Image reading device, image forming system

Abstract

Adjusts the geometric properties of an image without using a dedicated format. [Solution] The image reading device includes a document tray (121) on which a document is placed, an ADF 20 that transports the document from the document tray (121), a surface reading unit 104 that reads the document transported by the ADF 20, and a CPU 301 that detects a first vertex of the document from the document reading result by the surface reading unit 104 and derives a correction value for correcting the difference between the detected first vertex and a second vertex of a reference image.

Description

Technical Field

[0001] The present invention relates to an image reading device that reads an image from a document, and an image forming system including the image reading device.

Background Art

[0002] Some image reading devices installed in image forming systems such as copiers have a configuration in which an automatic document feeder (hereinafter referred to as "ADF") is provided in a reading unit. When using the ADF, the image reading device reads a document conveyed by the ADF. The image reading device reads the image of the document line by line in the main scanning direction, with the direction intersecting the conveyance direction of the document being the main scanning direction. The conveyance direction of the document becomes the sub-scanning direction. By using the ADF, the image reading device can read the images on both sides of the document in a single document conveyance.

[0003] When using the ADF, in order to accurately read the document at each position in the sub-scanning direction, image position adjustment is performed to match the operation timings of the document conveyance mechanism and the image reading mechanism. When reading images from both sides of the document, adjustment processes such as alignment of the reading results (hereinafter referred to as "read images") of the front surface image and the back surface image of the document, and tilt correction are required.

[0004] Generally, the image position adjustment of the image reading device is performed by fine adjustment of height, angle, and relative position when attaching the ADF to the main body of the image reading device during manufacturing. In Patent Document 1, a method of performing image position adjustment by comparing the reading result of a document conveyed by the ADF with the theoretical value of the read image is disclosed.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0006] In the method described in Patent Document 1, a document with an image position adjustment format printed on it is read. The document with the image position adjustment format printed on it has an image that serves as a reference for geometric characteristics such as the position, tilt, and magnification of the image. Based on the reading result of the document with the image position adjustment format printed on it, theoretical values ​​of the read image are generated.

[0007] Moving the image reader due to relocation or other reasons can cause the ADF (Automatic Document Feeder) to shift position. In this case, it is inconvenient that image position adjustment cannot be performed unless a document with a printed format for image position adjustment is prepared. Furthermore, if the document is made of a type of paper that is prone to shifting during transport, such as thick paper, it is difficult to achieve high-precision image position adjustment unless a document with a printed format for image position adjustment tailored to that type of paper is used.

[0008] In view of the above-mentioned problems, the primary objective of the present invention is to provide an image reading device that can adjust the geometric characteristics of an image without using a dedicated format such as a document on which a format for image position adjustment is printed. [Means for solving the problem]

[0009] The image reading device of the present invention is characterized by comprising: a document tray on which a document is placed; a document transport means for transporting the document from the document tray; a reading means for reading the document transported by the document transport means; and a derivation means for detecting a first vertex of the document from the reading result of the document by the reading means and deriving a correction value for correcting the difference between the detected first vertex and a second vertex of a reference image. The present invention provides an image forming system comprising: an image reading device; and an image forming means for forming an image on paper based on image data representing a document read by the image reading device, wherein the image reading device has a document tray on which a document is placed; a document transport means for transporting the document from the document tray; a reading means for reading the document transported by the document transport means; a derivation means for detecting a first vertex of the document from the reading result of the document by the reading means and deriving a correction value for correcting the difference between the detected first vertex and a second vertex of a reference image; and an image processing means for correcting the reading result of the document transported by the document transport means by the reading means and generating the image data based on the correction value. [Effects of the Invention]

[0010] According to the present invention, it is possible to adjust the geometric properties of an image without using a dedicated format. [Brief explanation of the drawing]

[0011] [Figure 1] (a) and (b) are diagrams showing the configuration of an automatic document scanning device. [Figure 2] Diagram explaining the control panel. [Figure 3] Internal diagram of an automatic document scanning device. [Figure 4] Configuration diagram of the control system. [Figure 5] (a) to (g) are example diagrams of the input screen. [Figure 6] A flowchart illustrating the automatic geometric adjustment process. [Figure 7] (a) and (b) are diagrams illustrating the "Create Reference Coordinates" process. [Figure 8] A flowchart illustrating the "Create Reference Coordinates" process. [Figure 9] (a) to (d) are diagrams illustrating the "Coordinate Acquisition for Adjustment" process. [Figure 10] A flowchart illustrating the process of "acquiring coordinates for adjustment." [Figure 11] (a) to (c) are diagrams illustrating the coordinate information detection process. [Figure 12] Flowchart representing coordinate information detection processing. [Figure 13] (a) to (e) are explanatory diagrams of correction value calculation processing. [Figure 14] (a) to (c) are exemplary diagrams of a reference coordinate registration screen. [Figure 15] Configuration diagram of an image forming system.

Embodiments for Carrying Out the Invention

[0012] Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings.

[0013] FIG. 1 is a configuration diagram of an automatic document reading apparatus including an image reading apparatus according to the present embodiment. FIG. 1(a) is an external perspective view of the automatic document reading apparatus 10. The automatic document reading apparatus 10 includes an image reading apparatus that reads a document (hereinafter referred to as the "reader 40") and an ADF 20 that automatically conveys the document to the reading position by the reader 40. The ADF 20 functions as a sheet conveying apparatus that conveys a sheet-like document to the reader 40. The ADF 20 is openable and closable with respect to the reader 40.

[0014] FIG. 1(b) shows the state where the ADF 20 is in the open state. The reader 40 is provided with a document table glass 101, a front reading glass 102, and a abutting portion 110 on the surface on the ADF 20 side. The document table glass 101 can place a document. When placing a document on the document table glass 101, the document is abutted at the corner portion against the abutting portion 110 and placed with the image surface to be read facing the document table glass 101. A front reading unit 104 is provided inside the housing of the reader 40. The front reading unit 104 reads the document placed on the document table glass 101. Since the front reading unit 104 reads in a state where the document is abutted against the abutting portion 110, it can read the document without deviation in geometric characteristics such as the position and shape of the image. Further, the front reading unit 104 can read the document conveyed by the ADF 20 through the front reading glass 102.

[0015] The automatic document scanning device 10 has an operating unit. Figure 2 is an explanatory diagram of the operating unit. The operating unit 90 is a user interface having an input interface and an output interface. In Figure 2, the input interface is a group of operation keys 92 including a start key 93. The input interface may also include, for example, a touch panel. In Figure 2, the output interface is a display unit 91. The output interface may also include, for example, a speaker.

[0016] Figure 3 is an internal configuration diagram of the automatic document reader 10. As described above, the reader 40 is equipped with a document glass 101, etc., between it and the ADF 20. The ADF 20 is rotatably supported by the reader 40 by a hinge so that it can be opened and closed relative to the document glass 101. The reader 40 and the ADF 20 each have a reading unit for reading a document. The reader 40 is equipped with the above-mentioned front reading unit 104 for reading the first side of the document (hereinafter, the first side may also be called the "front side"). The ADF 20 is equipped with a back reading unit 212 for reading a second side of the document that is different from the first side (hereinafter, the second side may also be called the "back side"). The document may be blank, or an image may be formed on one side or both sides.

[0017] The ADF20 includes a document tray 121 on which documents can be placed and an output tray 122 on which scanned documents are ejected. The document tray 121, together with the transport guide 123 of the ADF20, constitutes a sheet stacking section 120. The document tray 121 has a width restriction plate 125 on the document stacking surface that is movable in the width direction intersecting the document transport direction. The width restriction plate 125 restricts the position of documents placed on the document tray 121 in the width direction. Two width restriction plates 125 are provided in the width direction, and the two width restriction plates 125 sandwich the documents placed on the document tray 121.

[0018] The two width-regulating plates 125 move in conjunction with each other via an interlocking mechanism (not shown) located within the document tray 121, such as a rack and pinion mechanism, when one moves. In this embodiment, the document transport center is in the center in the width direction, and the two width-regulating plates 125 move closer to or further apart from the center in the width direction. This ensures that the document transport center is in the same position regardless of the size of the document. Furthermore, the width of the document placed in the document tray 121 can be detected by measuring the distance between the two width-regulating plates 125.

[0019] The ADF20 includes a pickup roller 111 as a transport rotating body for feeding paper from the document tray 121, and a separation drive roller 112 and a retard roller 113 which constitute a separation roller pair. The ADF20 also includes a registration roller pair 114a, 114b, a lead roller pair 115a, 115b, a transport roller pair 117a, 117b, and an ejection roller pair 119a, 119b for transporting the fed documents.

[0020] A first document presence detection sensor 204 and a second document presence detection sensor 205 are provided between the pickup roller 111 and the separation roller pair. A post-separation sensor 207 is provided in the transport path between the separation roller pair and the registration roller pair 114a, 114b. A lead sensor 210 is provided in the transport path between the registration roller pair 114a, 114b and the lead roller pair 115a, 115b. A backside reading unit 212 is provided in the transport path between the lead roller pair 115a, 115b and the transport roller pair 117a, 117b.

[0021] The reader 40 has the above-mentioned surface reading glass 102 on the same surface as the document glass 101 and has a surface reading unit 104 inside. The surface reading glass 102 is provided with a white reference member 103 for the surface facing inward from the reader 40 and a white reference member 218 for the back facing outward from the reader 40. The white reference member 103 for the surface and the white reference member 218 for the back are used for shading processing.

[0022] The surface scanning unit 104 is positioned on the guide 109 and is movable along the guide 109. When scanning documents transported by the ADF 20, the surface scanning unit 104 is positioned directly below the surface scanning glass 102, and when scanning documents placed on the document glass 101, it moves along the guide 109. In this embodiment, the operation mode for scanning documents transported by the ADF 20 is called "swimming mode," and the operation mode for scanning documents placed on the document glass 101 is called "fixed scanning mode." Furthermore, the scanning operation performed in the swimming mode is called "swimming," and the scanning operation performed in the fixed scanning mode is called "fixed scanning." The swimming mode is set when the first document presence detection sensor 204 detects documents loaded in the document tray 121 or when the user explicitly instructs it using the operation unit 90 or the like. The fixed scanning mode is set when a document placed on the document glass 101 is detected or when the user explicitly instructs it using the operation unit 90.

[0023] The document transport operation by the ADF20 will now be described. The pickup roller 111 is mounted to be oscillating freely by the arm 111a. The arm 111a is driven to move up and down, causing the pickup roller 111 to come into contact with / separate from the topmost document in the stack of documents loaded on the document tray 121. The arm 111a is equipped with a locking mechanism that locks a regulating plate 130 that comes into contact with the edge of the document. This locking mechanism locks the regulating plate 130 in a position that restricts the leading edge of the document when the pickup roller 111 is raised. The locking mechanism also releases the lock on the regulating plate 130 when the pickup roller 111 is lowered, allowing the document to pass through.

[0024] Upstream of the regulating plate 130 in the document transport direction, the detection member 160 of the first document presence detection sensor 204 and the detection member 150 of the second document presence detection sensor 205 are arranged side by side in the width direction. The detection members 160 and 150 are positioned downstream of the pickup roller 111 in the document transport direction, specifically downstream of the position where the pickup roller 111 descends and contacts the document. The first document presence detection sensor 204 outputs an ON signal when the detection member 160 is pressed downward by the document and rotated. The second document presence detection sensor 205 outputs an ON signal when the detection member 150 is pressed upward by the document and rotated. The presence or absence of a document on the document tray 121 is determined by the signals output from the first document presence detection sensor 204 and the second document presence detection sensor 205.

[0025] The original document fed by the pickup roller 111 is separated into individual sheets by the separation drive roller 112 and the retard roller 113 and then transported. A torque limiter is positioned in the rotation support structure of the retard roller 113, so that the retard roller 113 follows the separation drive roller 112 when only one document is fed, and does not rotate when two or more documents are fed. In this way, the documents are separated one by one. The retard roller 113 may also be driven in the opposite direction to the transport direction. In fixed reading mode, the surface reading unit 104 moves directly below the transport guide plate 211 to read the document being transported simultaneously with the descent of the pickup roller 111.

[0026] The leading and trailing ends of the document after passing through the separation roller pair are detected by the post-separation sensor 207. The detection results from the post-separation sensor 207 serve as a reference for the raising and lowering timing of the pickup roller 111, as well as the start and stop timing of its drive. The pickup roller 111 and the separation drive roller 112 are driven by the same drive source.

[0027] The registration roller pair 114a and 114b correct the skew of the document. The document, corrected for skew by the registration roller pair 114a and 114b, is transported toward the surface reading glass 102 by the lead roller pair 115a and 115b. The driving timing of the registration roller pair 114a and 114b and the lead roller pair 115a and 115b is controlled by the detection result of the separation sensor 207. A transport guide plate 211 is positioned opposite the surface reading glass 102. The transport guide plate 211 guides the document so that it does not lift away from the surface reading glass 102 as it passes between it and the transport guide plate 211. The position of the transport guide plate 211 becomes the reading position of the document by the surface reading unit 104. The registration roller pair 114a and 114b and the lead roller pair 115a and 115b are driven by the same drive source.

[0028] When scanning one side of a document, the image on the front of the document is read by the surface reading unit 104 via the surface reading glass 102. The surface reading unit 104 includes an LED (Light Emitting Diode) 105 which is a light source, a lens array 107, and a reading sensor 108 which is a line sensor.

[0029] The surface reading unit 104 illuminates the surface (image surface) of the document with an LED 105. The reflected light from the surface of the document passes through the lens array 107 and is received by the reading sensor 108. The reading sensor 108 has multiple light-receiving elements arranged in a straight line, and each light-receiving element receives the reflected light. Each light-receiving element converts the reflected light into photoelectric energy to generate and output image data as the result of reading the document. The multiple light-receiving elements are arranged in a direction intersecting the document transport direction, and photoelectric conversion is performed line by line. The LED 105 illuminates the light linearly in the same direction as the arrangement of the light-receiving elements. Due to this configuration, the direction intersecting the document transport direction becomes the main scanning direction, and the document transport direction becomes the sub-scanning direction.

[0030] The original document whose surface has been read by the surface reading unit 104 is discharged to the discharge tray 122 by the transport roller pair 117a, 117b and the discharge roller pair 119a, 119b. The transport roller pair 117a, 117b and the discharge roller pair 119a, 119b are driven by the same drive source.

[0031] When scanning both sides of a document, the image on the front side of the document is read by the front-side reading unit 104, and the image on the back side of the document is read by the back-side reading unit 212. The back-side reading unit 212 has the same configuration as the front-side reading unit 104 and includes an LED 214 as a light source, a lens array 215, and a reading sensor 216 as a line sensor. The back-side reading unit 212 reads the back side of the document whose front side has been read by the front-side reading unit 104. The document reading operation by the back-side reading unit 212 is the same as that of the front-side reading unit 104. The main scanning direction and sub-scanning direction are also the same.

[0032] In fixed scanning mode without using the ADF20, the document is placed on the document glass 101 with the image side facing the reader 40. In this case, the document on the document glass 101 does not move, and the surface scanning unit 104 scans the document by moving directly below the document glass 101 along the guide 109. The scanning operation of the surface scanning unit 104 itself is the same as in the scrolling scanning mode. In fixed scanning mode, the direction of movement of the surface scanning unit 104 is the sub-scanning direction, and the direction intersecting the direction of movement is the main scanning direction.

[0033] (Control system) Figure 4 is a diagram showing the configuration of a control system that controls the operation of the automatic document reader 10. The control system includes a configuration provided on the reader 40 side and a controller 310. The controller 310 may be provided inside the automatic document reader 10, but if the automatic document reader 10 is connected to an external device, for example, an image forming apparatus, the controller 310 may be provided on the image forming apparatus side. When the automatic document reader 10 is connected to such an image forming apparatus, it constitutes, for example, a copier, a facsimile machine, a multifunction printer, or an MFP (Multi Function Peripheral).

[0034] The reader 40 is an information processing device comprising a CPU (Central Processing Unit) 301, a ROM (Read Only Memory) 302, and a RAM (Random Access Memory) 303. The reader 40 also includes an image memory 305, an image processing unit 306, a shading memory 307, and an image transfer unit 304 as part of its control system. The image transfer unit 304 is connected to the controller 310 via an image transfer line 402 for communication. In addition to the configuration described in Figure 3, the reader 40 also includes an optical HP sensor 226 and an optical motor 225. The ADF 20 includes a feed clutch 223 and a transport motor 224 in addition to the configuration described in Figure 3. The automatic document reader 10 also includes a non-volatile memory 308.

[0035] The CPU 301 controls the operation of the reader 40 and ADF 20 by executing a computer program stored in the ROM 302. The RAM 303 provides a workspace for the CPU 301 when it performs processing. The CPU 301 is communicated via a bus to each part of the reader 40 and ADF 20. The CPU 301 is communicated via a communication line 401 to the controller 310.

[0036] The CPU 301 controls the drive of the transport motor 224, which drives each of the rollers that transport the document, in order to realize the document transport function of the ADF 20. The transport motor 224 is connected to the pickup roller 111 and the separation drive roller 112 via the feed clutch 223. By disengaging the feed clutch 223, the transport of the document can be stopped at position P (see Figure 3), just before reaching the registration roller pair 114a, 114b. The transport motor 224 is also connected to and drives the registration roller pair 114a, 114b, the lead roller pair 115a, 115b, the transport roller pair 117a, 117b, and the discharge roller pair 119a, 119b.

[0037] The transport motor 224 is a pulse motor. The CPU 301 controls the transport motor 224 by controlling the number of drive pulses. The number of drive pulses correlates with the transport distance of the document being transported. Therefore, the CPU 301 controls each load based on the transport distance derived from the number of drive pulses, and transports the document.

[0038] The CPU 301 detects the presence or absence of a document in the document tray 121 based on the detection results of the first document presence detection sensor 204 and the second document presence detection sensor 205. The CPU 301 also detects the position of the document being transported along the transport path based on the detection results of the separation sensor 207 and the read sensor 210.

[0039] The optical motor 225 is a drive source for moving the surface reading unit 104 along the guide 109 in the sub-scanning direction. The optical motor 225 is driven and controlled by the CPU 301. The optical HP sensor 226 is a sensor for detecting when the surface reading unit 104 is in the home position (HP).

[0040] The front-side reading unit 104 has an LED 105 and a reading sensor 108 as described above. The front-side reading unit 104 receives the reflected light from the LED 105 onto the transported document using the reading sensor 108 and generates image data as the reading result. The back-side reading unit 212 has an LED 214 and a reading sensor 216 as described above. The back-side reading unit 212 receives the reflected light from the LED 214 onto the transported document using the reading sensor 216 and generates image data as the reading result. The image data output from the front-side reading unit 104 and the back-side reading unit 212 is temporarily stored in the image memory 305. The image data represents the read image.

[0041] The document reading process by the front reading unit 104 and the back reading unit 212 is controlled by the CPU 301. The CPU 301 prompts the front reading unit 104 and the back reading unit 212 to perform the document reading process when the read sensor 210 detects a document.

[0042] The image processing unit 306 performs corrections on the read data (read image) stored in the image memory 305 using predetermined image processing. The shading memory 307 is connected to the image processing unit 306 and holds shading correction data. The CPU 301 can read and write shading correction data held in the shading memory 307 via the image processing unit 306. The image transfer unit 304 transfers the image data processed by the image processing unit 306 to the controller 310 via the image transfer line 402.

[0043] The non-volatile memory 308 stores backup data such as model-specific adjustment values ​​and counter values ​​for the automatic document reader 10. The non-volatile memory 308 is a rewritable memory such as EEPROM (Electrically Erasable Programmable Read-Only Memory) or flash memory.

[0044] The controller 310 controls the overall operation of the image reading system, including the reader 40 and ADF 20. The controller 310 comprises a CPU 311, ROM 312, RAM 313, image transfer unit 314, and image memory 315. These are communicated together via a bus. The operation unit 90 is also connected to this bus.

[0045] The CPU 311 controls the operation of the controller 310 by executing a computer program stored in the ROM 312. The CPU 311 receives instructions and other inputs from the operation unit 90. The CPU 311 displays messages, read images, etc., on the display unit 91 of the operation unit 90. The RAM 313 provides a workspace for the CPU 311 when it performs processing. The image transfer unit 314 acquires image data from the image transfer unit 304 of the reader 40 via the image transfer line 402 and stores it in the image memory 315.

[0046] CPU 311 works in cooperation with CPU 301 to control image reading by the automatic document reader 10. To this end, CPU 311 sends and receives control data, such as instructions related to image reading control, to and from CPU 301 via the communication line 401.

[0047] For example, CPU 311 receives an instruction to start an image reading job from the operation unit 90 and sends a reading start instruction to CPU 301. The instruction to start an image reading job includes information such as whether to read in black and white or color, the reading resolution, the size of the document to be read, and the instruction to start the image reading job. The reading start instruction includes control information such as an instruction to start feeding the document and an instruction to read the document. The instruction to start feeding the document includes information on the reading resolution, and the instruction to read the document includes information on the size of the document to be read. In addition, CPU 311 receives information from CPU 301 that represents the status of the automatic document reader 10 and displays a message to the user on the operation unit 90 according to the status of the device.

[0048] (Automatic geometric adjustment process) In the ADF20's stream scanning mode, the document transport condition is crucial for reading images from the document as it is being transported. For example, the document transport speed and its tilt relative to the transport direction can cause deformation of the scanned image in the sub-scanning direction or misalignment of the image. Such deformation and misalignment of the scanned image are corrected by an automatic geometric adjustment process.

[0049] The automatic geometric adjustment process begins when the CPU 311 receives an automatic geometric adjustment instruction from the operation unit 90. Upon receiving the instruction, the CPU 311 sends a control signal to the CPU 301 to perform the automatic geometric adjustment. The CPU 301 receives this control signal and performs the automatic geometric adjustment. The CPU 301 notifies the CPU 311 of the result. The CPU 311 displays the result on the display unit 91 of the operation unit 90.

[0050] Figure 5 is an example diagram of an input screen for inputting automatic geometric adjustment instructions. The input screen is displayed on the display unit 91 of the operation unit 90 under the control of the CPU 311. The user inputs automatic geometric adjustment instructions using an input interface such as the operation key group 92, according to the display on the display unit 91.

[0051] Figure 5(a) shows the standby screen. When the "Settings / Registration" button on the standby screen is pressed, the display on the display unit 91 transitions to the settings / registration screen shown in Figure 5(b). When the "Adjust" button on the settings / registration screen is pressed, the display on the display unit 91 transitions to the adjustment selection screen shown in Figure 5(c).

[0052] On the adjustment selection screen, initially only the "Create Reference Coordinates" button is clickable. This is because ADF coordinate adjustment, described later, cannot be performed without reference coordinates. Reference coordinates are coordinates that indicate the shape (vertices) and position of the image when no deformation or positional shift has occurred, and they serve as the basis for the shape (vertices) and position of the image when performing automatic geometric adjustment. ADF coordinate adjustment is the process of aligning the shape (vertices) and position of the read image obtained by skimming with the reference coordinates. ADF coordinate adjustment is performed using reference coordinates. For this reason, on the adjustment selection screen, the "Create Reference Coordinates" button is pressed first, and then the "ADF Coordinate Adjustment" button becomes clickable.

[0053] When the "Create Reference Coordinates" button is pressed, CPU 311 sends a "Preparation for Creation Notification" to CPU 301 and displays the reference coordinate creation screen shown in Figure 5(d) on the display unit 91 of the operation unit 90. The reference coordinate creation screen in Figure 5(d) prompts the user to place the document to be adjusted on the document glass 101 and press the "Start" button.

[0054] When the user places the document on the document glass 101 and presses the "Start" button, CPU 311 sends a "Creation Start Notification" to CPU 301. CPU 301 performs the reference coordinate creation process in response to the "Creation Start Notification". When the reference coordinate creation is successfully completed, CPU 301 sends a "Creation OK Notification" to CPU 311 to inform it that the reference coordinate creation has been successfully completed. Upon receiving the "Creation OK Notification" from CPU 301, CPU 311 changes the display on the display unit 91 of the operation unit 90 to the ADF coordinate adjustment screen shown in Figure 5(e) and sends a "Preparation for Adjustment Notification" to CPU 301. The ADF coordinate adjustment screen instructs the user to place the document on the document tray 121 of the ADF 20.

[0055] When the user presses the "Start" button, CPU 311 sends a "Start Adjustment Notification" to CPU 301. CPU 301 performs ADF coordinate adjustment processing in response to the "Start Adjustment Notification". When CPU 301 completes the ADF coordinate adjustment successfully, it sends an "Adjustment OK Notification" to CPU 311 to inform it that the adjustment has been completed successfully. Upon receiving the "Adjustment OK Notification" from CPU 301, CPU 311 terminates the automatic geometric adjustment and returns the display on the display unit 91 of the operation unit 90 to the standby screen shown in Figure 5(a).

[0056] If the creation of reference coordinates does not complete successfully, CPU 301 sends a "creation NG notification" to CPU 311 indicating that the creation of reference coordinates could not be completed successfully. Upon receiving the "creation NG notification," CPU 311 changes the display on the display unit 91 of the operation unit 90 to the screen shown in Figure 5(f), prompting the user to correctly place the document on the document glass 101.

[0057] If the ADF coordinate adjustment does not complete successfully, CPU 301 sends an "adjustment NG notification" to CPU 311 indicating that the ADF coordinate adjustment could not be completed successfully. Upon receiving the "adjustment NG notification" from CPU 301, CPU 311 terminates the automatic geometric adjustment and transitions the display on the display unit 91 of the operation unit 90 to the screen shown in Figure 5(g), prompting the user to correctly place the document in the document tray 121 of the ADF 20.

[0058] Figure 6 is a flowchart representing this automatic geometric adjustment process. This process represents the operation of the CPU 301. This process is started when the "Settings / Registration" button is pressed on the standby screen in Figure 5(c), and then the "Adjust" button is pressed on the settings / registration screen in Figure 5(b). The adjustment selection screen in Figure 5(c) is displayed on the display unit 91 of the operation unit 90.

[0059] When CPU 301 receives a "Preparation for Creation Notification" from CPU 311 (S401:Y), it waits until the document is placed on the document glass 101 (S402:N). As described above, when the "Create Reference Coordinates" button is pressed from the adjustment selection screen, CPU 311 sends a "Preparation for Creation Notification" to CPU 301. The display on the display unit 91 of the operation unit 90 switches to the reference coordinate creation screen shown in Figure 5(d). When the "Start" button on the reference coordinate creation screen is pressed, CPU 301 detects that the document has been placed on the document glass 101. When the document is placed on the document glass 101 (S402:Y), CPU 301 creates reference coordinates using the "Create Reference Coordinates" process described below (S403).

[0060] If the creation of the reference coordinates is completed successfully (S404:Y), CPU301 issues a "Creation OK notification" to CPU311 (S405) and terminates the automatic geometric adjustment process. If the creation of the reference coordinates is not completed successfully (S404:N), CPU301 issues a "Creation NG notification" to CPU311 (S406) and terminates the automatic geometric adjustment process.

[0061] Upon receiving a "Creation OK notification" from CPU 301, CPU 311 transitions the display on the display unit 91 of the operation unit 90 to the ADF coordinate adjustment screen shown in Figure 5(e), and sends a "Preparation for Adjustment notification" to CPU 301. The ADF coordinate adjustment screen instructs the user to place the document on the document tray 121 of the ADF 20. If neither the "Preparation for Creation notification" nor the "Preparation for Adjustment notification" is sent (S401:N, S407:N), CPU 301 terminates the automatic geometric adjustment process.

[0062] When CPU301 receives a "preparation for adjustment notification" from CPU311 (S401:N, S407:Y), it waits until a document is placed on the document tray 121 of the ADF20 (S408:N). CPU301 detects that a document has been placed on the document tray 121 based on the detection results of the first document presence detection sensor 204 and the second document presence detection sensor 205. When a document is placed on the document tray 121 (S408:Y), CPU301 acquires a read image (adjustment coordinates) for ADF coordinate adjustment through the "acquisition of adjustment coordinates" process described later (S409).

[0063] If the adjustment coordinates are successfully acquired (S410:Y), the CPU 301 compares the reference coordinates with the adjustment coordinates using the "Correction Value Calculation" process described later, and calculates various correction values ​​to correct the geometric characteristics of the ADF-read image (S411). The ADF-read image is the reading result (read image) obtained by skimming using the ADF 20.

[0064] If the correction value calculation is completed successfully (S412:Y), CPU301 issues an "Adjustment OK notification" to CPU311 (S413) and terminates the automatic geometric adjustment process. If the acquisition of adjustment coordinates or the calculation of correction values ​​fails (S410:N or S412:N), CPU301 issues an "Adjustment NG notification" to CPU311 (S414) and terminates the automatic geometric adjustment process.

[0065] Figure 7 is an explanatory diagram of the "reference coordinate creation" process in S403. In the "reference coordinate creation" process, the automatic document reader 10 operates in reference coordinate creation mode (fixed reading mode) using the document glass 101. Figure 8 is a flowchart representing the "reference coordinate creation" process.

[0066] Here, the document used for adjustment is rectangular, and the coordinates of the four vertices of the document's outline are used as the "reference coordinates" and "adjustment coordinates." As shown in Figure 7(a), the document used for adjustment is placed with one of its vertices resting against the abutment portion 110 on the document glass 101.

[0067] When the "Create Reference Coordinates" button is pressed on the adjustment selection screen in Figure 5(c), CPU 301 is instructed by CPU 311 to prepare for the creation of reference coordinates (S601:Y). CPU 301 then executes the preparation settings for the creation of reference coordinates in the image processing unit 306. An example of the necessary device settings is as follows.

[0068] The CPU 301 lowers the threshold in known shading correction so that it can read darker than normal scanning operation, making the vertices and edges of the document more easily visible in the scanned image (S602). The CPU 301 sets the maximum standard size even larger than the maximum standard size for fixed scanning so that the four vertices of the document can be read even when the maximum standard size document is placed on the document glass 101 (S603).

[0069] The area enclosed by the dotted line in Figure 7(a) is the reading range for the reference coordinate creation mode. The reading range is set from downstream of the abutment section 110 in the transport direction to upstream of the upstream end of the maximum standard-sized document in the transport direction, so that a maximum standard-sized document placed on the document glass 101 after being abutted against the abutment section 110 can be read. The reading start position in the transport direction (sub-scanning direction) is set downstream of the abutment section 110 in the transport direction (towards the surface reading glass 102), and the reading end position is set upstream of the upstream end of the maximum standard-sized document in the transport direction that is abutted against the abutment section 110. This reading range is set to ensure that the document vertices corresponding to the abutment section 110, which serves as the reference point, are reliably read as images.

[0070] After the setup is complete, CPU 301 waits until the "Start" button on the reference coordinate creation screen in Figure 5(d) is pressed (S604:N). When the "Start" button is pressed (S604:Y), CPU 301 is instructed by CPU 311 to start the reference coordinate creation operation. CPU 301 reads the document placed on the document glass 101 in fixed reading mode and stores the reading result (read image) in the image memory 305 (S605).

[0071] Figure 7(b) is an example of a scanned image stored in the image memory 305 during processing S605. When the image surface of the original document is placed facing the document glass 101 and scanned, the coordinates of the scanned image are reversed in the direction of the main scan. That is, the document abutment point 110 (position of the black triangle) is located in the upper left when the original document is placed so that the image surface of the original document faces the document glass 101 (Figure 7(a)), but in the scanned image it is located in the lower left (Figure 7(b)). The position of the black triangle is used as the reference point for the image position in the geometric characteristic adjustment operation.

[0072] To read an image covering a wider area than the normal maximum reading range, four vertices of the original document are detectable within the read image. The CPU 301 detects the coordinates of these four vertices using the coordinate information detection process described later and stores them in the RAM 303 (S606). In the read image shown in Figure 7(b), the detection results of the coordinates of the vertex (reference point) P0 and the other three vertices P1, P2, and P3 in the main scanning direction and sub-scanning direction are stored in the RAM 303. The coordinates of the reference point P0 are (XP0, YP0), and the coordinates of the three vertices P1, P2, and P3 are (XP1, YP1), (XP2, YP2), and (XP3, YP3).

[0073] Figure 9 is an explanatory diagram of the "adjustment coordinate acquisition" process in S409. In the "adjustment coordinate acquisition" process, the automatic document reader 10 operates in adjustment coordinate acquisition mode (swiping mode) using the ADF 20. Figure 10 is a flowchart representing the "adjustment coordinate acquisition" process.

[0074] The document used for adjustment is placed on the document tray 121 of the ADF 20, as illustrated in Figure 9(a), and the main scanning direction is restricted by two width restriction plates 125. The CPU 301 detects that a document has been placed on the document tray 121 based on the detection results of the first document presence detection sensor 204 and the second document presence detection sensor 205 (S801).

[0075] When the "ADF Coordinate Adjustment" button is pressed on the adjustment selection screen in Figure 5(c), CPU 301 is instructed by CPU 311 to prepare for acquiring adjustment coordinates (S802:Y). CPU 301 then configures the image processing unit 306 to prepare for acquiring adjustment coordinates. An example of the necessary preparation settings is as follows.

[0076] The CPU 301, as in the reference coordinate creation mode, lowers the threshold in the known shading correction so that it can read darker than normal in order to make the vertices and edges of the document more visible in the scanned image (S803). The CPU 301 sets the image reading range to be even larger than the maximum standard size set in the skimming mode so that the four vertices of the document can be read even when the maximum standard size document is placed on it (S804). The CPU 301 sets the coordinates of the scanned image on the back side to be reversed in the main scanning direction (S805).

[0077] The S805 process will now be explained. As shown in Figure 3, the front reading unit 104 and the back reading unit 212 are located at different positions in the transport direction. During transport of the document from the reading position of the front reading unit 104 to the reading position of the back reading unit 212, the condition of the document may change due to skew or other reasons. Therefore, it is preferable that the scanned image of the front and the scanned image of the back be adjusted independently.

[0078] The reference point for measuring the document position is the position of the black triangle in the scanned image of Figure 7(b) read in fixed-read mode. Figure 9(a) shows the same document as in Figure 7(a) placed on the document tray 121. Documents placed in this manner are scanned in flow-read mode. The scanned image of the front side is Figure 9(b), and the reference point is in the same position as in Figure 7(b). In the scanned image of the back side, as shown in Figure 9(c), the position of the black triangle is at the top of the figure. In order to compare the difference with the reference coordinates in Figure 7(b), the scanned image of the back side needs to be read inverted in the main scanning direction (up and down direction in Figure 9(c)) (Figure 9(d)). For this purpose, processing in S805 sets the coordinates of the scanned image of the back side in the main scanning direction to be inverted.

[0079] After the setup is complete, CPU 301 waits until the "Start" button on the ADF coordinate adjustment screen in Figure 5(e) is pressed (S806:N). When the "Start" button is pressed (S806:Y), CPU 301 is instructed by CPU 311 to start the coordinate acquisition operation for adjustment. CPU 301 reads both the front and back sides of the document placed on the document tray 121 in a continuous reading mode, and stores the reading results for the front side (front reading image) and the reading results for the back side (back reading image) in the image memory 305 (S807).

[0080] Figure 9(b) shows an example of a front-side scanned image stored in the image memory 305, and Figure 9(d) shows an example of a back-side scanned image stored in the image memory 305. Because the image is read over a wider area than the normal maximum scanning range, the four vertices of the document can be detected within the scanned image even if the document is skewed. The CPU 301 detects the four vertices of both the front-side scanned image and the back-side scanned image using the coordinate information detection process described later and stores them in the RAM 303 (S808).

[0081] In the surface reading image shown in Figure 9(b), the detection results of the coordinates of the reference point F0 and the other three vertices F1, F2, and F3 in the main scanning direction and sub-scanning direction are stored in RAM303. The coordinates of the reference point F0 are (XF0, YF0), and the coordinates of the three points F1, F2, and F3 are (XF1, YF1), (XF2, YF2), and (XF3, YF3). Similarly, in the back-side read image shown in Figure 9(d), the detection results of the coordinates of the reference point B0 and the other three vertices B1, B2, and B3 in the main scan direction and sub-scan direction are stored in RAM 303. The coordinates of the reference point B0 are (XB0, YB0), and the coordinates of the three points B1, B2, and B3 are (XB1, YB1), (XB2, YB2), and (XB3, YB3).

[0082] Figure 11 is an explanatory diagram of the coordinate information detection process in S606 of Figure 8 and S808 of Figure 10. Here, we will explain using the case where vertices A0 to A3 are detected as coordinate information from the read image stored in the image memory 305 as an example.

[0083] Figure 11(a) is an explanatory diagram of edge detection of a scanned image read in the reference coordinate creation mode of Figure 8 and stored in the image memory 305. In fixed reading in reference coordinate creation mode, the document is placed with its leading edge abutting against the stopper 110. Therefore, the boundary between the outer edge of the device, indicated by the diagonal lines, and the document glass 101 is also detected as an edge. The stopper 110 is provided along the edge of the document glass 101 and is fixed in both the main scanning direction and the sub-scanning direction. Therefore, in the case of fixed reading, vertex detection is performed with the position of the stopper 110 as the reference, within a range that does not include the outer edge of the device, indicated by the diagonal lines.

[0084] Figures 11(b) and (c) are explanatory diagrams for edge detection of a scanned image read in the adjustment coordinate acquisition mode of Figure 10 and stored in the image memory 305. Vertex detection of the scanned image in the scrolling mode is performed over the entire range of the scanned image. In Figures 11(b) and 11(c), the skew direction of the scanned document is reversed.

[0085] Figure 12 is a flowchart representing the coordinate information detection process. Note that in the reference coordinate creation mode and adjustment coordinate acquisition mode of this embodiment, the document is scanned darker than during normal scanning, as described above. Therefore, the scanned image is prone to showing shadows at the edges of the document.

[0086] The CPU 301 binarizes the entire scanned image stored in the image memory 305 and extracts the edges that form the outer frame of the original document (S901). In this example, the binarization process is performed on the scanned image stored in the image memory 305, but the scanned image may be binarized by the image processing unit 306 before being stored in the image memory 305.

[0087] The CPU 301 extracts an image of one line in the main scanning direction (the image corresponding to L in Figure 11(b)) from the read image in order from the downstream side in the transport direction and performs edge detection. Here, if the read image to be searched is a fixed-read image as shown in Figure 11(a) (S902:Y), the CPU 301 targets edge detection from line L=Lt corresponding to the stopper 110 (S903). For this reason, lines to the left of line Lt are not targeted for detection. If the read image to be searched is a flow-read image as shown in Figure 11(b) (S902:N), the CPU 301 starts edge detection from the edge of the read image, i.e., line L=0 (the leftmost line) (S904).

[0088] Next, the CPU 301 performs edge detection processing while moving line L one line at a time in the sub-scan direction until it detects the first edge (S905:N, S906). When the first edge is detected (S905:Y), the CPU 301 determines the edge position of line L where the first edge was detected to be the coordinates of vertex A0 shown in Figure 11(b) and stores it in the RAM 303 (S907).

[0089] The CPU 301 further moves line L in the sub-scanning direction and performs edge detection processing (S908:N, S909). As a result, the CPU 301 detects two edges on each line L in the main scanning direction (S908:Y). The two detected edges are designated as edges E0 and E1, as shown in Figure 11(b). The CPU 301 counts the number of pixels D between edges E0 and E1 for each line L, as shown in Figure 11(b) (S910). While the number of pixels D between edges is increasing (S911:Y), the CPU 301 repeatedly performs the process in S910. During this time, the edge detection range has not yet reached the next vertex (vertex A1 in this case).

[0090] When the number of pixels D stops increasing (remains constant or decreases) (S911:N), the CPU 301 determines that the detection position has reached line L1 in Figure 11(b). The CPU 301 determines the coordinates of the second vertex A1 at the edge on line L1 that has moved towards the center in the main scanning direction and stores it in the RAM 303 (S912).

[0091] After passing line L1, CPU 301 continues to perform edge detection processing line by line while counting the number of pixels D between edges (S913, S914:N). When the number of pixels D decreases (S914:Y), CPU 301 determines that line L2 in Figure 11(b) has been reached. CPU 301 determines the coordinates of the third vertex A2 of the edge that has moved more towards the center in the main scanning direction on line L2 and stores it in RAM 303 (S915).

[0092] The CPU 301 moves line L one line at a time in the sub-scanning direction and performs edge detection processing until only one edge remains to be detected (S916:N, S917). When only one edge has been detected (S916:Y), the CPU 301 determines the position of the detected edge as the coordinates of vertex A3 and stores it in the RAM 303 (S918).

[0093] CPU301 performs edge detection processing up to the last line of the read image (S919:N). When the edge detection processing up to the last line of the read image is completed (S919:Y), CPU301 checks the number of vertices detected up to that point by referring to RAM303 (S920). If all four vertices A0 to A3 have been detected (S920:Y), CPU301 completes the coordinate information detection process successfully. If the number of detected vertices is not four (S920:N), CPU301 terminates the coordinate information detection process with an error. The coordinates of the vertices are detected similarly for the read image in Figure 11(c). The number of detected vertices may not be four if, for example, the document is skewed and the vertices are not read correctly (are not in the reading range), or if the edges of the read image are unclear.

[0094] In the image read in reference coordinate creation mode, the area from line L=Lt onwards becomes the target for edge detection, as shown in the processing of S903. Edges are detected in the vicinity of line Lt in this read image. The positions of the detected edges are stored in RAM303, which is determined to be the coordinates of vertices P0 and P1 in Figure 11(a). Subsequently, CPU301 performs detection processing while moving line L one line at a time in the sub-scan direction until no more edges are detected. CPU301 determines the positions of the edges where no more edges are detected to be the coordinates of vertices P2 and P3 in Figure 11(a) and stores them in RAM303.

[0095] In the process shown in Figure 8, the four vertices detected by the coordinate information detection process are mapped to vertices Pn(XPn,YPn) (n=0~3) obtained from the fixed reading (see Figure 7(b)). In the process shown in Figure 10, the four vertices A0~A3 detected by the coordinate information detection process are mapped to vertices Fn(XFn,YFn) (n=0~3) obtained from the skimming reading (see Figures 9(b) and (d)).

[0096] The correspondence between vertices A0 to A3 and, for example, vertices F0 to F3 depends on the tilt direction of the original document. Using a surface scan image obtained by skimming as an example, we will explain which of vertices F0 to F3 in Figure 9(b) correspond to vertices A0 to A3.

[0097] In the case of a surface scan image read from an obliquely oriented document as shown in Figure 11(b), the first detected vertex A0 corresponds to the reference point F0 in the surface scan image in Figure 9(b). Similarly, vertex A1 corresponds to vertex F1, vertex A2 to vertex F2, and vertex A3 to vertex F3.

[0098] Figure 11(c) is a surface scan image read from a document with the opposite direction of oblique scanning compared to Figure 11(b). In this case, the vertex A0, which is the first to be detected in the coordinate information detection process in Figure 12, corresponds to F1, which is located away from the reference point F0 in the main scanning direction in the surface scan image of Figure 9(b). The other three points correspond to vertex A1 as vertex F0, vertex A2 as vertex F3, and vertex A3 as vertex F2, respectively.

[0099] Based on the coordinates of each vertex set in this way, image correction parameters for the front and back surface images are obtained. The CPU 301 determines the corresponding coordinates based on the front-to-back positional relationship between vertex A0 and vertex A1, and stores the coordinate information in the RAM 303.

[0100] (Calculation of correction value) Figure 13 is an explanatory diagram of the correction value calculation process, which calculates various geometric adjustment parameters that are image correction parameters for the front-side and back-side-side-read images. The correction value calculation process is the process S411 in Figure 6. Geometric adjustment parameters are obtained from the difference between the reference coordinates and the adjustment coordinates. Here, the following five types of correction values ​​are derived as geometric adjustment parameters. (1) Main scanning direction position shift correction value ΔX (2) Slope correction value θ (3) Sub-scan direction position shift correction value ΔY (4) Sub-scan magnification correction value α (5) Right-angle correction value ΔF

[0101] These five correction values ​​are derived independently for both the front-side and back-side images. The following explanation uses the process of calculating the correction values ​​by comparing the coordinates F0 to F3 of the front-side image with the reference coordinates P0 to P3 as an example.

[0102] Figure 13(a) shows the difference between the reference coordinates P0-P3, which are acquired before the correction value calculation process in S411 of Figure 6 and stored in RAM 303, and the adjustment coordinates F0-F3 of the surface reading image, which are shown on the same plane. The solid line indicates the edge position of the reading image (hereinafter referred to as the reference image) obtained by fixed reading in the process of Figure 8, which is composed of connecting the reference coordinates P0-P3. The dotted line indicates the edge position of the surface reading image (hereinafter referred to as the adjustment image) composed of connecting the adjustment coordinates F0-F3.

[0103] The adjusted image has positional shifts and tilts in the transport direction compared to the reference image. The CPU 301 calculates the five correction values ​​described above using the following procedure, with the goal of matching the edge positions of the adjusted image to those of the reference image.

[0104] (1) Main scanning direction position shift correction value ΔX In Figure 13(a), the CPU 301 first compares the main scan direction coordinates XP0 and XF0 of the reference point P0 (XP0, YP0) of the reference image and the vertex F0 (XF0, YF0) of the adjustment image. In this example, XF0 is shifted in the main scan direction relative to XP0 by ΔX = |XF0 - XP0|. The amount of shift ΔX is stored in the RAM 303 as a main scan direction position shift correction value.

[0105] To compensate for the displacement ΔX, the four vertices F0 to F3 in the adjustment image are uniformly moved by the displacement ΔX in the main scanning direction, as indicated by the arrows in Figure 13(a). If each vertex in the adjustment image after the movement is called Fan (n=0 to 3), then the X and Y coordinates (XFan, YFan) of vertex Fan are expressed by the following equations. XFan = XFn - ΔX YFan = YFn (n = 0 to 3)

[0106] The CPU 301 temporarily stores the coordinates of each vertex Fan in the adjustment image after correcting the displacement amount ΔX in the RAM 303. This correction causes the position of the adjustment image to be shifted in the main scan direction by the displacement amount ΔX.

[0107] (2) Slope correction value θ The CPU 301 calculates a tilt correction value θ to match the edges P0-P1 in the main scanning direction of the reference image, which include the reference point, with the edges Fa0-Fa1 in the main scanning direction of the adjustment image. Figure 13(b) shows the difference between the vertices P0-P3 of the reference image and the vertices Fa0-Fa3 of the adjustment image after applying the main scanning direction shift correction value (shift amount ΔX). The angle θ between each line of edge P0-P1 and edge Fa0-Fa1 represents the "tilt" of the adjustment image. The tilt correction value θ is expressed by the following formula. θ = tan -1 {(XFa1-XFa0) / (YFa1-YFa0)}

[0108] The tilt correction amount θ is stored in RAM303. The CPU301 rotates vertices Fa1, Fa2, and Fa3 of the four vertices Fan(n=0~3) of the adjusted image, excluding the reference point Fa0, by an angle θ around the reference point Fa0, in order to correct using the tilt correction amount θ. The coordinates after the rotation are denoted as Fbn(n=0~3). The X coordinate XFbn and Y coordinate YFbn of Fbn are expressed by the following equations. XFbn=cosθ(XFa1-XFa0)-sinθ(YFa1-YFa0) YFbn=sinθ(XFa1-XFa0)+cosθ(YFa1-YFa0) n:1~3

[0109] The CPU 301 temporarily stores the coordinates of each vertex Fbn of the adjustment image after correcting the tilt correction value θ in the RAM 303. This correction makes the leading edge in the sub-scan direction of the adjustment image parallel to the leading edge in the sub-scan direction of the reference image.

[0110] (3) Sub-scan direction position shift correction value ΔY Figure 13(c) shows the difference between the vertices P0 to P3 of the reference image and the vertices Fb0 to Fb3 of the adjusted image after applying the tilt correction amount θ. The CPU 301 compares the coordinates YP0 and YFb0 in the subscan direction of the reference point P0 (XP0, YP0) of the reference image and the vertex Fb0 (XFb0, YFb0) of the adjusted image. In this example, YFb0 is shifted in the subscan direction relative to YP0 by ΔY = |YFb0 - YP0|. The shift amount ΔY is stored in the RAM 303 as a subscan direction position shift correction value.

[0111] CPU301 moves the four vertices Fb0 to Fb3 of the adjustment image by the amount of displacement ΔY in the sub-scan direction to correct the displacement ΔY. If each vertex of the adjustment image after the movement is Fcn (n=0 to 3), then the X and Y coordinates (XFcn, YFcn) of vertex Fcn are expressed by the following equations. XFcn=XFbn YFcn = YFn - ΔY (n = 0 to 3)

[0112] The CPU 301 temporarily stores the vertices Fcn of the adjustment image, after correcting the displacement amount ΔY, in the RAM 303. This correction causes the position of the adjustment image to be shifted in the sub-scan direction by the displacement amount ΔY.

[0113] (4) Sub-scan magnification correction value α Figure 13(d) shows the difference between the vertices P0 to P3 of the reference image and the vertices Fc0 to Fc3 of the adjusted image after applying the sub-scan direction position shift correction (shift amount ΔY). Due to the corrections using the correction values ​​(1) to (3), the line P0-P1 in the main scan direction connecting the reference point P0 and vertex P1 of the reference image and the line FC0-FC1 in the adjusted image corresponding to line P0-P1 overlap. In this state, as shown in Figure 13(d), if there is a shift in the sub-scan direction (line P2-P3 and line Fc2-Fc3), there will be a difference in the magnification α in the sub-scan direction.

[0114] The sub-scanning direction magnification correction value α for the vertices of the adjusted image is expressed as α = 100 × (YFc2 - YFc0) / (YP2 - YP0), where the magnification of the vertices of the reference image is 100. The sub-scanning direction magnification correction value α is stored in RAM303 as a magnification correction value.

[0115] CPU301 corrects the right-hand vertices Fc2 and Fc3 in Figure 13(d) to correct the sub-scan direction magnification correction value α. After the correction, the sub-scan direction length of the adjusted image matches the sub-scan direction length of the reference image. The coordinates (XFdn, YFdn) of the corrected vertices Fd2 and Fd3 are expressed by the following formula. XFdn=XFcn YFdn=YFc0+(YFcn-YFc0) / α (n=2~3)

[0116] (5) Right-angle correction value ΔF Figure 13(e) shows the difference between the vertices P0 to P3 of the reference image and the vertices Fd0 to Fd3 of the adjusted image after applying the sub-scan direction magnification correction value α. The difference in the main scan direction between the vertex P2 of the reference image and the vertex Fd2 of the adjusted face image, indicated by the arrow, is considered to be the right-angle error ΔF of the scanned image due to the oblique feed of the document. Comparing vertex P2 and vertex Fd2, ΔF = |XFd2 - XP2|. The right-angle error ΔF is stored in RAM303 as a right-angle correction value.

[0117] CPU301 moves vertices Fd2 and Fd3 in Figure 13(e) to correct the right-angle correction amount. The coordinates Fe2 and Fe3 after the move are expressed by the following equations. XFen=XFdn YFen = YFdn - ΔF (n = 2~3)

[0118] The processes (1) to (5) yield geometric adjustment parameters (ΔX, θ, ΔY, α, ΔF) to align the four vertices of the surface reading image (adjustment image) with the four vertices of the reference image. If the calculation is completed successfully up to this point, the CPU 301 saves these parameters (correction values), which were temporarily stored in the RAM 303, as backup data in the non-volatile memory 308.

[0119] The above explanation focused on the front-side image, but geometric adjustment parameters (correction values) are also acquired for the back-side image in a similar manner and stored in the non-volatile memory 308 as parameters independent of the correction values ​​obtained from the front-side image. In this case, as explained in Figure 9(c), the correction values ​​are acquired using coordinates obtained by inverting the back-side image in the main scanning direction.

[0120] The correction values ​​obtained as described above are used to correct the scanned image when scanning a document in normal skimming mode. The corrected scanned image is sent to the controller 310. Correction is performed, for example, by setting each correction value in the image processing unit 306. The image processing unit 306 performs the correction before writing the scanned image to the image memory 305, or before sending the scanned image written to the image memory 305 to the controller 310. Alternatively, the correction may be performed on the controller 310 side. In this case, the correction values ​​are set in the CPU 311, and the correction is performed by the CPU 311.

[0121] (Input of reference coordinates) In the above explanation, the reference coordinates are obtained based on the scanned image of a document read in fixed-reading mode using the document glass 101, and the correction values ​​are derived based on the scanned image of the same document read in flow-reading mode using the ADF 20. The reference coordinates can also be obtained by other methods.

[0122] For example, if you always use documents of the same size, you don't need to generate reference coordinates each time. You can back up the reference coordinates in non-volatile memory 308, and then afterwards, you can simply adjust the ADF coordinates and compare them with the backed-up contents (correction values) to derive new correction values ​​(parameters). Also, if the document size is a standard size such as A4 or is known from the beginning, the reference coordinates may be registered according to the size entered from the operation unit 90.

[0123] Figure 14 is an example of the reference coordinate registration screen. When "Create Reference Coordinates" is selected from the adjustment selection screen in Figure 5(c), the reference coordinate registration screen in Figure 14(a) is displayed on the display unit 91. On the reference coordinate registration screen, the "Create Reference Coordinates" button and the "Enter Coordinates Numerically" button are selectable. If the "Create Reference Coordinates" button is selected, as described above, the reference coordinates are obtained based on the scanned image of the document scanned in fixed reading mode using the document glass 101.

[0124] When the "Enter coordinates numerically" button is selected, the reference coordinates are numerically entered by the operation unit 90. Therefore, when the "Enter coordinates numerically" button is selected, the screen of the display unit 91 transitions to the reference coordinate input screen shown in Figure 14(b). On the reference coordinate input screen, it is possible to select a predetermined size from the list of standard sizes or to numerically input the size in millimeters. In the example in Figure 14(b), numerical input can be set for the length in the main scanning direction and the sub-scanning direction, assuming a rectangle. In addition, by directly inputting the coordinates of the three vertices P1 to P3, excluding the reference point P0, of the four vertices of the reference image, it is also possible to register the coordinates of documents other than rectangles, such as parallelograms and trapezoids.

[0125] Once the input of the reference coordinates is complete, the display unit 91 transitions to the adjustment selection screen shown in Figure 14(c). From the adjustment selection screen, it is possible to continue with ADF coordinate adjustment, or to end the work after only creating the reference coordinates. The reference coordinates are transmitted from CPU 311 to CPU 301 and stored in the non-volatile memory 308.

[0126] When the "Adjust" button is pressed from the settings / registration screen in Figure 5(b) while the reference coordinates are retained, the display unit 91 of the operation unit 90 will show the adjustment selection screen in Figure 14(c) instead of the adjustment selection screen in Figure 5(c). Therefore, it is not necessary to create the reference coordinates again, and ADF coordinate adjustment can be performed directly.

[0127] The above describes a configuration in which the CPU 301 in the reader 40 performs vertex extraction and correction value acquisition from the read image. These processes may also be performed by the CPU 311 of the controller 310. In this case, the read image is transferred directly to the controller 310, just as in normal image reading operations. The CPU 311 analyzes the transferred read image and performs vertex extraction and correction acquisition.

[0128] As explained above, correction values ​​are generated to adjust the geometry of the scanned image obtained by scanning a document using a fixed-position scanning method, based on the reference coordinates obtained from the scanned image obtained by scanning a document using a fixed-position scanning method. Subsequently, the scanned image obtained by scanning is corrected using these correction values ​​to obtain a geometrically adjusted scanned image. For this reason, it is possible to easily correct positional shifts, skew, and other geometric deviations in the scanned image obtained by scanning a document using a scanned image without having to prepare a document with a format for image position adjustment printed on it.

[0129] (Image forming system) Figure 15 is a diagram of the configuration of an image forming system comprising the automatic document reader 10, an image forming apparatus, and an operation unit 90. In this image forming system 1, a reader 40 is provided on top of the image forming apparatus 50, and an ADF 20 is provided on top of the reader 40. The operation unit 90 is provided on the front of the automatic document reader 10 and the image forming apparatus 50. Such an image forming system 1 realizes a high-performance image forming apparatus such as a copier, a multifunction printer, or an MFP (Multifunction Peripheral). The image forming apparatus 50 has an image forming unit 51 for forming an image on paper.

[0130] When copying a document, the image data generated by the reader 40 reading the document is transmitted to the image forming apparatus 50. This image data is generated by correcting the scanned image obtained by flow reading using the ADF 20 with correction values. The image forming apparatus 50 forms an image on the paper using the image forming unit 51 based on the image data acquired from the reader 40. The image forming apparatus 50 forms the image on the paper using a method commonly used for image formation, such as electrophotography or inkjet.

[0131] The image forming apparatus 50 may be connected to an external device via a predetermined network. In this case, the image forming apparatus 50 may transmit image data acquired from the reader 40 to the external device.

Claims

1. A document transport means having a document tray on which a document is placed, and transporting the document from the document tray, A reading means for reading the document being transported by the document transport means, The invention is characterized by comprising: a derivation means for detecting a first vertex of the document from the reading result of the document by the reading means, and deriving a correction value for correcting the difference between the detected first vertex and a second vertex of a reference image. Image reading device.

2. A storage means for storing the aforementioned correction value, The system is characterized by comprising an image processing means that corrects the reading result of the document being transported by the document transport means based on the correction value stored in the storage means, The image reading device according to claim 1.

3. The reading means sets the reading range of the document to be even larger than the maximum standard size that can be read by the reading operation using the document transport means. The derivation means is characterized by detecting the first vertex from the reading result of the document read within a reading range set to be even larger than the maximum standard size. The image reading device according to claim 1.

4. The aforementioned document is rectangular, The derivation means is characterized by detecting four vertices of the original document as the first vertices and deriving the correction value from the four first vertices and the four second vertices. The image reading device according to claim 1.

5. The derivation means is characterized by deriving the correction value based on the coordinates of the first vertex and the coordinates of the second vertex. The image reading device according to claim 1.

6. It is equipped with a document glass on which a document can be placed, The reading means reads the document placed on the document glass, The derivation means is characterized by detecting the second vertex based on the reading result of the document placed on the document glass by the reading means. The image reading device according to claim 1.

7. The reading means sets the reading range of the document to be even larger than the maximum standard size that can be read by the reading operation using the document glass, The derivation means is characterized by detecting the second vertex from the reading result of the document read within a reading range set to be even larger than the maximum standard size. The image reading device according to claim 6.

8. The system is characterized by comprising an input means for inputting information about the second vertex, The image reading device according to claim 1.

9. The reading means includes a first reading means for reading the first side of the original document, and a second reading means for reading a second side of the original document that is different from the first side. The derivation means is characterized by deriving a first correction value for correcting the difference between the first vertex and the second vertex detected from the reading result of the document by the first reading means, and a second correction value for correcting the difference between the first vertex and the second vertex detected from the reading result of the document by the second reading means. The image reading device according to claim 1.

10. Image reading device and The system includes an image forming means that forms an image on paper based on image data representing a document read by the aforementioned image reading device, The aforementioned image reading device is A document transport means having a document tray on which a document is placed, and transporting the document from the document tray, A reading means for reading the document being transported by the document transport means, A derivation means for detecting a first vertex of the document from the reading result of the document by the reading means, and for deriving a correction value for correcting the difference between the detected first vertex and a second vertex of a reference image, The system is characterized by comprising an image processing means that corrects the reading result of the document transported by the document transport means by the reading means based on the correction value and generates the image data, Image forming system.

Images