Image reading device, control method thereof, and program

Abstract

To implement highly accurate image connecting processing between line image sensors without being affected by a skew of a document during conveyance.SOLUTION: One embodiment of the present invention relates to an image reading apparatus including: a plurality of line image sensors arrayed in a first direction; conveyance means configured to convey a document in a second direction intersecting with the first direction; calculation means configured to calculate a correction value for correcting a deviation at the time of connecting image data obtained by each of the plurality of line image sensors on the basis of read data obtained by reading, by the plurality of line image sensors, a chart on which a plurality of dot patterns are printed and which is conveyed by the conveyance means; measurement means configured to measure an amount of conveyance in the first direction in a case where the document is conveyed in the second direction by the conveyance means; and adjusting means configured to adjust the correction value on the basis of the amount of conveyance in the first direction in a case where the read data is obtained and the amount of conveyance in the first direction in a case where the image data is obtained.SELECTED DRAWING: Figure 18

Description

[Technical Field]

[0001] The present disclosure relates to image correction in an image reading device having a line image sensor. [Background technology]

[0002] Image reading devices for large-format documents generally employ a configuration that uses multiple small-sized line image sensors, which offers significant cost benefits. Because multiple line image sensors are used, a process is required to stitch together the data read by each line image sensor. In this process, if there is a mechanical mounting tolerance for the line image sensor, the position of the line image sensor may shift in the X or Y direction, or the tilt of the line image sensor may deviate by an angle Φ from the ideal. This results in errors in the stitching process.

[0003] Figure 23 shows an example of a typical configuration of an image reading device using multiple line image sensors and the resulting reading. As shown in Figure 23(a), an original 110 is conveyed by upstream original conveying rollers 107 and downstream original conveying rollers 108, and the original 110 is read by multiple line image sensors 2304. The results read by the multiple line image sensors 2304 are stitched together at stitching positions 113, but if there is a positional misalignment 2301 between the line image sensors 2304, an error occurs in the stitching positions. For example, as shown in Figure 23(b), when a straight line pattern 2302 is read, due to the positional misalignment of the line image sensors 2304, the resulting line is read as a jagged line 2303, which results in a misalignment in the stitching positions when the line image sensors 2304 are joined together.

[0004] To address the above issue, a process called calibration is used to calculate in advance the error components that occur during reading due to misalignment of the line image sensor, thereby achieving accurate stitching. For example, to perform calibration, a document containing a specific pattern is used, and the pattern is read. After that, the stitching position is calculated based on the position data of the specific pattern that was read.

[0005] Furthermore, since the document 110 is read while being transported, any change in the transport speed of the document will result in errors in the joining process. To address this issue, a means for detecting the transport position of the document has been provided, enabling the process of joining images read by each line image sensor 2304 with high precision even when the transport speed of the document changes (Patent Document 1). [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Japanese Patent Application Laid-Open No. 2015-119388 Summary of the Invention [Problem to be solved by the invention]

[0007] Conventionally, even when the transport speed of a document changes, a highly accurate stitching process has been realized by deriving the position of the document in the transport direction, as in the image reading device of Patent Document 1. However, in reality, the amount of skew caused by roller meshing during document transport differs for each read, which causes errors in the stitching process and results in image misalignment.

[0008] Therefore, an object of the present disclosure is to realize highly accurate image stitching processing between line image sensors regardless of skew of the document during transport. [Means for solving the problem]

[0009] One embodiment of the present invention is a scanning device including: a plurality of line image sensors arranged in a staggered pattern along a first direction; a conveying unit that conveys a chart or document on which a plurality of dot patterns are printed in a second direction perpendicular to the first direction; a deriving means for deriving a first direction transport amount of transport in the first direction associated with transport in the second direction when the chart or the document is transported in the second direction by the transport means, the deriving means deriving the first direction transport amount for the chart and the first direction transport amount for the document; a skew amount calculating means for calculating a skew amount in the first direction of each of the plurality of line image sensors, the skew amount calculating means calculating the skew amount during calibration processing based on the skew amount calculated by linear approximation using reference coordinates of a reference dot pattern and coordinates of dot patterns surrounding the reference dot pattern in read data of the transported chart; and a correction value calculating means for calculating a correction value for a joining position of images when read by each of the plurality of line image sensors based on the skew amount, the correction value calculating means calculating a correction value for a joining position of images when read by each of the plurality of line image sensors based on the skew amount. a correction value calculation means for deriving, by coordinate conversion, processing target coordinates for each of a first line image sensor and a second line image sensor whose positions overlap in the first direction, and calculating a first offset value for the first direction and a second offset value for the second direction based on a difference in the first direction and a difference in the second direction between a first processing target coordinate for the first line image sensor and a second processing target coordinate for the second line image sensor, and calculating the correction value based on the first offset value and the second offset value; and an adjustment means for adjusting the correction value calculated by the correction value calculation means, which calculates a skew correction amount based on a difference between the first direction transport amount for the chart and the first direction transport amount for the document, adjusts the correction value based on the skew correction amount, and writes the adjusted correction value to a memory. of and when reading the adjusted correction values ​​written in the memory, a process of stitching together images obtained when the original is read by each of the plurality of line image sensors is performed by a method of reading while changing an access location of the memory or a method of physically adjusting an inclination of each of the plurality of line image sensors. The image reading device is characterized by the above. [Effects of the Invention]

[0010] According to the present disclosure, highly accurate image stitching between line image sensors can be achieved regardless of skew of the document during transport. [Brief explanation of the drawings]

[0011] [Figure 1] FIG. 1 is a diagram showing the overall configuration of an image reading apparatus according to a first embodiment; [Figure 2] FIG. 1 is a block diagram showing a hardware configuration of an image reading apparatus according to a first embodiment; [Figure 3] 1 is a flowchart of a calibration process of an image reading device according to a first embodiment; [Figure 4] FIG. 10 is a diagram showing a correction value calculation processing pattern; [Figure 5] 10 is a flowchart illustrating a correction value calculation process according to the first embodiment. [Figure 6] FIG. 1 is a diagram showing a circular dot pattern to be read in the first embodiment; [Figure 7] 10 is a flowchart of a process for deriving the center coordinates of a circular dot pattern in the first embodiment. [Figure 8] FIG. 10 is a diagram showing a pattern used to derive the tilt angle of a CIS in the first embodiment. [Figure 9] FIG. 10 is a diagram showing an algorithm used to derive the tilt angle of a CIS in the first embodiment. [Figure 10] 10 is a flowchart illustrating calculation of a correction value accompanied by derivation of a tilt angle of a CIS in the first embodiment. [Figure 11] FIG. 10 is a diagram for explaining coordinate transformation based on the tilt of a CIS in the first embodiment. [Figure 12]FIG. 10 is a diagram for explaining derivation of a connection position between CISs in the first embodiment; [Figure 13] 1 is a flowchart showing a flow of deriving a correction value for correcting a joint position in the first embodiment; [Figure 14] Diagram showing the arrangement of circular dot patterns when there is misalignment when placing the original [Figure 15] FIG. 10 is a diagram for explaining a measure to prevent erroneous derivation of a connection position in the first embodiment; [Figure 16] 1 is a diagram showing the internal structure of a CIS and the arrangement of circular dot patterns in a first embodiment; [Figure 17] Schematic diagram showing the effect of skew [Figure 18] FIG. 10 is a schematic diagram showing a means for deriving a transport amount in the first embodiment. [Figure 19] 10 is a flowchart of a calibration process for suppressing the influence of skew in the first embodiment. [Figure 20] 10 is a flowchart of a normal reading process for suppressing the influence of skew in the first embodiment. [Figure 21] 10 is a flowchart of a process for deriving an X-direction transport amount in the first embodiment. [Figure 22] Flowchart of skew correction processing in the first embodiment [Figure 23] Typical configuration of an image reading device with multiple line image sensors and an example of the reading result DETAILED DESCRIPTION OF THE INVENTION

[0012] [First embodiment] In this embodiment, by reading a specific pattern in advance, the inclination of each line image sensor and the connecting position of image data between each line image sensor are derived, which are used to accurately perform connecting processing between multiple line image sensors.

[0013] Specifically, the system uses coordinates within the pattern to derive the inclination of each line image sensor and the joint position between the line image sensors. The system also adjusts the joint position based on the difference in the transport amount in the line image sensor arrangement direction (specifically, the X direction shown in FIG. 1) between when the joint position is derived and when the derived joint position is applied. This allows for highly accurate joint processing, regardless of the influence of skew in the scanned document.

[0014] <Configuration of image reading device> First, the basic configuration of the image reading device will be described using an overall view of the image reading device and a view showing the image reading mode.

[0015] FIG. 1(a) is a perspective view showing the exterior of a sheet-fed scanner 100. As shown in FIG. 1(a), the scanner 100 has a document feed slot 101 and a document feed tray 102 on the front side of the main body. The user places the leading edge of a document on the document feed tray 102 so that the center of the document is aligned with the center of the feed slot, and then slides the document along the tray to insert it into the document feed slot 101. The document feed slot 101 is designed to allow for a certain degree of misalignment and tilt during insertion relative to the width of the document in the main scanning direction (X direction) that the scanner 100 can read. The configuration of the document feed path will be described in detail in FIG. 1(b). For the sake of explanation, coordinate axes are set as shown in FIG. 1(a), and these coordinate axes will also be applied to other drawings.

[0016] Scanner 100 has an operation unit 103 on the top surface of the main body, which is composed of physical keys, an LCD panel, etc., and allows the user to set reading conditions and input the document size. In addition, a top cover 104 is provided on the top surface of scanner 100, and opening top cover 104 upward allows access to the reading unit, etc., allowing maintenance of the main body to be performed.

[0017] 1(b) and 1(c) are schematic diagrams showing the internal configuration of scanner 100, with FIG. 1(b) being a cross-sectional view and FIG. 1(c) being a top view. In the cross-sectional view of FIG. 1(b), the left side is the upstream side of the document feed and the right side is the downstream side, and the document is transported in the Y direction, which intersects with the X direction. A document 110 fed by a user along document feed tray 102 passes through a flat transport path and is ejected from the rear of the main body.

[0018] The document detection sensor 105 detects the insertion of the document 110, and when the insertion of the document 110 is detected, the control unit 202 (see FIG. 2) of the scanner 100 rotates the upstream transport rollers 107 to draw the document into the main body. The edge detection sensor 112 is used to detect the leading edge of the document 110 that has been drawn into the main body by the rotation of the upstream document transport rollers 107. The detection result of the edge detection sensor 112 is also used to determine the reading start position of the document 110, detect the position of the trailing edge of the document 110, etc.

[0019] Inside the main body, the original 110 passes between the glass plate 109 and the original pressing plate 111. The original pressing plate 111 serves to press the original 110 against the glass plate 109 with a predetermined pressure. The CIS 106 is a line image sensor in which light receiving elements are arranged in the main scanning direction (X direction in the figure), and is composed of multiple chips each made up of multiple light receiving elements. The reading surface of the CIS 106 faces the glass plate 109, and it is designed so that the reading focal position is located at the contact surface between the original 110 and the glass plate 109.

[0020] The downstream document transport rollers 108 are configured to be driven by the upstream document transport rollers 107 via a belt (not shown), and have the role of discharging documents downstream that have passed through the area where the document presser plate 111 presses the glass plate 109. The control unit 202 (described later) is configured by a circuit board for controlling each detection sensor, a motor (not shown) for rotating the upstream document transport rollers 107, the CIS 106, and the operation unit 103.

[0021] 1(c) shows a top view of the scanner 100, in which multiple (five in this example) CISs 106 are arranged in a staggered pattern in the main scanning direction. The scanner 100 reads data using each CIS 106, and the control unit 202 performs processing to splice the data read by each CIS 106 at a splicing position 113.

[0022] 2 is a block diagram showing the hardware configuration of the scanner 100 according to this embodiment. The control unit 202, which controls image scanning and other operations in the scanner 100, includes a CPU 204, a memory 208, a motor driver 207, an interface (hereinafter referred to as IF) unit 203, an A / D conversion unit 206, and a power supply unit 205. The operation unit 103 is configured with a touch panel equipped with an LCD (short for Liquid Crystal Display). Information regarding the document to be scanned and settings of the scanning device is displayed on the LCD of the operation unit 103 in accordance with instructions from the CPU 204. The user can input information to the scanner 100, such as changing various settings, by operating the touch panel of the operation unit 103 while checking the information displayed on the LCD of the operation unit 103.

[0023] The transport motor 201 is controlled by the CPU 204 via a motor driver 207, and rotates the upstream transport rollers 107 and the downstream transport rollers 108. The outputs of the document detection sensor 105 and the edge detection sensor 112 are input to the CPU 204, and the CPU 204 performs control such as determining the drive timing of the multiple CISs 106 based on changes in the output signals of these sensors and the state of the transport motor 201.

[0024] The multiple CISs 106 output the read images as analog signals to the control unit 202. The analog signals output from the multiple CISs 106 are converted into digital signals by each A / D conversion unit 206 and input to the CPU 204. The CPU 204 processes the data converted into digital signals by each A / D conversion unit 206 and can transmit it as image data to an external device connected via USB, LAN, or the like via the IF unit 203. The power supply unit 205 generates the voltage required for each unit and supplies power. The memory 208 is capable of storing image data for multiple lines.

[0025] <Calibration> The flow of reading the document 110 using the CIS 106 and calculating the correction values ​​will be described below with reference to Fig. 3. The correction values ​​may be calculated in advance, or may be calculated each time the document is read.

[0026] When calculating the correction value in advance, the correction value is calculated by scanning a specific document prepared in advance at the factory or at the user's site, and the same correction value is applied to each subsequent scan. In this case, since there is no need to calculate the correction value for each scan, the scan time can be reduced.

[0027] On the other hand, if the correction value is calculated for each scan, the correction value is calculated by scanning a specified document before scanning or by scanning a document with a pattern for calculating the correction value printed in the header section of the document. In this case, the current error component can be corrected as needed, achieving highly accurate scanning.

[0028] First, in step S301, CPU 204 accepts input from the user pressing the calibration start button on operation unit 103. This step puts scanner 100 into a state where it waits for the insertion of a dedicated document to be used for calibration. Hereinafter, for simplicity, "step S~" will be abbreviated to "S~".

[0029] In S302, the CPU 204 determines whether insertion of the document 110 set by the user has been detected. If the determination result in this step is true, the process proceeds to S303. On the other hand, if the determination result in this step is false, the insertion detection determination of the document 110 is performed again.

[0030] In S303, the CPU 204 controls the transport motor 201 to transport the document 110 to the reading start position.

[0031] In S304, the CPU 204 starts the image reading operation and stores the data obtained by reading (referred to as read data) in the memory 208.

[0032] In S305, the CPU 204 determines whether reading of the predetermined length has been completed. If the determination result of this step is true, the process proceeds to S306. On the other hand, if the determination result of this step is false (i.e., if reading of the predetermined length has not been completed), the reading operation continues until reading of the predetermined length has been completed.

[0033] In S306, the CPU 204 ends the image reading operation and causes the calibration document 110 to be transported to the paper discharge position.

[0034] In S307, the CPU 204 performs a correction value calculation process. The correction value obtained in this step is stored in the memory 208, and is read out and applied during normal reading operations.

[0035] Next, the flow of calculating correction values ​​based on the read pattern (details of S307 in FIG. 3) will be described with reference to FIG. 4 and FIG. 5. FIG. 4 is a diagram showing a correction value calculation process pattern (also called a calibration chart) in this embodiment.

[0036] 4, a plurality of circular dot patterns 401 each consisting of a plurality of ON dots (pixel value 1) are printed isolated from one another on the calibration chart, and the circular dot patterns 401 are read while the document is conveyed by the document conveying roller. As a pattern arrangement on the document, the plurality of circular dot patterns 401 are printed across the width (denoted by Xr) of the readable area and the circumference of the document conveying roller 104 (circumferential length denoted by Yr). Note that a plurality of dot patterns 401 may be printed that is greater than the circumference of the document conveying roller 104. In this specification, the circular dot patterns 401 are also referred to as lattice points.

[0037] First, in S501, the CPU 204 functions as a derivation unit that derives the center coordinates of the circular dot pattern 401, and derives the center coordinates of each dot pattern circle 401 from the read data acquired by image reading. The center coordinates derived in this step are used in the process of deriving each correction value, which will be described later.

[0038] In S502, the CPU 204 performs a process of deriving the tilt angle of the CIS 106. Specifically, the tilt angle is calculated by utilizing the fact that multiple circular dot patterns are concentrically arranged so that the sum of their coordinates from the reference coordinates is zero. The process of deriving the tilt angle in this step is a process for suppressing deviation of the stitching position 113 when the scanned data is stitched together. Information on the tilt angle of the CIS 106 obtained in this step makes it possible to stitch the scanned images together later with high accuracy. In the process of deriving each correction value, which will be described later, the correction value according to the tilt angle of the CIS 106 derived in S502 is applied in advance before processing.

[0039] In S503, the CPU 204 performs a process of deriving the sub-scanning direction magnification due to the document conveying rollers. The process of deriving the sub-scanning direction magnification due to the document conveying rollers is a process of deriving the sub-scanning direction magnification due to the diameter error of the document conveying rollers 107 and 108, which affects the overall reading result of the scanner 100.

[0040] In S504, the CPU 204 performs processing to suppress the effects of eccentricity of the document feed roller. This step is processing to correct reading errors in the sub-scanning direction due to eccentricity of the document feed roller, which affects the overall reading result of the scanner 100, for example, by calculating the eccentricity rate of the document feed roller.

[0041] In S505, the CPU 204 performs a process of deriving the main scanning direction magnification due to the step between chips. The process of deriving the main scanning direction magnification due to the step between chips is a process for correcting reading errors in the main scanning direction due to the gaps between each chip inside the CIS 106.

[0042] In S506, CPU 204 performs a process of deriving a joint position. The process of deriving a joint position is a process for accurately joining the reading results of each CIS 106, and is a process of applying in advance a correction value corresponding to each step found from the results of S502 to S505, and deriving joint position 113. This completes the calibration involving the calculation of each correction value.

[0043] <Center coordinate derivation process> The process of deriving the center coordinates of a circular dot pattern based on the read data (S501 in Fig. 5) will be specifically described below using the circular dot pattern 401 shown in Fig. 6 and the flowchart in Fig. 7. As shown in Fig. 6, the circular dot pattern 401 needs to be a pattern that is somewhat large relative to the pixels to be read by the scanner.

[0044] First, in S701, the CPU 204 extracts all pixel data in the main scanning direction (1) of the CIS 106 at a position of interest in the sub-scanning direction (2) of the CIS 106 from all the read data.

[0045] In S702, the CPU 204 determines whether there are any consecutive pixels in the main scanning direction (1) whose gradation values ​​exceed a threshold value Xth, based on the pixel data extracted in S701, and performs binarization on each pixel as shown in Fig. 6. If the determination result in this step is true, the process proceeds to S703, whereas if the determination result is false, the process proceeds to S705. The threshold value Xth used in this step is set in advance, and the data is stored in the memory 208.

[0046] In S703, the CPU 204 derives the position of the central pixel of the continuous pixels whose gradation value exceeds the threshold value Xth as the central coordinate in the main scanning direction (1).

[0047] In S704, the CPU 204 determines whether the derivation of the center coordinates in the main scanning direction (1) has been completed for all lines in the sub-scanning direction (2).

[0048] In S705, the CPU 204 advances the position of interest in the sub-scanning direction (2) by one pixel.

[0049] In S706, the CPU 204 calculates the average of the derived central coordinates in the main scanning direction, and sets the calculated average as the central coordinate 601 of the circular dot pattern 401.

[0050] If reading errors due to dust are to be taken into consideration when calculating the central coordinate in the main scanning direction, this can be addressed by making the dot pattern larger. Also, if reading errors due to gaps between chips in the CIS 106 are to be taken into consideration, when calculating the coordinate in the main scanning direction (1), it is necessary to select a location that does not straddle between chips and calculate the central coordinate.

[0051] Furthermore, the shape of the dot pattern is preferably a substantially circular shape as shown in FIG. 6. The reason for this is that the substantially circular shape is less affected by error components during reading when deriving the center coordinates 601. For example, when there is an inclination during manuscript setting, if the shape of the dot pattern is square, it is difficult to determine whether the pixel data in the main scanning direction (1) of the read image data of the CIS 106 has pixels whose gradation values continuously exceed the threshold value Xth in the main scanning direction (1). The substantially circular shape is easier to distinguish continuous pixel data whose gradation values continuously exceed the threshold value Xth in the main scanning direction (1) than a square shape. Also, when the dot pattern is substantially circular, it is not necessary to perform the process of deriving the center coordinates for all lines in the sub-scanning direction (2) as in S704 to S706 of FIG. 7. That is, on the premise that the dot pattern is substantially circular, it is possible to derive the center coordinates 601 of the dot pattern by inferring pixel data whose gradation values continuously exceed the threshold value Xth in the main scanning direction (1), so it is possible to shorten the time required for center coordinate derivation.

[0052] <Derivation of the inclination angle of the CIS> Using the center coordinates 601 derived by the above-described center coordinate derivation process, the processes of S502 to S506 are performed. As described above, since this embodiment derives the inclination angle of the CIS 106 in advance, hereinafter, the derivation process of the inclination angle of the CIS in S502 in the flow of correction value calculation shown in FIG. 5 will be described.

[0053] Here, using the inclination angle derivation pattern of the CIS 106 shown in FIG. 8(a), the algorithm shown in FIG. 9, and the flowchart shown in FIG. 10, the process of reading the circular dot pattern 401 and deriving the inclination angle of the CIS 106 is shown. The circular dot pattern group shown in FIG. 8(a) includes a central circular dot pattern (referred to as a reference dot pattern) and four circular dot patterns around this reference dot pattern (referred to as surrounding dot patterns). By using such an arrangement of the circular dot pattern 401, it is possible to accurately derive the inclination of the CIS 106 regardless of the inclination during manuscript setting or the diameter error of the manuscript conveyance roller.

[0054] First, in S1001, the CPU 204 extracts a specified number (assumed to be X) of center coordinates of the circular dot pattern 401 for each CIS 106.

[0055] In S1002, the CPU 204 sets the central coordinate among the specified number X of coordinates for each CIS 106 as the reference coordinate P, and the coordinates around the reference coordinate as the tilt angle correction value calculation coordinates A1 to A1. Fig. 8(a) shows a case where four tilt angle correction value calculation coordinates A1 to A4 exist around the reference coordinate P as the center.

[0056] In S1003, the CPU 204 performs linear approximation using the reference coordinate P set in S1002 and the tilt angle correction value calculation coordinates A1 to AN, with the reference coordinate P as a passing point.

[0057] The change in coordinates during linear approximation and the algorithm for deriving the skew angle will be explained using Figures 8(b) to 8(d) and 9. Figure 8(b) shows the change in coordinates for calculating the skew angle correction value with reference to the reference coordinate P when the document 110 is scanned at an angle, and shows the case where the document skew angle is 45 degrees. Figure 8(c) shows the change in coordinates for calculating the skew angle correction value with reference to the reference coordinate P when there is a diameter error in the upstream document transport roller 107, and shows the case where the document transport roller's sub-scanning direction magnification is μ.

[0058] 9 shows the formula used in the algorithm for deriving the tilt angle φ of the CIS 106. Formula (1) indicates that the sum of the coordinates for calculating the tilt angle correction value based on the reference coordinate P is 0. The calculation formula (1) does not change even if there are effects of the document tilt angle θ and the sub-scanning direction magnification μ due to the document feed roller shown in FIGS. 8(b) and 8(c).

[0059] Next, Fig. 8(d) shows the coordinate change of the tilt angle correction value calculation coordinates with respect to the reference coordinate P when reading is performed with the tilt angle φ of CIS 106. Equations (2) to (5) in Fig. 9 show the coordinates after the change when the tilt angle correction value calculation coordinates change due to the tilt angle φ of CIS 106.

[0060] Equation (6) in Fig. 9 is an approximation equation used when performing linear approximation using the tilt angle correction value calculation coordinates A1 to AN with the reference coordinate P as the reference in S1003, and equation (7) in Fig. 9 represents the slope and intercept in equation (6). Equations (8) to (10) are obtained by substituting equations (2) to (5) into each of the constituent equations in equation (7) in Fig. 9.

[0061] 9 into equation (7), equation (7) can be simplified as in equation (11). Equation (12) is an equation for deriving the tilt angle φ of the CIS 106 based on the tilt shown in equation (11).

[0062] In S1003, the CPU 204 calculates the slope a of the line by performing linear approximation according to the algorithm shown in FIG.

[0063] In S1004, the CPU 204 calculates the tilt angle φ of the CIS 106 using equation (12) based on a calculated in S1003.

[0064] In S1005, the CPU 204 calculates a correction value for correcting the tilt angle of the CIS 106 based on φ calculated in S1004.

[0065] In S1006, the CPU 204 writes the correction value calculated in S1005 into the memory 208 to store it.

[0066] As a method for applying the correction value, it is possible to employ a method of reading while changing the memory access location at each reading based on the correction value written to the memory in S1006, a method of physically adjusting the tilt of the CIS 106, etc. Alternatively, it is also possible to directly change the memory access location or physically adjust the tilt of the CIS 106 using the tilt a of the straight line calculated in S1003 without calculating the tilt angle φ of the CIS 106.

[0067] <Connection position derivation process> The process of deriving a joint position in this embodiment will be described below with reference to Figs. 11 to 16. Fig. 11 is a diagram showing coordinate transformation based on the tilt angle of the CIS 106 obtained in the immediately preceding calculation of the correction value. Fig. 12 is a diagram showing an overview of deriving a joint position between the CISs 106. Fig. 13 is a flowchart showing the flow of a process of deriving a joint position based on read data.

[0068] First, in S1301, the CPU 204 performs coordinate transformation based on the tilt angle of CIS 106 obtained in the immediately preceding calculation of the correction value to transform the center coordinate 601 of the circular dot pattern 401 described with reference to FIGS. 6 and 7 back to the state before the change due to the tilt of CIS 106. FIG. 11(a) is a diagram showing the change in the center coordinate 601 of the circular dot pattern 401 due to the tilt angle φ of CIS 106. FIG. 11(b) is a diagram showing the result of coordinate transformation to transform the center coordinate 601 of the circular dot pattern 401 back to the state before the change due to the tilt of CIS 106, using the tilt angle φ of CIS 106 calculated in the immediately preceding calculation of the correction value. The coordinate transformation based on the tilt angle φ of CIS 106 is performed according to the calculation formula shown in FIG. 11(c).

[0069] Here, the embodiment is described in which the connection position is derived by converting the coordinates of the center coordinates 601 of the circular dot pattern 401. However, the connection position may be derived by re-reading the document 110 on which the circular dot pattern 401 is printed, with the inclination correction value of the CIS 106 obtained in the immediately preceding correction value calculation being applied.

[0070] In S1302, the CPU 204 selects processing target coordinates for the Nth CIS (referred to as CIS[N]) 1201 and the N+1th CIS (referred to as CIS[N+1]) 1202 based on the center coordinate 601 of the circular dot pattern 401 that was coordinate-converted in S1301. FIG. 12(a) is a diagram showing the arrangement of the circular dot pattern 401 relative to the overlapping portion of multiple CISs 106. FIG. 12(b) shows the read result of CIS[N] 1201, and FIG. 12(c) shows the read result of CIS[N+1] 1202. As shown in FIG. 12(b), for the data read by CIS[N] 1201, the center coordinate 601 of the circular dot pattern 401 that is closest to a predetermined fixed reference position 1203 of CIS[N] is selected as the processing target coordinate 1205 of CIS[N]. Similarly, as shown in FIG. 12C, in CIS[N+1] 1202, the center coordinates 601 of the circular dot pattern 401 closest to the connection reference position 1204 of CIS[N+1] are selected as the processing target coordinates 1206 of CIS[N+1].

[0071] In S1303, the CPU 204 calculates offset values ​​in the X and Y directions based on the difference in the X and Y coordinates between the processing target coordinates 1205 of CIS[N] selected in S1302 and the processing target coordinates 1206 of CIS[N+1].

[0072] In S1304, the CPU 204 calculates a correction value for correcting the joint position based on the offset values ​​in the X and Y directions calculated in S1303.

[0073] In S1305, the CPU 204 determines whether or not the correction values ​​for correcting the joint positions have been calculated for all overlapping portions of the multiple CISs 106. If the determination result in this step is true, the process proceeds to S1306. On the other hand, if the determination result in this step is false (i.e., if the calculation of the correction values ​​for correcting the joint positions has not been completed for all overlapping portions of the multiple CISs 106), the process returns to S1302, and the correction values ​​for correcting the joint positions of the next overlapping portion are calculated.

[0074] In S1306, the CPU 204 writes and stores the correction values ​​for correcting the joint positions derived for all overlapping portions of the multiple CISs 106 in the memory 208.

[0075] The correction value thus obtained can be applied by, for example, changing the memory access location during each read based on the correction value written to the memory in S1306, or by physically adjusting the position of the CIS 106.

[0076] Furthermore, in deriving the boundary position, in order to suppress the influence of quantization errors when deriving the center coordinate 601 of the circular dot pattern 401, it is advisable to arrange multiple circular dot patterns 401 in the overlapping portion. In the above-described arrangement configuration, in S1302, the boundary reference positions of CIS[N] and CIS[N+1] are used to select processing target coordinates 1205 and 1206 for CIS[N] and CIS[N+1] for the number of lattice points read in the overlapping portion. When calculating a correction value for boundary position correction in S1304, the average result of the multiple offset values ​​in the X and Y directions calculated in S1303 is used.

[0077] FIG. 14 shows the arrangement of circular dot patterns when there is a misalignment 1401 when the document is set. Here, as shown in FIG. 14(a), a case will be considered in which a document set misalignment 1401 occurs when the user sets the document 110. In such a case, if only one circular dot pattern 401 is arranged in the overlapping portion of the CIS 106 as shown in FIG. 12(a), there is a possibility that no pattern will fit into the overlapping portion. To address this issue, as shown in FIG. 14(b), two patterns, a first pattern 1402 for calculating a boundary position correction value and a second pattern 1403 for calculating a boundary position correction value, are arranged in the overlapping portion of the CIS 106. The solution shown in FIG. 14(b) makes it possible to derive the boundary position between the CISs 106 even when there is a misalignment 1401 when the document is set, as shown in FIG. 14(c).

[0078] FIG. 15 shows a pattern arrangement that prevents erroneous determination of the connection position in a configuration in which multiple dot patterns are arranged in the overlapping portion of the CIS 106. As shown in FIG. 15(a), multiple dot patterns may be arranged in the overlapping portion of the CIS 106 to reduce the effects of the quantization error described above or to reduce the effects of misalignment when the document is set. Ideally, the connection reference position 1203 of CIS[N] and the connection reference position 1204 of CIS[N+1] are located on the same straight line in the document transport direction (the same in the main scanning direction). However, in reality, the positions of multiple CISs 106 are misaligned due to mechanical installation tolerances, so these reference positions are misaligned from the straight line. Therefore, depending on the amount of misalignment of the CIS positions, it may be impossible to select the same dot pattern in step S1302, such as dot pattern 1501 closest to the reference coordinates of CIS[N] and dot pattern 1502 closest to the reference coordinates of CIS[N+1].

[0079] FIG. 15(b) shows a configuration consisting of multiple dot patterns with different sizes. FIG. 15(c) shows a configuration consisting of multiple dot patterns with different optical densities. By adopting the configuration of FIG. 15(b), for example, when selecting dot patterns with a diameter greater than Xb that are closest to the connection reference position 1203 of CIS[N] and the connection reference position 1204 of CIS[N+1], it is possible to select the same patterns in S1302. Also, even when adopting the configuration of FIG. 14(c), for example, when selecting dot patterns with an optical density greater than Db that are closest to the connection reference position 1203 of CIS[N] and the connection reference position 1204 of CIS[N+1], it is possible to select the same patterns in S1302.

[0080] As explained above, it is possible to prevent erroneous derivation of the connection position by using a configuration in which multiple dot patterns are arranged in the overlapping portion of the CIS 106. In addition to dot patterns with different sizes or optical densities, dot patterns with different colors are also effective in preventing erroneous derivation.

[0081] FIG. 16 shows the internal structure of CIS[N] 1201 and CIS[N+1] 1202 and the arrangement of circular dot patterns 401. CIS 106 is generally composed of multiple chips, and there is a positional misalignment dx between the chips for each internal chip 1601 in CIS[N] 1201 and for each internal chip 1602 in CIS[N+1] 1102. As shown in FIG. 16, circular dot patterns 401 are arranged in the overlapping portion (referred to as the chip overlapping portion) between chip 1601 in CIS[N] and chip 1602 in CIS[N+1]. This allows the connection position between CISs 106 to be derived while ignoring the influence of the gap dx between the chips. To suppress the influence of the document set misalignment 1401 shown in FIG. 14, two circular dot patterns 401 may be arranged in the overlapping portion between chip 1601 in CIS[N] and chip 1602 in CIS[N+1].

[0082] As described above, by deriving the joint position between the CISs 106 after accurately deriving the inclination of the CISs 106, regardless of the inclination when the document is set or the diameter error of the document transport roller, highly accurate joint processing between the line image sensors can be realized.

[0083] Furthermore, since the pattern used when deriving the connection position between the CISs 106 and the pattern used when deriving the inclination of the CISs 106 can be unified, the cost of creating the calibration chart can be reduced.

[0084] Next, a method for adjusting the joint position based on the difference in the transport amount in the arrangement direction (X direction) of the line image sensor between when the joint position is derived and when the derived joint position is applied will be described.

[0085] FIG. 17 is a schematic diagram illustrating the effect of skew on an image reading device. FIG. 17(a) is a cross-sectional view of the image reading device when the top cover is opened and closed. A typical image reading device has a top cover 104 that can be opened and closed to remove documents from the inside or wipe off dirt when clearing a jam. Each of the upstream document transport roller 107 and the downstream document transport roller 108 is composed of two rollers, one above the other. Each of the upstream document transport roller 107 and the downstream document transport roller 108 has some play due to mechanical installation tolerances. Therefore, when the top cover 104 is opened and closed, the meshing positions of the two rollers change. If a document is transported with the roller meshing positions different, the load on each document transport roller shifts to the right or left during transport, causing the document to be transported skewed.

[0086] Fig. 17(b) shows the calibration process when the document is skewed. In the connection position derivation process, which is one of the calibration processes shown in Fig. 12 and Fig. 13, if there is a skew component as explained in Fig. 17(a), the correction value for correcting the connection position is derived with a deviation equal to the skew component.

[0087] In contrast, FIG. 17(c) shows the normal scanning process when the original is skewed. In the normal scanning process, the original is scanned while applying the correction value for correcting the stitching position derived by scanning the calibration chart as shown in FIG. 17(b). If the skew component in the normal scanning process is equivalent to the skew component in the calibration process, the amount of deviation due to the skew component in the normal scanning process will be equivalent to the amount of deviation due to the skew component in the calibration process. This allows the stitching process to be performed without being affected by the skew. However, as shown in FIG. 17(a), the skew component during document transport changes when the upper cover 104 is opened or closed. Therefore, if the upper cover 104 is opened or closed after the calibration process, the skew component may differ between the calibration process and the normal scanning process. If the normal scanning process is performed with the skew component different, a stitching deviation 1701 will occur due to the difference in the skew component.

[0088] Fig. 18 is a schematic diagram showing a means for deriving a transport amount required to suppress the effects of skew in an image reading device. Fig. 18(a) is a diagram explaining a calibration process for suppressing the effects of skew, and Fig. 19 is a flowchart of the calibration process. Fig. 21 is a flowchart of a transport amount derivation process executed by an image reading device.

[0089] As shown in FIG. 18(a), the image reading device includes a transport distance measuring unit 1801 that directly measures the actual X-direction transport distance of the document in order to suppress the effects of skew during calibration processing. The transport distance measuring unit 1801 includes an image sensor and a lens, and is configured to capture images recorded on paper at regular time intervals Δt and calculate a movement distance ΔX based on the correlation between the captured images. Using the transport distance measuring unit 1801 configured in this way, the amount of movement of the document per unit time can be calculated, and the X-direction transport distance of the document can be derived based on this movement distance. The transport distance measuring unit 1801 is located upstream of the CIS 106 in the document transport direction, and may be configured to read either the front or back side of the document.

[0090] Next, the calibration process for suppressing the influence of skew will be described with reference to Fig. 19 and Fig. 21. Note that S1901 to S1904 and S1906 to S1908 in Fig. 19 are the same as S301 to S304 and S305 to S307 in Fig. 3, and therefore description thereof will be omitted.

[0091] After S1904, in S1905, the CPU 204 executes a process for deriving the transport amount in the X direction of the original document. Fig. 21 is a detailed flowchart of the process for deriving the transport amount in the X direction of the original document (S1905).

[0092] In the process of deriving the X-direction transport amount of the original, in S2101, the CPU 204 captures an image using the image sensor of the transport amount measurement means 1801 to obtain a calibration chart image (referred to as the first calibration chart image) at the start of the process of deriving the X-direction transport amount of the original.

[0093] In S2102, the CPU 204 determines whether reading of the calibration chart by the CIS 106 for a predetermined length in the Y direction has been completed. If the determination result in this step is true, the process proceeds to S2103. On the other hand, if the determination result in this step is false, reading of the calibration chart continues.

[0094] In S2103, the CPU 204 acquires a calibration chart image (hereinafter referred to as a second calibration chart image) by capturing an image using the image sensor of the transport amount measuring unit 1801. Then, the CPU 204 derives the X-direction transport amount of the document using the first calibration chart image acquired in S2101 and the second calibration chart image acquired in this step.

[0095] In S2104, the CPU 204 writes the transport amount derived in S2103 to the memory 208. Through the series of processes described above, the transport amount (referred to as Xc) of the document in the X direction during calibration processing is derived.

[0096] Fig. 18(b) is a diagram for explaining normal reading processing for suppressing the effects of skew, Fig. 20 is a flowchart of the normal reading processing, and Fig. 22 is a flowchart of skew correction processing executed by the image reading device.

[0097] 18(b), the image reading device has a transport amount measuring means 1801. In order to suppress the influence of skew even during normal reading processing, the transport amount measuring means 1801 is used to derive the X-direction transport amount similar to that shown in FIG.

[0098] Hereinafter, the normal reading process for suppressing the influence of skew will be described with reference to FIGS.

[0099] In S2001, the CPU 204 receives an instruction from the user to press the read start button on the operation unit 103. Note that S2002 to S2004 and S2007 to S2008 in Fig. 20 are similar to S302 to S304 and S305 to S306 in Fig. 3, and therefore description thereof will be omitted.

[0100] In S2005 after S2004, the CPU 204 executes the transport amount derivation process described with reference to Fig. 21. Through the process of this step, the transport amount (referred to as Xs) of the document in the X direction during normal reading is derived.

[0101] After deriving Xs in S2005, the CPU 204 executes skew correction processing in S2006.

[0102] The skew correction process in S2006 will now be described with reference to Fig. 22. Fig. 22 is a detailed flowchart of the skew correction process.

[0103] In the skew correction process, first, in S2201, the CPU 204 compares the X-direction conveyance amount Xc derived in S1905 with the X-direction conveyance amount Xs derived in S2005, and derives the difference between Xc and Xs (=Xc-Xs).

[0104] In S2202, the CPU 204 calculates a skew correction value (denoted as Xt) based on the difference between Xc and Xs calculated in S2201. Specifically, as shown in FIG. 18(c), the skew correction value Xt is calculated based on the Y-direction gap (denoted as Yg) of each CIS 106 and the document conveyance speed.

[0105] In S2203, the CPU 204 adjusts the correction value for deriving the joint position stored in S1306 based on the skew correction value Xt calculated in S2202. Specifically, the CPU 204 adds the skew correction value Xt to the correction value for deriving the joint position stored in S1306.

[0106] In S2204, the CPU 204 updates the correction value for deriving the joint position, that is, overwrites the memory 208 with the correction value for deriving the joint position adjusted in S2203.

[0107] <Effects of this embodiment> As described above, according to this embodiment, based on the difference between the X-direction transport amount during calibration processing and the X-direction transport amount during normal reading, it is possible to suppress the influence of skew during document transport when correcting the joint positions between the CISs 106. Note that the joint positions between the CISs 106 can be corrected by changing the read destination within the image data storage unit in which the image data acquired by each line image sensor is stored.

[0108] In this embodiment, a calibration chart consisting of dot pattern 401 is used as the pattern for calibration processing, but other patterns such as straight line patterns may be used in addition to dot patterns as long as the purpose is to reduce the effects of skew.

[0109] Regarding the error component in the Y direction due to skew, which was not mentioned in this embodiment, the Y direction transport amount can be directly derived by using the transport amount measurement unit 1801, or by providing a separate transport amount measurement unit, just like the X direction transport amount directly derived by the transport amount measurement unit 1801. This makes it possible to suppress the influence of skew based on the Y direction transport amount difference, as described above. Alternatively, the Y direction transport amount can be estimated from the X direction transport amount derived by the transport amount measurement unit 1801, and the error component in the Y direction due to skew can be suppressed based on the estimated Y direction transport amount.

[0110] [Other embodiments] The present invention can also be realized by supplying a program that realizes one or more functions of the above-described embodiments to a system or device via a network or a storage medium, and having one or more processors in the computer of the system or device read and execute the program. It can also be realized by a circuit (e.g., ASIC) that realizes one or more functions. The contents of the above-described embodiments may also be combined as appropriate. [Explanation of symbols]

[0111] 100 scanners 106 CIS 107 Upstream document transport roller 108 downstream document transport roller 110 manuscripts 204 CPU1801 Transport amount measuring means

Claims

1. a plurality of line image sensors arranged in a staggered pattern along a first direction; a conveying means for conveying a chart or a document on which a plurality of dot patterns are printed in a second direction perpendicular to the first direction; a deriving unit that derives a first direction transport amount of the first direction transport associated with the transport in the second direction when the chart or the document is transported in the second direction by the transport unit, the deriving unit deriving the first direction transport amount for the chart and the first direction transport amount for the document; a skew amount calculation unit that calculates a skew amount in the first direction for each of the plurality of line image sensors, the skew amount calculation unit calculating the skew amount during calibration processing based on a slope of a straight line calculated by linear approximation using reference coordinates of a reference dot pattern and coordinates of dot patterns surrounding the reference dot pattern in read data of the conveyed chart; a correction value calculation means for calculating, based on the amount of skew, a correction value for a joint position of images read by each of the plurality of line image sensors, the correction value calculation means deriving processing target coordinates for each of a first line image sensor and a second line image sensor whose positions overlap in the first direction by coordinate transformation based on the amount of skew, calculating a first offset value for the first direction and a second offset value for the second direction based on a difference in the first direction and a difference in the second direction between a first processing target coordinate for the first line image sensor and a second processing target coordinate for the second line image sensor, and calculating the correction value based on the first offset value and the second offset value; an adjustment unit that adjusts the correction value calculated by the correction value calculation unit, the adjustment unit calculating a skew correction amount based on a difference between the first direction transport amount for the chart and the first direction transport amount for the document, adjusting the correction value based on the skew correction amount, and writing the adjusted correction value to a memory; Equipped with When reading the adjusted correction values ​​written in the memory, a method of reading while changing an access location of the memory, or a method of physically adjusting the inclination of each of the plurality of line image sensors is used to stitch together images obtained when the document is read by each of the plurality of line image sensors. An image reading device characterized by:

2. The transport means includes a first document transport roller and a second document transport roller disposed downstream of the first document transport roller in the second direction.

2. The image reading device according to claim 1, wherein:

3. Further comprising an imaging element and a lens, the imaging element and the lens are disposed upstream of the plurality of line image sensors and downstream of the first document transport roller in the second direction; 3. The image reading device according to claim 2, wherein:

4. The skew amount calculation means acquires a plurality of images of the conveyed chart spaced apart in the second direction, the plurality of images includes a first image and a second image different from the first image, the second image is an image captured when the sheet is transported a predetermined length in the second direction after the first image is captured.

4. The image reading device according to claim 3, wherein:

5. the adjusting means adds the skew correction amount to the correction value; 5. The image reading device according to claim 4, wherein:

6. the plurality of line image sensors are disposed between the first document transport roller and the second document transport roller in the second direction; 6. The image reading device according to claim 5, wherein:

7. the plurality of dot patterns are arranged isolated from one another on the chart, Each of the plurality of dot patterns is circular.

6. The image reading device according to claim 5, wherein:

8. a plurality of line image sensors arranged in a staggered pattern along a first direction; a conveying means for conveying a chart or a document on which a plurality of dot patterns are printed in a second direction perpendicular to the first direction; A control method for an image reading apparatus comprising: a deriving step of deriving a first direction transport amount of the first direction transport accompanying the transport in the second direction when the chart or the document is transported in the second direction by the transport unit, the deriving step deriving the first direction transport amount for the chart and the first direction transport amount for the document; a skew amount calculation step of calculating a skew amount in the first direction for each of the plurality of line image sensors, the skew amount being calculated based on a slope of a straight line calculated by linear approximation using reference coordinates of a reference dot pattern and coordinates of dot patterns surrounding the reference dot pattern in read data of the conveyed chart; and a correction value calculation step of calculating, based on the amount of skew, a correction value for a joint position of images read by each of the plurality of line image sensors, the correction value calculation step deriving processing target coordinates for each of a first line image sensor and a second line image sensor whose positions overlap in the first direction by coordinate transformation based on the amount of skew, calculating a first offset value for the first direction and a second offset value for the second direction based on a difference in the first direction and a difference in the second direction between a first processing target coordinate for the first line image sensor and a second processing target coordinate for the second line image sensor, and calculating the correction value based on the first offset value and the second offset value; an adjusting step of adjusting the correction value calculated in the correction value calculating step, wherein a skew correction amount is calculated based on a difference between the first direction transport amount for the chart and the first direction transport amount for the document, the correction value is adjusted based on the skew correction amount, and the adjusted correction value is written to a memory; a step of executing a process of stitching together images obtained by reading the document with each of the plurality of line image sensors by a method of reading the adjusted correction values ​​written in the memory while changing an access location of the memory or by a method of physically adjusting an inclination of each of the plurality of line image sensors; having 2. A method for controlling an image reading device, comprising:

9. A program for causing a computer to execute the image reading device control method according to claim 8.

Images