Image reading device

Abstract

To obtain shading data with reduced influence of skew as compared with shading data obtained by an image sensor array receiving reflected light from a white reference arc surface which is an outer peripheral surface of a shading roller even when the shading roller is skewed.SOLUTION: A read data acquisition unit 40 acquires arc surface read data based on light reflected from a white reference arc surface 30, and plane read data based on light reflected from a white reference plane 32. A difference data calculation unit 42 calculates difference data representing a difference between the arc surface read data and the plane read data based on the arc surface read data and the plane read data. The difference data can be said to be data representing a fluctuation amount of a pixel value due to the effect of skew of a shading roller 20. A shading data acquisition unit 44 acquires shading data based on the arc surface read data and the difference data.SELECTED DRAWING: Figure 5

Description

[Technical Field]

[0001] The present invention relates to an image reading device. [Background technology]

[0002] Conventionally, image reading devices have been proposed that use a contact image sensor (CIS) to read the surface of a document while it is being conveyed. The CIS includes an image sensor array consisting of multiple image sensors aligned in the main scanning direction. Image reading devices may also have a shading correction function that corrects for variations in the light intensity of a light source in the main scanning direction or variations in sensitivity between image sensors in the image data obtained by scanning.

[0003] For example, Patent Document 1 discloses an image reading device that uses a CIS to read the surface of a document as it is conveyed, and that includes a shading roller that faces the CIS across the document conveyance path. The shading roller has a reference surface on its outer circumferential surface for obtaining correction data for shading correction. The shading roller in Patent Document 1 is configured so that the distance from the CIS to the reference surface (outer circumferential surface) changes as the shading roller rotates.

[0004] Furthermore, Patent Document 2 discloses an image reading device in which a shading roller is provided opposite a CIS, similar to Patent Document 1. In the image reading device of Patent Document 2, a correction curve consisting of a plurality of correction pixel values ​​for shading correction is prepared in advance for pixel values ​​output by each of a plurality of image sensors arranged in the main scanning direction of the CIS, and pixel values ​​affected by dust are detected from among the plurality of pixel values ​​obtained by reading the reference surface of the shading roller based on a pixel value curve consisting of a plurality of pixel values ​​obtained by reading the reference surface of the shading roller and the correction curve. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Laid-Open No. 2013-81121 [Patent Document 2] Japanese Patent Application Laid-Open No. 2010-011297 Summary of the Invention [Problem to be solved by the invention]

[0006] Image sensors used in scanning, which receive reflected light emitted from a light source toward a document, are arranged in the main scanning direction. This forms an image sensor array consisting of multiple image sensors. Meanwhile, a shading roller, which is arranged opposite the image sensor array across the document transport path, has a columnar shape extending in the longitudinal direction. The outer peripheral surface of the shading roller serves as a reference surface for obtaining correction data for shading correction (hereinafter referred to as shading data). This reference surface is white.

[0007] Ideally, the longitudinal direction of the shading roller would be parallel to the main scanning direction (i.e., the alignment direction of the image sensor array). However, as shown in FIG. 20, due to factors such as the installation accuracy of the shading roller SDR, the longitudinal direction of the shading roller SDR and the alignment direction of the image sensor array ISA may not be parallel. In FIG. 20 (and FIG. 21, described later), the X-axis direction represents the main scanning direction, the Y-axis direction represents the sub-scanning direction, and the Z-axis direction represents the height direction. In this specification, when the longitudinal direction of the shading roller and the alignment direction of the image sensor array are not parallel, this may be expressed as the shading roller being skewed.

[0008] The amount of light received by the image sensor varies depending on the distance between the image sensor and the outer surface of the shading roller or the document. Specifically, the amount of light received by the image sensor decreases in proportion to the square of the distance. Therefore, to obtain appropriate shading data, it is ideal to obtain the shading data by making the distance from the image sensor to the document the same as the distance from the image sensor to the outer surface of the shading roller.

[0009] Here, if the outer surface of the shading roller facing the image sensor array is an arc-shaped surface in a cross section in the short direction, if the shading roller is skewed, a problem occurs in that the distance from the image sensor to the outer surface of the shading roller varies depending on the position in the main scanning direction.

[0010] This problem will be explained with reference to Figure 21. Figure 21 is a perspective view showing a state in which a shading roller SDR having a circular cross section is skewed. Assume that the positions of the image sensor array ISA and the shading roller SDR in the Y-axis direction coincide at a position where the X coordinate is x1. Therefore, at the position where the X coordinate is x1, the distance d1 from the image sensor array ISA to the outer peripheral surface of the shading roller SDR is the distance from the image sensor array ISA to the point on the outer peripheral surface of the shading roller SDR that is closest to the image sensor array ISA (for convenience, this will be expressed as "directly below").

[0011] On the other hand, at a position where the X coordinate is x2, away from x1, the shading roller SDR is skewed, so the position of the shading roller SDR is shifted in the Y-axis direction relative to the image sensor array ISA. Therefore, at the position where the X coordinate is x2, the distance d2 from the image sensor array ISA to the outer peripheral surface of the shading roller SDR is the distance from the image sensor array ISA to a point slightly shifted to the side from directly below the shading roller SDR. Here, because the outer peripheral surface of the shading roller SDR is arc-shaped, the distance d2 is longer than the distance d1.

[0012] As mentioned above, when the shading roller is skewed, the distance from the image sensor to the outer peripheral surface of the shading roller varies depending on the position in the main scanning direction, which causes the problem that appropriate shading data cannot be obtained. Fig. 22 is a diagram showing an example of shading data. In Fig. 22, the solid line represents shading data when the shading roller is skewed, and the dashed line represents appropriate shading data.

[0013] The object of the present invention is to make it possible to obtain shading data with reduced effects of skew, even when the shading roller is skewed, compared to shading data obtained by an image sensor array receiving reflected light from the white reference arc surface, which is the outer surface of the shading roller. [Means for solving the problem]

[0014] The invention according to claim 1 is a shading roller that is columnar and extends in a longitudinal direction and is rotatable around a rotation axis extending in the longitudinal direction, the shading roller having, on its outer peripheral surface, a white reference arc surface that is white, has an arc-shaped cross section in the short side direction, and extends in the longitudinal direction, and a white reference flat surface that is white, flat, and extends in the longitudinal direction; an image sensor array that is disposed opposite the shading roller across a document transport path and that receives reflected light of light irradiated from a light source toward the document or the shading roller, and that is composed of a plurality of image sensors aligned in a main scanning direction; and a processor, wherein the processor is configured to receive a signal obtained by the image sensor array receiving the reflected light from the white reference arc surface. The image reading device further calculates difference data representing the difference between the arc surface read data and the plane read data based on arc surface read data indicating the pixel values ​​output by each of the image sensors and plane read data obtained by the image sensor array receiving the reflected light from the white reference plane, obtains corrected arc surface read data based on the arc surface read data and the difference data, corrects the document read data obtained by the image sensor array receiving the reflected light from the document based on the corrected arc surface read data to obtain corrected document read data, and forms image data corresponding to the document based on the corrected document read data. The invention of claim 2 is the image reading device described in claim 1, characterized in that the processor calculates the difference data based on arc surface skeleton data obtained by applying a low-pass filter to the arc surface reading data, and planar skeleton data obtained by applying a low-pass filter to the planar reading data. The invention according to claim 3 is characterized in that the processor corrects the arc surface read data based on the difference data to obtain intermediate arc surface read data, and takes a difference between the intermediate arc surface read data and intermediate arc surface skeleton data obtained by applying a low-pass filter to the intermediate arc surface read data, thereby obtaining arc surface foreign matter data indicating the image sensor affected by a foreign matter present between the white reference arc surface and the image sensor array and the degree of influence of the foreign matter on the image sensor, and takes a difference between the plane read data and the plane skeleton data, thereby obtaining the plane read data: The image reading device according to claim 2, characterized in that planar foreign object data indicating the image sensor affected by a foreign object present between the white reference plane and the image sensor array and the degree of influence of the foreign object on the image sensor from the foreign object are obtained, and the corrected arcuate surface read data is obtained by combining pixel values ​​of the intermediate arcuate surface read data for image sensors where the influence of the foreign object indicated by the arcuate surface foreign object data is smaller than the influence of the foreign object indicated by the planar foreign object data, and by using pixel values ​​of the planar read data for image sensors where the influence of the foreign object indicated by the planar foreign object data is smaller than the influence of the foreign object indicated by the arcuate surface foreign object data. The invention of claim 4 is the image reading device described in claim 1, characterized in that the processor determines a first representative pixel value for each of the image sensors by dividing a sum of N pixel values ​​in a time series obtained by the image sensor repeatedly outputting pixel values ​​related to the reflected light from the white reference arc surface N times by a first correction coefficient, and obtains the first representative pixel value for each of the image sensors as the arc surface reading data, and determines a second representative pixel value for each of the image sensors by dividing a sum of M pixel values ​​in a time series obtained by the image sensor repeatedly outputting pixel values ​​related to the reflected light from the white reference arc surface M times, different from N, by a second correction coefficient different from the first correction coefficient, and obtains the second representative pixel value for each of the image sensors as the plane reading data. The invention of claim 5 is an image reading device as described in claim 4, characterized in that each image sensor repeatedly outputs pixel values ​​related to the reflected light from the white reference arc surface of the rotating shading roller N times, thereby obtaining N pixel values ​​in a time series. [Effects of the Invention]

[0015] According to the invention of claim 1, even if the shading roller is skewed, shading data can be obtained that has reduced effects of skew compared to shading data obtained by an image sensor array receiving reflected light from the white reference arc surface, which is the outer surface of the shading roller. According to the invention of claim 2, it is possible to calculate the difference data after reducing the influence of foreign matter or high frequency noise contained in the arcuate surface read data and the plane read data. According to the invention of claim 3, it is possible to acquire corrected arcuate surface read data after reducing the influence of foreign matter contained in at least one of the arcuate surface read data and the plane read data. According to the invention of claim 4, when the number (N) of time-series pixel values ​​acquired by each image sensor from the white reference arc surface is different from the number (M) of time-series pixel values ​​acquired by each image sensor from the white reference plane, the scales of the arc surface reading data and the plane reading data can be aligned. According to the invention of claim 5, it is possible to acquire arcuate surface reading data while reducing the influence of foreign matter adhering to the white reference arcuate surface. [Brief explanation of the drawings]

[0016] [Figure 1] 1 is a schematic diagram illustrating the configuration of an image reading apparatus according to an embodiment of the present invention. [Figure 2] 1 is a conceptual diagram showing a schematic shape of a cross section in the short side direction of a shading roller according to an embodiment of the present invention. [Figure 3] FIG. 2 is a perspective view of a shading roller according to the embodiment. [Figure 4]FIG. 10 is a diagram showing a state in which the shading roller is skewed with respect to the image sensor array. [Figure 5] FIG. 2 is a schematic diagram illustrating the configuration of a processor. [Figure 6] FIG. 10 is a diagram showing an example of arc surface read data. [Figure 7] FIG. 10 is a diagram showing an example of arc surface skeleton data. [Figure 8] FIG. 10 is a diagram showing an example of plane read data. [Figure 9] FIG. 10 is a diagram illustrating an example of planar skeleton data. [Figure 10] FIG. 10 is a diagram illustrating an example of difference data. [Figure 11] FIG. 10 is a diagram illustrating an example of shading data. [Figure 12] FIG. 10 is a diagram showing an example of arc surface read data in which the influence of a foreign object appears; [Figure 13] FIG. 10 is a diagram showing an example of intermediate arc surface read data. [Figure 14] FIG. 10 is a diagram showing an example of intermediate arc surface skeleton data. [Figure 15] FIG. 10 is a diagram showing an example of arc surface foreign substance data; [Figure 16] FIG. 10 is a diagram showing an example of planar foreign object data. [Figure 17] 10 is a conceptual diagram showing N pixel values ​​in time series relating to reflected light from a white reference arc surface output by each image sensor. FIG. [Figure 18] FIG. 10 is a conceptual diagram showing M pixel values ​​in time series relating to reflected light from a white reference plane output by each image sensor. [Figure 19] 10 is a conceptual diagram showing how an image sensor receives reflected light from a white reference arc surface of a rotating shading roller. FIG. [Figure 20] FIG. 10 is a diagram showing a state in which the shading roller is skewed with respect to the image sensor array. [Figure 21] FIG. 10 is a perspective view showing a state in which a shading roller having a circular cross section is skewed. [Figure 22] FIG. 10 is a diagram illustrating an example of shading data. DETAILED DESCRIPTION OF THE INVENTION

[0017] FIG. 1 is a schematic diagram of an image reading device 10 according to this embodiment. In each drawing in this specification, the X-axis direction indicates the main scanning direction, the Y-axis direction indicates the sub-scanning direction, and the Z-axis direction indicates the height direction. The image reading device 10 optically reads an original document, which is a medium such as paper (in other words, performs a scanning process) and forms image data representing the original document. The image reading device 10 is, for example, a scanner or a multifunction peripheral with a scanning function. As shown in FIG. 1, the image reading device 10 includes an original feed roller 12, a CIS 18 including a light source 14 and an image sensor array 16, a shading roller 20, a memory 22, and a processor 24.

[0018] The document feed rollers 12 transport the document along the document transport path P. In the example of Fig. 1, a plurality of pairs of document feed rollers 12 are provided, and each pair of document feed rollers 12 transports the document by rotating while sandwiching the document in its nip portion.

[0019] For scanning, the light source 14 emits light toward the original or the shading roller 20. A plurality of light sources 14 may be provided along the X axis.

[0020] The image sensor array 16 is composed of a plurality of image sensors (for example, a dozen or so) lined up in the X-axis direction. Each image sensor receives reflected light emitted from the light source 14 toward the original or the shading roller 20, and outputs a pixel value based on the received reflected light. The image sensor array 16 is disposed opposite the shading roller 20 across the conveying path P.

[0021] The shading roller 20 is a columnar component that extends in the longitudinal direction generally along the image sensor array 16. The shading roller 20 is provided opposite the image sensor array 16 across the transport path P.

[0022] A white reference surface is provided on the outer peripheral surface of the shading roller 20 to obtain shading data for shading correction. Shading correction is a correction to suppress fluctuations in pixel values ​​output by the image sensor array 16 that occur due to differences in the light intensity of the light sources 14 (including differences between light sources 14 and differences due to changes over time within the same light source 14) or differences in sensitivity of each image sensor (including differences between image sensors and differences due to changes over time within the same image sensor). The shading data is formed from pixel values ​​output by the image sensor array 16 after receiving light from the light source 14 that is reflected by the white reference surface of the shading roller 20.

[0023] Fig. 2 is a conceptual diagram showing the outline shape of a lateral cross section of the shading roller 20. As shown in Fig. 2, the shading roller 20 according to this embodiment has, on its outer peripheral surface, a white reference arc surface 30 that is white, has an arc-shaped lateral cross section, and extends in the longitudinal direction. The shading roller 20 also has, on its outer peripheral surface, a white reference flat surface 32 that is white, flat, and extends in the longitudinal direction.

[0024] The shading roller 20 is rotatable about a rotation axis R extending in the longitudinal direction, thereby making it possible to change the surface facing the image sensor array 16 (or the light source 14). In this embodiment, the cross-sectional shape of the white reference arc surface 30 is a semicircle with the rotation axis R as the center.

[0025] In addition to the white reference arc surface 30 and the white reference plane 32, the shading roller 20 may have flat planes 34 and 36 extending in the longitudinal direction on its outer peripheral surface. For example, the plane 34 is black, and by facing the image sensor array 16, the plane 34 is used to improve the accuracy of position detection of a document transported along the transport path P. Furthermore, for example, the plane 36 is used as a soiling surface. When a document passes between the image sensor array 16 and the shading roller 20, foreign matter (e.g., paper dust) may come off the document and adhere to the shading roller 20. To prevent such foreign matter from adhering to the white reference arc surface 30 and the white reference plane 32, the plane 36 is made to face the image sensor array 16 when the document passes between the image sensor array 16 and the shading roller 20.

[0026] In this embodiment, the distance from the rotation axis R to the white reference arc surface 30 is the same as the distance from the rotation axis R to the transport path P (see FIG. 1). As described above, the amount of light received by each image sensor included in the image sensor array 16 varies depending on the distance between the image sensor and the outer circumferential surface of the shading roller 20 or the original. Therefore, shading data can be obtained by making the distance from each image sensor to the original and the distance from each image sensor to the outer circumferential surface of the shading roller 20 the same. For the same reason, it is preferable that the distance from the rotation axis R to the white reference plane 32 is the same as the distance from the rotation axis R to the transport path P.

[0027] On the other hand, the distance from the rotation axis R to at least one of the white reference plane 32 and the planes 34 and 36 is shorter than the distance from the rotation axis R to the transport path P. For example, in this embodiment, the distance from the rotation axis R to the plane 36, which is the soiled surface, is shorter than the distance from the rotation axis R to the transport path P. As a result, by arranging the plane 36 to face the image sensor array 16, it is possible to prevent interference between the document being transported along the transport path P and the shading roller 20.

[0028] 3 is a perspective view of the shading roller 20 according to this embodiment. More specifically, a groove 38 extending in the longitudinal direction is provided in the circumferential center of the white reference arc surface 30 of the shading roller 20, and the white reference arc surface 30 is divided into two by the groove 38.

[0029] FIG. 4 is a diagram showing a state in which the shading roller 20 (20b in the example of FIG. 4) is skewed. As described above, it is ideal for the longitudinal direction of the shading roller 20 to be parallel to the main scanning direction (i.e., the alignment direction of the image sensor array 16 (16b in the example of FIG. 4)). However, due to factors such as the accuracy of installation of the shading roller 20, the longitudinal direction of the shading roller 20 and the alignment direction of the image sensor array 16 may not be parallel. In this embodiment, even if the longitudinal direction of the shading roller 20 and the alignment direction of the image sensor array 16 are not parallel and the shading roller 20 has an arcuate surface (white reference arcuate surface 30) on its outer circumferential surface, it is still possible to obtain appropriate shading data, as will be described below.

[0030] Returning to FIG. 1 , the image reading device 10 according to this embodiment has two combinations of a CIS 18 and a shading roller 20. That is, the image reading device 10 has a CIS 18a (i.e., the light source 14a and the image sensor array 16a) for scanning the front side of a document, a shading roller 20a for generating shading data for the CIS 18a, a CIS 18b (i.e., the light source 14b and the image sensor array 16b) for scanning the back side of a document, and a shading roller 20b for generating shading data for the CIS 18b. However, the image reading device 10 may have only one combination of a CIS 18 and a shading roller 20. In this embodiment, the shading data for each CIS 18 is generated in the same manner, and therefore, in the following description, no distinction will be made between the CIS 18a and the CIS 18b, and between the shading roller 20a and the shading roller 20b.

[0031] The memory 22 includes, for example, a hard disk drive (HDD), a solid state drive (SSD), an embedded multi media card (eMMC), a read only memory (ROM), a random access memory (RAM), etc. The memory 22 stores an image reading program for operating each unit of the image reading device 10.

[0032] 5 is a schematic diagram of the configuration of the processor 24. The processor 24 performs the functions of a read data acquisition unit 40, a difference data calculation unit 42, a shading data acquisition unit 44, and an image data formation unit 46 by using an image reading program stored in the memory 22. Below, these functions performed by the processor 24, as well as the shading data generation process performed by the processor 24, will be described.

[0033] The read data acquisition unit 40 rotates the shading roller 20 so that the white reference arc surface 30 of the shading roller 20 faces the image sensor array 16. In this state, the read data acquisition unit 40 causes the light source 14 to irradiate the white reference arc surface 30, and acquires data indicating pixel values ​​output by each image sensor included in the image sensor array 16, which is obtained when the image sensor array 16 receives the light reflected from the white reference arc surface 30. In this specification, this data is referred to as arc surface read data.

[0034] Fig. 6 is a diagram showing an example of arc surface read data. In Fig. 6 (and also in Figs. 7 to 13 described below), the horizontal axis represents the X-axis direction, i.e., the main scanning direction (of the image sensors aligned in that direction), and the vertical axis represents the pixel values ​​output by the image sensors. The pixel values ​​are, for example, luminance values.

[0035] FIG. 6 shows arc surface read data when the shading roller 20 is skewed. In the arc surface read data in FIG. 6, as the X coordinate increases, the pixel value output by the image sensor becomes extremely low. This is because the shading roller 20 is skewed and the outer peripheral surface of the shading roller 20 facing the image sensor array 16 is arc-shaped in cross section in the short side direction (white reference arc surface 30), so as the X coordinate increases, the distance between the image sensor and the outer peripheral surface of the shading roller 20 increases (see FIG. 21). If shading correction is performed using such arc surface read data as shading data, appropriate shading correction cannot be performed.

[0036] High-frequency noise resulting from the image sensor may be superimposed on the arc surface read data. In FIG. 6, the state where such high-frequency noise is superimposed is indicated by a thick line. The read data acquisition unit 40 may apply a low-pass filter to the arc surface read data to remove such high-frequency noise so that the difference data calculation unit 42, which will be described later, can calculate suitable difference data (described later). In this specification, data obtained by applying a low-pass filter to the arc surface read data is referred to as arc surface skeleton data. The arc surface skeleton data obtained by applying a low-pass filter to the arc surface read data shown in FIG. 6 is shown in FIG. 7.

[0037] It is not necessary for the read data acquisition unit 40 to apply a low-pass filter to the arc surface read data. For example, if there is not much high-frequency noise superimposed on the arc surface read data, the read data acquisition unit 40 does not need to apply a low-pass filter to the arc surface read data.

[0038] The read data acquisition unit 40 also rotates the shading roller 20 so that the white reference plane 32 of the shading roller 20 faces the image sensor array 16. In this state, the read data acquisition unit 40 causes the light source 14 to irradiate the white reference plane 32, and acquires data indicating pixel values ​​output by each image sensor included in the image sensor array 16, which is obtained when the image sensor array 16 receives the light reflected from the white reference plane 32. In this specification, this data is referred to as plane read data.

[0039] FIG. 8 is a diagram showing an example of planar scan data. FIG. 8 shows planar scan data when the shading roller 20 is skewed. Even if the shading roller 20 is skewed, if the outer peripheral surface of the shading roller 20 facing the image sensor array 16 is flat (the white reference plane 32), the outer peripheral surface of the shading roller 20 is flat, and therefore the distance between the image sensor and the outer peripheral surface of the shading roller 20 does not vary with changes in the X coordinate. Therefore, the effect of the skew of the shading roller 20 does not appear in the planar scan data. Note that in FIG. 8, the pixel values ​​at both ends of the image sensor array 16 are lower than those at the center of the X coordinate (main scanning direction), but this is due to factors such as the position of the light source 14, and is not due to the skew of the shading roller 20.

[0040] As described above, since the planar scan data is not affected by the skew of the shading roller 20, it is conceivable to perform shading correction using the planar scan data as shading. However, as will be described in detail later, if a foreign object is present between the outer peripheral surface of the shading roller 20 and the image sensor array 16, it is more difficult to reduce the influence of the foreign object in the planar scan data than in the arcuate surface scan data. Note that the concept of foreign object includes not only dirt and dust adhered to the shading roller 20, but also floating dust located between the outer peripheral surface of the shading roller 20 and the image sensor array 16. Therefore, the influence of the foreign object may appear in the planar scan data (the protruding portions in the vertical direction in FIG. 8 are the influence of the foreign object), and therefore, if the planar scan data is used as shading data to perform shading correction, appropriate shading correction may not be performed.

[0041] Like arcuate surface scan data, planar scan data may also be superimposed with high-frequency noise resulting from the image sensor. In Figure 8, the superimposed high-frequency noise is indicated by a thick line. Furthermore, the planar scan data in the example of Figure 8 is affected by foreign matter that existed between the white reference plane 32 and the image sensor array 16 when the planar scan data was acquired. The scan data acquisition unit 40 may apply a low-pass filter to the planar scan data to remove the effects of such high-frequency noise and foreign matter, enabling the difference data calculation unit 42 to calculate suitable difference data. In this specification, the data obtained by applying a low-pass filter to the planar scan data is referred to as planar skeleton data. Figure 9 shows the planar skeleton data obtained by applying a low-pass filter to the planar scan data shown in Figure 8.

[0042] It is not necessary for the read data acquisition unit 40 to apply a low-pass filter to the plane read data. For example, the read data acquisition unit 40 does not need to apply a low-pass filter to the plane read data when the plane read data is not significantly affected by high-frequency noise and foreign matter.

[0043] The difference data calculation unit 42 calculates difference data representing the difference between the arc surface read data and the plane read data based on the arc surface read data and the plane read data acquired by the read data acquisition unit 40. When the read data acquisition unit 40 acquires arc surface skeleton data based on the arc surface read data and plane skeleton data based on the plane read data, the difference data calculation unit 42 calculates the difference data based on the arc surface skeleton data and the plane skeleton data. In this embodiment, the difference data calculation unit 42 calculates the difference data by dividing the arc surface skeleton data (see FIG. 7) by the plane skeleton data (see FIG. 9).

[0044] 10 is a diagram showing an example of calculated difference data. As described above, the arc surface read data or arc surface skeleton data is data that has been affected by the skew of the shading roller 20, while the planar read data or planar skeleton data is data that has not been affected by the skew of the shading roller 20. Therefore, it can be said that the difference data is data that represents the amount of variation in pixel value at each main scanning direction position due to the effect of the skew of the shading roller 20.

[0045] The difference data calculation unit 42 may calculate the difference data by dividing the arc surface read data by the plane read data.

[0046] The shading data acquisition unit 44 acquires shading data as corrected arc surface read data based on the arc surface read data acquired by the read data acquisition unit 40 (see FIG. 6) and the difference data calculated by the difference data calculation unit 42 (see FIG. 10). As described above, the arc surface read data or arc surface skeleton data is data affected by the skew of the shading roller 20, and the difference data is data representing the amount of variation in pixel values ​​at each main scanning direction position due to the effect of the skew of the shading roller 20. Therefore, the shading data acquisition unit 44 can acquire arc surface read data (i.e., shading data as corrected arc surface read data) in which the effect of the skew of the shading roller 20 has been reduced, based on the arc surface read data and the difference data. In this embodiment, the shading data acquisition unit 44 acquires shading data by dividing the arc surface skeleton data by the difference data.

[0047] FIG. 11 is a diagram showing an example of shading data. As can be seen by comparing the arc surface scan data shown in FIG. 6 with the shading data shown in FIG. 11, the influence of skew of the shading roller 20 is reduced in the shading data. Note that, because high-frequency noise is superimposed on the arc surface scan data, high-frequency noise is also superimposed on the shading data in FIG. 11. High-frequency noise caused by the image sensor is also superimposed on the image data obtained by the image sensor array 16 actually reading the document. Therefore, the superimposition of high-frequency noise on the shading data may enable more suitable shading correction.

[0048] The image data forming unit 46 irradiates light from the light source 14 onto the document transported along the transport path P, and acquires document reading data obtained by the image sensor array 16 receiving the light reflected from the document.

[0049] Furthermore, the image data forming unit 46 corrects the acquired original scan data based on the shading data (see FIG. 11) acquired by the shading data acquiring unit 44, and acquires corrected original scan data. Since a conventional method can be adopted for correcting the original scan data based on the shading data, a detailed description thereof will be omitted here. For example, in the shading data of FIG. 11, the pixel values ​​at both ends are lower than those at the center in the main scanning direction, so the image data forming unit 46 corrects the original scan data based on the shading data so as to raise the pixel values ​​at both ends compared to those at the center in the main scanning direction.

[0050] Furthermore, the image data forming unit 46 forms image data corresponding to the document based on the corrected document read data. The formed image data is stored in the memory 22 or transmitted to a device outside the image reading device 10 (for example, a personal computer or a server) by a communication unit (not shown) included in the image reading device 10.

[0051] As described above, according to this embodiment, even if the shading roller 20 is skewed and the outer surface of the shading roller 20 for obtaining shading data is an arc surface (a white reference arc surface 30), shading data can be obtained with reduced effects due to the skewed shading roller 20.

[0052] If a foreign object is present between the white reference arc surface 30 and the image sensor array 16 when acquiring the arc surface read data, the effect of the foreign object will also appear in the arc surface read data. FIG. 12 shows the arc surface read data in which the effect of the foreign object appears. As with the plane read data shown in FIG. 8, the protruding portions in the vertical direction in FIG. 12 are the effect of the foreign object. If the arc surface read data is affected by a foreign object, when the shading data acquisition unit 44 acquires shading data based on the arc surface read data (see FIG. 12) and the difference data (see FIG. 10) as in the method described above, the effect of the foreign object will remain in the shading data. Therefore, in order to suppress the effect of the foreign object in the shading data, the processor 24 may perform the following processing.

[0053] As in the basic embodiment described above, the shading data acquisition unit 44 corrects the arcuate surface read data (see FIG. 12) based on the difference data (see FIG. 10) to acquire data in which the influence of the skew of the shading roller 20 is reduced, although the influence of the foreign matter that existed between the white reference arcuate surface 30 and the image sensor array 16 is not eliminated. In this specification, this data is referred to as intermediate arcuate surface read data. In this embodiment, the shading data acquisition unit 44 acquires the intermediate arcuate surface read data by dividing the arcuate surface read data by the difference data. An example of intermediate arcuate surface read data is shown in FIG. 13.

[0054] The shading data acquisition unit 44 applies a low-pass filter to the intermediate arc surface read data. In this specification, the intermediate arc surface read data after application of the low-pass filter is referred to as intermediate arc surface skeleton data. Since the influence of foreign matter appearing in the intermediate arc surface read data is a high-frequency component, applying a low-pass filter to the intermediate arc surface read data reduces the influence of high-frequency noise and foreign matter from the intermediate arc surface read data. Figure 14 shows an example of intermediate arc surface skeleton data.

[0055] The shading data acquisition unit 44 obtains arc surface foreign matter data indicating the image sensor affected by the foreign matter and the degree of influence of the foreign matter on the image sensor in the arc surface read data by taking the difference between the intermediate arc surface read data (see Figure 13) and the intermediate arc surface skeleton data (see Figure 14).

[0056] FIG. 15 shows an example of arc surface foreign substance data. In FIG. 15 (and FIG. 16, described later), the horizontal axis represents the X-axis direction, i.e., the main scanning direction (image sensors aligned in the X-axis direction), and the vertical axis represents the difference in pixel values. The intermediate arc surface read data is data in which the influence of foreign substances is apparent, and the intermediate arc surface skeleton data is data in which the influence of foreign substances has been reduced. Therefore, it can be said that the arc surface foreign substance data is data that primarily represents the influence of foreign substances. In FIG. 15, the main scanning direction position (in other words, the image sensor) where the difference in pixel values ​​appears is the image sensor affected by the foreign substance, and the difference in pixel values ​​represents the degree of influence of the foreign substance on that image sensor. For example, the example in FIG. 15 shows that image sensors Sb, Sd, and Sf are affected by foreign substances. The arc surface foreign matter data obtained by subtracting the intermediate arc surface read data from the intermediate arc surface skeleton data may include the effects of not only foreign matter but also high-frequency noise. However, because the amount of fluctuation in pixel values ​​due to high-frequency noise is small, the effects of high-frequency noise are not shown in Figure 15.

[0057] In addition, the shading data acquisition unit 44 obtains planar foreign object data indicating the image sensors affected by foreign objects and the degree of influence of the foreign objects on the image sensors in the planar read data by calculating the difference between the planar read data (see FIG. 8) and the planar skeleton data (see FIG. 9) obtained by applying a low-pass filter to the planar read data. An example of planar foreign object data is shown in FIG. 16. For example, the example in FIG. 16 shows that image sensors Sa, Sc, and Se are affected by foreign objects. Note that the planar foreign object data may also include the influence of high-frequency noise in addition to the influence of foreign objects, but because the amount of variation in pixel values ​​due to high-frequency noise is small, the influence of high-frequency noise is not shown in FIG. 16.

[0058] Next, the shading data acquisition unit 44 acquires shading data as corrected arcuate surface read data by combining the intermediate arcuate surface read data and the planar read data based on the arcuate surface foreign object data and the planar foreign object data. Specifically, the shading data acquisition unit 44 acquires shading data by combining pixel values ​​of the intermediate arcuate surface read data for image sensors in which the influence of foreign objects indicated by the arcuate surface foreign object data is smaller than that of foreign objects indicated by the planar foreign object data, and by combining pixel values ​​of the planar read data for image sensors in which the influence of foreign objects indicated by the planar foreign object data is smaller than that of foreign objects indicated by the arcuate surface foreign object data. For example, if the arcuate surface foreign object data is data such as that shown in FIG. 15 and the planar foreign object data is data such as that shown in FIG. 16, the shading data acquisition unit 44 acquires shading data by combining pixel values ​​of the arcuate surface read data for image sensors Sa, Sc, and Se, and pixel values ​​of the planar read data for image sensors Sb, Sd, and Sf.

[0059] The read data acquiring unit 40 may acquire noise-reduced arc surface read data. For example, the read data acquiring unit 40 may cause each image sensor to output pixel values ​​multiple times while irradiating light from the light source 14 onto the white reference arc surface 30, and may calculate the output pixel value of each image sensor based on the multiple pixel values ​​output in chronological order by each image sensor.

[0060] For example, the read data acquisition unit 40 repeatedly outputs pixel values ​​relating to light reflected from the white reference arc surface 30 N times for each image sensor to acquire N pixel values ​​in time series. FIG. 17 shows how image sensor S1 outputs N pixel values ​​PV11, PV12, PV13, ... PV1N, image sensor S2 outputs N pixel values ​​PV21, PV22, PV23, ... PV2N, and image sensor S3 outputs N pixel values ​​PV31, PV32, PV33, ... PV1N. Then, the read data acquisition unit 40 determines a first representative pixel value for each image sensor by calculating the sum of the N pixel values ​​using a predetermined first correction coefficient. For example, the read data acquisition unit 40 can determine the first representative pixel value by dividing the sum of the N pixel values ​​by N, which is the first correction coefficient. The first representative pixel value for each image sensor is then acquired as arc surface read data. This makes it possible to reduce noise in the arc surface read data that is suddenly output by the image sensor, for example.

[0061] Similarly, the read data acquisition unit 40 may acquire plane read data with reduced noise. For example, the read data acquisition unit 40 may cause each image sensor to output pixel values ​​multiple times while irradiating light from the light source 14 onto the white reference plane 32, and may calculate the output pixel values ​​of the image sensors based on the multiple pixel values ​​output by each image sensor in chronological order.

[0062] For example, the read data acquisition unit 40 repeatedly outputs pixel values ​​related to light reflected from the white reference plane 32 for each image sensor M times to acquire M pixel values ​​in time series. FIG. 18 shows how image sensor S1 outputs M pixel values ​​PV11, PV12, PV13, ... PV1M, image sensor S2 outputs M pixel values ​​PV21, PV22, PV23, ... PV2M, and image sensor S3 outputs M pixel values ​​PV31, PV32, PV33, ... PV1M. Then, the read data acquisition unit 40 determines a second representative pixel value for each image sensor by calculating the sum of the M pixel values ​​using a predetermined second correction coefficient. For example, the read data acquisition unit 40 can determine the second representative pixel value by dividing the sum of the M pixel values ​​by M, which is the second correction coefficient. The second representative pixel value for each image sensor is then acquired as planar read data. This makes it possible to reduce noise that is suddenly output by an image sensor, for example, in plane read data as well.

[0063] In this embodiment, two types of read data, namely, arc surface read data and planar read data, are used to acquire shading data, and the number of times (N) that each image sensor outputs a pixel value when acquiring the arc surface read data may be different from the number of times (M) that each image sensor outputs a pixel value when acquiring the planar read data. In this case, the first correction coefficient used when calculating the first representative pixel value that is the basis of the arc surface read data will be different from the second correction coefficient used when calculating the second representative pixel value that is the basis of the planar read data.

[0064] In addition, the read data acquisition unit 40 may rotate the shading roller 20 and repeatedly output pixels related to reflected light from the white reference arc surface 30 for each image sensor multiple times (N times), thereby acquiring N pixel values ​​in a time series.

[0065] FIG. 19 is a conceptual diagram showing how the image sensor S1 receives light reflected from the white reference arc surface 30 of the rotating shading roller 20. Assume that a foreign substance F is attached to the white reference arc surface 30. In this case, when the rotation angle of the shading roller 20 is in the state shown in FIG. 19(b), the foreign substance F is present at a position facing the image sensor S1, and the image sensor S1 outputs a pixel value affected by the foreign substance F. On the other hand, when the rotation angle of the shading roller 20 is in the state shown in FIG. 19(a) or (c), the foreign substance F is not present at a position facing the image sensor S1, and the image sensor S1 outputs a pixel value not affected by the foreign substance F. In other words, this can reduce the number of pixel values ​​affected by the foreign substance among the N pixel values ​​acquired by the image sensor S1. Therefore, by determining the first representative pixel value based on these N pixel values, the influence of the foreign substance F on the first representative pixel value can be reduced.

[0066] In addition, in this embodiment, the cross-sectional shape of the white reference arc surface 30 is semicircular with the rotation axis R as the center, so even if the shading roller 20 is rotated, the distance between the image sensor S1 and the white reference arc surface 30 does not change.

[0067] Although the embodiment of the present invention has been described above, the present invention is not limited to the above embodiment, and various modifications are possible without departing from the spirit of the present invention.

[0068] In the above embodiment, the processor refers to a processing device in a broad sense, and includes at least one of a general-purpose processing device (e.g., a CPU (Central Processing Unit)) and a dedicated processing device (e.g., a GPU (Graphics Processing Unit), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or a programmable logic device). The processor may not be a single processing device, but may be configured by the cooperation of multiple processing devices located at physically separate locations.

[0069] (Addendum) (((1))) a shading roller having a columnar shape extending in a longitudinal direction and rotatable around a rotation axis extending in the longitudinal direction, the shading roller having, on its outer peripheral surface, a white reference arc surface which is white, has an arc-shaped cross section in the short side direction, and extends in the longitudinal direction, and a white reference flat surface which is white, flat, and extends in the longitudinal direction; an image sensor array arranged opposite the shading roller across a document transport path, the image sensor array comprising a plurality of image sensors arranged in a main scanning direction, for receiving reflected light of light irradiated from a light source toward the document or the shading roller; a processor; Equipped with The processor: calculating difference data representing a difference between the arc surface read data and the plane read data based on arc surface read data indicating pixel values ​​output by each of the image sensors, which is obtained by the image sensor array receiving the reflected light from the white reference plane, and plane read data indicating pixel values ​​output by each of the image sensors, which is obtained by the image sensor array receiving the reflected light from the white reference plane; obtaining corrected arcuate surface reading data based on the arcuate surface reading data and the difference data; correcting document read data obtained by the image sensor array receiving the reflected light from the document based on the corrected arcuate surface read data to obtain corrected document read data; forming image data corresponding to the original based on the corrected original read data; An image reading device characterized by: (((2))) The processor: The difference data is calculated based on arc surface skeleton data obtained by applying a low-pass filter to the arc surface read data and on plane surface skeleton data obtained by applying a low-pass filter to the plane read data. The image reading device according to (((1))) is characterized in that: (((3))) The processor: correcting the arc surface reading data based on the difference data to obtain intermediate arc surface reading data; By calculating a difference between the intermediate arc surface read data and intermediate arc surface skeleton data obtained by applying a low-pass filter to the intermediate arc surface read data, arc surface foreign matter data is obtained, which indicates the image sensor affected by a foreign matter that existed between the white reference arc surface and the image sensor array when the arc surface read data was obtained, and the degree of influence of the foreign matter on the image sensor; By calculating a difference between the plane read data and the plane skeleton data, plane foreign object data is obtained, which indicates the image sensor affected by a foreign object that existed between the white reference plane and the image sensor array when acquiring the plane read data, and the degree of influence of the foreign object on the image sensor; For the image sensor in which the influence of the foreign matter indicated by the arcuate surface foreign matter data is smaller than the influence of the foreign matter indicated by the planar foreign matter data, pixel values ​​of the intermediate arcuate surface read data are used, and for the image sensor in which the influence of the foreign matter indicated by the planar foreign matter data is smaller than the influence of the foreign matter indicated by the arcuate surface foreign matter data, pixel values ​​of the planar read data are used to synthesize the data, thereby obtaining the corrected arcuate surface read data. The image reading device according to (((2))) is characterized in that: (((4))) The processor: determining a first representative pixel value for each of the image sensors by dividing a sum of N pixel values ​​in a time series obtained by outputting pixel values ​​relating to the reflected light from the white reference arc surface N times by a first correction coefficient, and acquiring the first representative pixel value for each of the image sensors as the arc surface read data; a second representative pixel value is determined for each of the image sensors by dividing a sum of M pixel values ​​in time series, obtained by outputting pixel values ​​relating to the reflected light from the white reference plane by a second correction coefficient different from the first correction coefficient, for each of the image sensors, M times, which is different from N, and acquiring the second representative pixel value for each of the image sensors as the plane read data; The image reading device according to any one of ((1))) to (((3))) is characterized in that: (((5))) Each of the image sensors repeatedly outputs pixel values ​​relating to the reflected light from the white reference arc surface of the rotating shading roller N times, thereby obtaining N pixel values ​​in time series. The image reading device according to (((4))) is characterized in that:

[0070] According to the invention (((1))), even if the shading roller is skewed, shading data can be obtained that has reduced effects of skew compared to shading data obtained by an image sensor array receiving reflected light from the white reference arc surface, which is the outer surface of the shading roller. According to the invention (((2))), it is possible to calculate the difference data after reducing the influence of foreign matter or high frequency noise contained in the arcuate surface read data and the plane read data. According to the invention related to ((3)), it is possible to obtain corrected arcuate surface read data after reducing the influence of foreign matter contained in at least one of the arcuate surface read data and the plane read data. According to the invention (((4))), when the number (N) of time-series pixel values ​​acquired by each image sensor from a white reference arc surface is different from the number (M) of time-series pixel values ​​acquired by each image sensor from a white reference plane, the scales of the arc surface read data and the plane read data can be matched. According to the invention (((5))), the influence of foreign matter adhering to the white reference arc surface can be reduced and arc surface reading data can be obtained. [Explanation of symbols]

[0071] 10 image reading device, 12 document feed roller, 14 light source, 16 image sensor array, 18 CIS, 20 shading roller, 22 memory, 24 processor, 30 white reference arc surface, 32 white reference plane, 40 read data acquisition unit, 42 difference data calculation unit, 44 shading data acquisition unit, 46 image data formation unit.

Claims

1. a shading roller having a columnar shape extending in a longitudinal direction and rotatable about a rotation axis extending in the longitudinal direction, the shading roller having, on an outer peripheral surface thereof, a white reference arc surface which is white, has an arc-shaped short-side cross section and extends in the longitudinal direction, and a white reference flat surface which is white, flat, and extends in the longitudinal direction; an image sensor array including a plurality of image sensors arranged in a main scanning direction, the image sensor array being disposed opposite the shading roller across a document transport path and configured to receive reflected light of light irradiated from a light source toward the document or the shading roller; A processor; Equipped with The processor, calculating difference data representing a difference between the arc surface read data and the plane read data based on arc surface read data indicating pixel values ​​output by each of the image sensors, the data being obtained by the image sensor array receiving the reflected light from the white reference plane, and plane read data indicating pixel values ​​output by each of the image sensors, the data being obtained by the image sensor array receiving the reflected light from the white reference plane; obtaining corrected arcuate surface reading data based on the arcuate surface reading data and the difference data; correcting document read data obtained by receiving the reflected light from the document by the image sensor array based on the corrected arcuate surface read data to obtain corrected document read data; forming image data corresponding to the document based on the corrected document read data; 1. An image reading apparatus comprising:

2. The processor, calculating the difference data based on arc surface skeleton data obtained by applying a low-pass filter to the arc surface read data and on plane skeleton data obtained by applying a low-pass filter to the plane read data; 2. The image reading apparatus according to claim 1, wherein the image reading apparatus comprises:

3. The processor, correcting the arcuate surface reading data based on the difference data to obtain intermediate arcuate surface reading data; a difference between the intermediate arcuate surface read data and intermediate arcuate surface skeleton data obtained by applying a low-pass filter to the intermediate arcuate surface read data is calculated to obtain arcuate surface foreign matter data indicating the image sensor affected by a foreign matter that existed between the white reference arcuate surface and the image sensor array when obtaining the arcuate surface read data, and the degree of influence of the foreign matter on the image sensor; By calculating a difference between the planar read data and the planar skeleton data, planar foreign object data is obtained that indicates the image sensor affected by a foreign object that existed between the white reference plane and the image sensor array when obtaining the planar read data, and the degree of the effect of the foreign object on the image sensor; For the image sensor in which the influence of the foreign object indicated by the arc surface foreign object data is smaller than the influence of the foreign object indicated by the planar foreign object data, pixel values ​​of the intermediate arc surface read data are used, and for the image sensor in which the influence of the foreign object indicated by the planar foreign object data is smaller than the influence of the foreign object indicated by the arc surface foreign object data, pixel values ​​of the planar read data are used to synthesize the data, thereby obtaining the corrected arc surface read data.

3. The image reading apparatus according to claim 2, wherein the image reading apparatus comprises:

4. The processor, determining a first representative pixel value for each of the image sensors by dividing a sum of N pixel values ​​in a time series, the sum being obtained by outputting pixel values ​​relating to the reflected light from the white reference arc surface N times by a first correction coefficient, and acquiring the first representative pixel value for each of the image sensors as the arc surface read data; a second representative pixel value is determined by dividing a sum of M pixel values ​​in a time series, the sum being obtained by outputting pixel values ​​relating to the reflected light from the white reference plane by the image sensor M times, different from N, by a second correction coefficient different from the first correction coefficient, for each of the image sensors, and the second representative pixel value for each of the image sensors is obtained as the plane read data.

2. The image reading apparatus according to claim 1, wherein the image reading apparatus comprises:

5. each of the image sensors repeatedly outputs pixel values ​​relating to the reflected light from the white reference arc surface of the rotating shading roller N times to obtain N pixel values ​​in a time series; 5. The image reading apparatus according to claim 4,

Images