set value set values θΔ and (xΔ, yΔ) and the surface as in the case of mixed loading of different widths. It may be calculated from the skew amount.

  As described above, in the image reading apparatus according to this embodiment, it is possible to reduce the time required for skew detection when the skew amount of a document is 3 ° or less in a double-sided mixed width job.

  The embodiment described above is for explaining the present invention more specifically, and the scope of the present invention is not limited to these examples.

  DESCRIPTION OF SYMBOLS 1 ... Paper feed roller, 2 ... Separation roller, 3 ... Extraction roller, 4 ... Conveyance roller, 5 ... Surface reading conveyance roller, 6 ... Surface glass facing member, 7 ...・ Back reading conveyance roller, 8... Back glass facing member, 12... Discharge roller, 13... Discharge tray, 14 .. document detection sensor, 15, 16, 17, 18, 19,. -Sensor, 21 ... Separation pad, 30 ... Document tray, 51 ... Front reading upstream roller, 52 ... Front reading downstream roller, 53 ... Back reading upstream roller, 54 ... Back reading Downstream roller, 101... Backside scanning glass, 102... Backside reading unit, 103... Backside LED, 108... Backside reading sensor, 110. 201... Surface flow reading glass, 202... Surface reading unit, 203... Surface LED, 208... Surface reading sensor, 209. 251 ... CPU (A), 252 ... ROM (A), 253 ... RAM (A), 255 ... Image transfer unit (A), 256 ... Backup unit (A), 260 .. Image memory (A), 261... Image processing unit (A), 270... Skew correction unit, 271... Skew detection unit (.alpha.), 272. ) 273... Image input selection unit, 226... Optical system motor, 300... Controller, 301... CPU (B), 302. ), 304... Operation unit, 306. Li (B), 308 ... image transfer unit (B), 401 ... communication line, 100 ... ADF, 200 ... reader, 1000 ... image reading apparatus.

Claims (8)

First reading means for reading an image on one side of the conveyed document;
First detection means for detecting a skew amount of the document based on image data of the image of the document output from the first reading means, corresponding to a first detection area;
Second detection means for detecting a skew amount of the document based on image data of the image of the document output from the first reading means, corresponding to a second detection area;
Correction means for performing skew correction on image data of the image of the document based on the skew amount detected by the first detection means or the second detection means;
Control means for controlling the first detection area and the second detection area,
The length of the first detection region in the document transport direction is longer than the length of the second detection region in the document transport direction.
Image reading device. The first detection area and the second detection area are on the front end side of the document in the document transport direction.
The image reading apparatus according to claim 1. The control means controls to input the read image to the first and second detection means when the original reading by the first reading means reads an original having a different width. To
The image reading apparatus according to claim 1. The correction means, when the skew amount of the document can be detected by the second detection means, performs skew correction based on the detection result of the second detection means,
When the second detection means cannot detect the skew amount of the original and the first detection means can detect the skew amount of the original, the detection result of the first detection means is obtained. Based on the skew correction based on,
The image reading apparatus according to claim 1, 2 or 3. Second reading means for reading an image on the other side of the document;
The control means controls the input of the read image to the first and second detection means when the original reading by the first and second reading means reads an original having a different width. Characterized by the
The image reading apparatus according to claim 1. Storage means for holding a relative skew amount of the image read by the second reading means with respect to the image read by the first reading means;
When the second detection unit can detect the skew amount of one side of the document, the correction unit corrects the skew correction of the read image on the one side as the detection result of the second detection unit. Based on
The skew feeding correction of the read image on the other side of the document is performed based on the skew feeding amount of the one side and the relative skew feeding amount held in the storage unit,
The image reading apparatus according to claim 5. When the second detection unit cannot detect the skew amount of the document and the first detection unit can detect the skew amount of the document, the correction unit reads the one surface. Performing skew correction of the image based on the detection result of the first detection means,
The skew feeding correction of the read image on the other side of the document is performed based on the skew feeding amount of the one side and the relative skew feeding amount held in the storage unit,
The image reading apparatus according to claim 6. The control means sets the first detection area in which the tilt amount generated in the read image is set to 12 [°] in the first detection means, and the second detection means adds the first detection area to the read image. The second detection area in which the amount of inclination generated is set to 3 [°] is set,
The image reading apparatus according to claim 1.