Image reading device, anomaly detection program, and anomaly detection method

Abstract

When figures, text, etc. 19 are visible within the area of ​​the original image, the distances WL and WR from the center position CL to each edge 24, 25 may be misidentified as edges, making it impossible to accurately determine the distances WL and WR. [Solution] The image reading device of the present invention comprises a transport unit and a reading unit 5 that reads an image of a medium, and a control unit 6 that processes the read image 14 and controls the transport of the medium. The control unit performs an abnormality determination process on an abnormality determination region 13 set for the leading edge region of the medium. The abnormality determination process includes a first step of processing the pixels of the read data included in the abnormality determination region to a state in which pixels within the medium region 20 and pixels outside the medium region 21 can be distinguished, a second step of obtaining the maximum distance WL from the center position in the width direction of the medium to one edge 24 and the maximum distance WR to the other edge 25, and a third step of determining whether or not there is an abnormality in the transport state based on each of the maximum distances.

Description

[Technical Field]

[0001] The present invention relates to an image reading apparatus that reads an image of a document. The present invention also relates to an abnormality determination program executed in an image reading apparatus. The present invention further relates to an abnormality determination method for an image reading apparatus. [Background Art]

[0002] As an example of the prior art for this type of apparatus, the one described in Patent Document 1 can be cited. Patent Document 1 discloses an image reading apparatus provided with a document protection mechanism. This document discloses the following contents. In a document image in an abnormality determination area, a distance WL from the center position in the medium width direction to one edge in the medium width direction and a distance WR from the center position to the other edge in the medium width direction are acquired, and a value R which is a ratio or difference between the distance WR and the distance WL is compared with a predetermined threshold R0 to determine whether an abnormality exists. [Prior Art Documents] [Patent Documents]

[0003] [Patent Document 1] Japanese Unexamined Patent Publication No. 2021-34842 [Summary of the Invention] [Problem to be Solved by the Invention]

[0004] However, when figures, characters, or the like are shown in the area of the document image, for example, when acquiring the distances WL and WR from the center position to each edge, there is a risk that images of the figures, characters, or the like are misjudged as the edges, and the distances WL and WR cannot be obtained accurately. [Means for Solving the Problem]

[0005] To solve the above problems, the image reading device according to the present invention comprises a transport unit that transports a medium along a transport path, a reading unit provided on the transport path that reads an image of the medium, and a control unit that processes the reading data read by the reading unit and controls the transport of the medium along the transport path, wherein the control unit sets an abnormality determination area for the leading edge region of the medium and performs an abnormality determination process for the abnormality determination area, and the abnormality determination process comprises a first step of processing the pixels of the reading data included in the abnormality determination area in a state in which pixels within the medium region and pixels outside the medium region can be distinguished, a second step after the first step of obtaining the maximum distance WL from the center position in the medium width direction which is a direction intersecting the transport direction of the medium to one edge, and the maximum distance WR from the center position to the other edge, and a third step of comparing a value R which is the ratio or difference between the maximum distance WR and the maximum distance WL with a predetermined threshold R1.

[0006] Furthermore, the abnormality determination program according to the present invention is an abnormality determination program executed in an image reading device comprising: a transport unit that transports a medium along a transport path; a reading unit provided on the transport path that reads an image of the medium; and a control unit that processes the read data read by the reading unit and controls the transport of the medium along the transport path, wherein the abnormality determination program sets an abnormality determination area for the leading edge region of the medium, performs an abnormality determination process on the abnormality determination area, and the abnormality determination process comprises: a first step of processing the pixels of the read data included in the abnormality determination area in a state in which pixels within the medium region and pixels outside the medium region can be distinguished; a second step after the first step of obtaining the maximum distance WL from the center position in the medium width direction which is a direction intersecting the transport direction of the medium to one edge, and the maximum distance WR from the center position to the other edge; and a third step of comparing a value R, which is the ratio or difference between the maximum distance WR and the maximum distance WL, with a predetermined threshold R1.

[0007] Furthermore, the abnormality determination method according to the present invention is an abnormality determination method in an image reading device comprising: a transport unit that transports a medium along a transport path; a reading unit provided on the transport path that reads an image of the medium; and a control unit that processes the read data read by the reading unit and controls the transport of the medium along the transport path, wherein an abnormality determination area is set for the leading edge region of the medium, an abnormality determination process is performed on the abnormality determination area, the abnormality determination process is characterized by processing the pixels of the read data included in the abnormality determination area in a state in which pixels within the medium region and pixels outside the medium region can be distinguished, obtaining the maximum distance WL from the center position in the medium width direction which is a direction intersecting the transport direction of the medium to one edge, and the maximum distance WR from the center position to the other edge, and comparing a value R which is the ratio or difference between the maximum distance WR and the maximum distance WL with a predetermined threshold R1. [Brief explanation of the drawing]

[0008] [Figure 1] A schematic diagram of the image reading device according to the embodiment. [Figure 2] Block configuration of the control unit according to the embodiment. [Figure 3] A diagram illustrating image processing for an example of an abnormality detection region in the embodiment. [Figure 4] A diagram illustrating image processing for another example of the abnormality detection region of the embodiment. [Figure 5] A flowchart showing the control flow of the abnormality detection processing mode of the embodiment. [Figure 6] A diagram illustrating image processing for another example of the abnormality detection region of the embodiment. [Figure 7] A diagram illustrating image processing for another example of the abnormality detection region of the embodiment. [Figure 8] A diagram illustrating image processing for another example of the abnormality detection region of the embodiment. [Modes for carrying out the invention]

[0009] The present invention will now be described in general terms. To solve the above problems, an image reading device according to a first aspect of the present invention comprises: a transport unit that transports a medium along a transport path; a reading unit provided on the transport path that reads an image of the medium; and a control unit that processes the reading data read by the reading unit and controls the transport of the medium along the transport path. The control unit sets an abnormality determination area for the leading edge region of the medium and performs an abnormality determination process on the abnormality determination area. The abnormality determination process includes: a first step of processing the pixels of the reading data included in the abnormality determination area to a state in which pixels within the medium region and pixels outside the medium region can be distinguished; a second step of obtaining, after the first step, the maximum distance WL from the center position in the medium width direction, which is a direction intersecting the transport direction of the medium, to one edge, and the maximum distance WR from the center position to the other edge; and a third step of comparing a value R, which is the ratio or difference between the maximum distance WR and the maximum distance WL, with a predetermined threshold R1.

[0010] According to this embodiment, the pixels of the read data included in the abnormality determination area are processed in a way that allows for the identification of pixels within the media area determined based on the first and second edges, and pixels outside the media area. This processing makes it possible to accurately search for the boundary line between the inner and outer parts of the media area, thereby enabling the precise identification of the positions of one edge and the other edge. As a result, the maximum distance WL from the center position to one edge and the maximum distance WR from the center position to the other edge in the media width direction can be accurately determined.

[0011] An image reading device according to a second aspect of the present invention is an aspect dependent on the first aspect, comprising an edge identification step of binarizing the pixels of the reading data included in the abnormality determination area to identify one edge and the other edge, wherein the first step is characterized in that the inside and outside of the media area within the abnormality determination area are determined based on the one edge and the other edge identified in the edge identification step.

[0012] According to this embodiment, the first step includes an edge identification step that identifies one edge and the other edge, and the first step is configured to define the inside and outside of the media area within the abnormality determination area based on the one edge and the other edge identified in the edge identification step. This makes it easy to define the inside and outside of the media area within the abnormality determination area.

[0013] An image reading device according to a third aspect of the present invention is an aspect dependent on the second aspect, characterized in that the edge identification step performs a binarization process in the abnormality determination region, in which pixels in the pixels of the read data where the brightness difference is greater than or equal to a threshold R2 are given a first color, and pixels in the parts where the brightness difference is less than the threshold R2 are given a second color.

[0014] According to this embodiment, the binarization process based on the brightness difference divides the pixels within the abnormality detection area into a first color portion and a second color portion, thereby facilitating subsequent image processing.

[0015] An image reading device according to a fourth aspect of the present invention is an aspect dependent on the third aspect, characterized in that the first step, after the binarization process, performs a homochromatization process in the abnormality determination region to make all pixels between one edge and the other edge the first color, thereby achieving the ability to identify.

[0016] According to this embodiment, the first step, after the binarization process, performs a homochromatization process in the abnormality determination region to make all pixels between one edge and the other edge the first color. This makes it easy to achieve a state in which pixels within the media region and pixels outside the media region can be distinguished.

[0017] An image reading device according to a fifth aspect of the present invention is an aspect dependent on the third aspect, characterized in that, before the binarization process, the pixels of the abnormality determination region are set to the second color. Furthermore, this embodiment can also be made dependent on the fourth embodiment.

[0018] According to this embodiment, before the first step, the configuration is such that pixels in the abnormality determination area are set to the second color. This makes it possible to reduce the risk of erroneous search caused by dust and vertical streaks when searching for and specifying the respective positions of the one edge and the other edge.

[0019] An image reading apparatus according to a sixth aspect of the present invention is an embodiment dependent on the fourth aspect, characterized in that, after the same-color processing, if there is the one edge and the other edge that have a spaced edge positioned away from the downstream side of the abnormality determination area in the conveyance direction, a second-stage same-color processing is performed to set all pixels between the spaced edge and one side of the abnormality determination area in a direction intersecting the conveyance direction to a first color. Furthermore, this embodiment can also be made dependent on the fifth embodiment.

[0020] According to this embodiment, if there is the one edge and the other edge that have a spaced edge positioned away from the downstream side of the abnormality determination area, the configuration is such that a second-stage same-color processing is performed to set all pixels between the spaced edge and the one side edge of the abnormality determination area to the first color. This makes it possible to cope even with a conveyance posture in which one of the left and right sides of the medium is not within the abnormality determination area.

[0021] An image reading apparatus according to a seventh aspect of the present invention is an embodiment dependent on the third aspect, characterized in that, when the medium is thin paper, the abnormality determination processing specifies a leading edge of the medium in the image after the binarization processing in the abnormality determination area, and forms the one edge and the other edge at both end positions of the leading edge; and the first step achieves the identifiability by performing same-color processing to set all pixels between the formed one edge and the other edge to the first color in the abnormality determination area. The reading unit irradiates light from upstream to downstream or from downstream to upstream in the transport direction of the medium, so that contrast is easily obtained and edges are easily extracted at the leading edge and trailing edge of the medium. However, contrast is difficult to obtain and edges are difficult to extract at the side edges of the medium. This tendency depends on the degree of light transmission of the light-transmitting medium. In this specification, "thin paper" refers to a medium in which the above tendency is strong and the side edges are difficult to extract. Furthermore, this embodiment may be subordinate to the fifth embodiment.

[0022] According to this embodiment, the system is configured to create one edge and the other edge within the abnormality detection area based on the positions of both ends of the tip edge, and to perform a homogeneous coloring process to make all the pixels between the created edge and the other edge the first color. This allows the system to handle even thin paper and to accurately determine the maximum distance WL and the maximum distance WR.

[0023] An abnormality determination program according to the eighth aspect of the present invention is an abnormality determination program executed in an image reading device comprising: a transport unit for transporting a medium along a transport path; a reading unit provided on the transport path for reading an image of the medium; and a control unit for processing the reading data read by the reading unit and controlling the transport of the medium along the transport path, wherein the abnormality determination program sets an abnormality determination area for the leading edge region of the medium, performs an abnormality determination process on the abnormality determination area, and the abnormality determination process comprises: a first step of processing the pixels of the reading data included in the abnormality determination area in a state in which pixels within the medium region and pixels outside the medium region can be distinguished; a second step after the first step of obtaining the maximum distance WL from the center position in the medium width direction which is a direction intersecting the transport direction of the medium to one edge, and the maximum distance WR from the center position to the other edge; and a third step of comparing a value R, which is the ratio or difference between the maximum distance WR and the maximum distance WL, with a predetermined threshold R1. According to this embodiment, the same effects as in the first embodiment can be obtained.

[0024] An abnormality determination method according to an eighth aspect of the present invention is an abnormality determination method in an image reading device comprising: a transport unit for transporting a medium along a transport path; a reading unit provided on the transport path for reading an image of the medium; and a control unit for processing the reading data read by the reading unit and controlling the transport of the medium along the transport path, wherein an abnormality determination area is set for the leading edge region of the medium, an abnormality determination process is performed on the abnormality determination area, the abnormality determination process is characterized by processing the pixels of the reading data included in the abnormality determination area to a state in which pixels within the medium region and pixels outside the medium region can be distinguished, obtaining the maximum distance WL from the center position in the medium width direction which intersects the transport direction of the medium to one edge, and the maximum distance WR from the center position to the other edge, and comparing a value R which is the ratio or difference between the maximum distance WR and the maximum distance WL with a predetermined threshold R1. According to this embodiment, the same effects as in the first embodiment can be obtained.

[0025] [Embodiment] Hereinafter, an image reading device according to an embodiment of the present invention will be specifically described with reference to Figures 1 to 8. In the following explanation, the three mutually orthogonal axes will be referred to as the X-axis, Y-axis, and Z-axis, as shown in each figure. The direction indicated by the arrows on the three axes (X, Y, Z) is the positive direction, and the opposite direction is the negative direction. The Z-axis direction corresponds to the vertical direction, i.e., the direction in which gravity acts, with the +Z direction indicating vertically upwards and the -Z direction indicating vertically downwards. The X-axis and Y-axis directions correspond to the horizontal direction. The +Y direction indicates the forward direction of the image reader, and the -Y direction indicates the backward direction of the image reader. The +X direction indicates the right direction of the image reader, and the -X direction indicates the left direction of the image reader.

[0026] [Embodiment] <Overall Overview of the Image Reading Device> The image reading device 1 of this embodiment is a scanner, as an example. As shown in Figure 1, the image reading device 1 includes a transport unit 4 that directs the medium 2, which is the original document, along the transport path 3 in the transport direction F, a reading unit 5 provided on the transport path 3 that reads the image of the medium 2, and a control unit 6 that processes the reading data read by the reading unit 5 and controls the transport of the medium 2 along the transport path 3. The transport unit 4 is a transport roller that is driven by rotational power transmitted from a motor (not shown) and applies a transport force to the medium 2 in the transport direction F. In this embodiment, only one located adjacent to the upstream of the reading unit 5 is shown, and the others are omitted from the illustration. That is, multiple transport rollers are arranged in the transport path 3 along the transport direction F from the medium placement unit (not shown), where the medium 2 is set for reading by the reading unit 5. In addition, multiple transport rollers (not shown) are also arranged in the transport path 3 downstream of the reading unit 5. The transport rollers of the transport unit 4 are arranged in pairs symmetrically with respect to the center position CL in the media width direction (X-axis direction). That is, in this embodiment, the center position CL becomes the reference position of the media being transported, and media of different width sizes are transported through the reading area of ​​the reading unit 5 without the center position CL changing.

[0027] The reading unit 5 is equipped with a contact-type image sensor module (CISM). The reading unit 5 illuminates the reading position 9 with light and receives reflected light from the transported medium 2, thereby reading the image of the medium 2 as it passes through the reading position 9 in the transport direction F. Since the reading position 9 is long in the width direction (X-axis direction) of the medium 2, it can also be called a reading line. Upstream of the reading unit 5 is a tip detection unit 8 that detects the passage of the leading edge 7 of the medium 2. The tip detection unit 8 determines the position of the leading edge 7 by rotating a lever as the leading edge 7 of the medium 2 passes by. The reading unit 5 performs a reading operation based on the position information of the leading edge 7 of the medium 2 detected by the tip detection unit 8. In Figure 1, reference numeral 10 denotes a glass document tray that supports the medium 2 during reading.

[0028] As shown in Figure 2, the control unit 6 receives position information of the tip 7 of the medium 2 and other sensing information from the tip detection unit 8, and controls the execution of various operations, including the reading operation by the reading unit 5, the transport operation of the medium 2 by the transport unit 4, and the operation of other drive units not shown in the figure. The control unit 6 has an abnormality determination processing mode 11 that performs an abnormality determination processing, which will be described later. The abnormality determination processing mode 11 executes abnormality determination processing based on the abnormality determination processing program 12. This processing is always performed when reading the medium 2. In other words, this processing is not limited to thin paper or thick paper, and may be performed when reading any type of medium. Furthermore, this processing may be performed even when the medium 2 is thin paper. Here, the control unit 6 includes a CPU, flash ROM, and RAM. The CPU performs various calculations according to various programs, such as the abnormality detection processing program 12, stored in the flash ROM, and controls the operation of the entire image reading device 1. Flash ROM, an example of a storage means, is a non-volatile memory that can be read from and written to. RAM, another example of a storage means, is used as the CPU's work area and also temporarily stores various information.

[0029] As shown in Figure 3, the abnormality determination processing mode 11 of the control unit 6 is configured to set an abnormality determination region 13 for the leading edge region of the medium 2 and to perform abnormality determination processing on the abnormality determination region 13. In this embodiment, the abnormality determination area 13 is a horizontally elongated rectangle enclosed by an upstream side 131 on the upstream side in the transport direction F, a downstream side 132 on the downstream side, and one side 133 and the other side 134 on the left and right sides. The abnormality determination area 13 contains a read image 14 of the leading edge region of the medium 2, i.e., an image corresponding to the read data. The read image 14 is an image read from the time the leading edge of the transported medium 2 passes the reading position 9 until a predetermined time T has elapsed. The predetermined time T is the length of time required to acquire the image data necessary for abnormality determination, and is set in advance. In Figure 3, the four sides that make up the abnormality detection area 13—the upstream side 131, the downstream side 132, one side 133, and the other side 134—are depicted as visible lines. However, this is to indicate the extent of the abnormality detection area 13, and the lines of these four sides do not actually exist in the image. Specifically, in states ST1 and ST2 of Figure 3, the four sides—the upstream side 131, the downstream side 132, one side 133, and the other side 134—do not exist in a visible state. Figure 3 shows the case where medium 2 is transported without any abnormalities such as skew.

[0030] State ST1 in Figure 3 shows the state in which the read image 14 within the abnormality detection region 13 has been processed into a binarized image 15. This binarized image 15 is obtained by assigning a first color C1 (white in the figure) to pixels in the read image 14 where the brightness difference is greater than or equal to the threshold R2 in the abnormality detection region 13, and a second color C2 (black in the figure) to pixels where the brightness difference is less than the threshold R2. The threshold R2 is set in advance so that the first color C1 and the second color C2 can be clearly distinguished. In this embodiment, all pixels within the abnormality detection region 13 are set to the second color C2, and the read image 14 is then taken into the abnormality detection region 13 and binarized.

[0031] In the binarization based on the luminance difference described above, for example, if the medium 2 is a white sheet of paper and has letters or figures written on it in black or red, the luminance difference will be large at the leading edge and side edges of the medium 2, as well as at the boundaries of the images such as letters and figures within the medium 2, i.e., the luminance difference will be greater than or equal to the threshold R2. On the other hand, in the white parts of the medium 2 where no letters or figures are written, there is almost no luminance difference, so the luminance difference will be less than the threshold R2. Therefore, as shown in state ST1 in Figure 3, the leading edge 16 of the medium 2, the positions of one side edge 17 and the other side edge 18 in the width direction, and the parts corresponding to the letters and figures 19 will be the first color C1 "white", and the other parts will be the second color C2 "black". Furthermore, the image of the letter "A" shown as one of the 19 examples of letters and shapes is presented here as an image with a clear outline, but in reality, the outline is not this clear, and adjacent "A" letters tend to be connected in the image.

[0032] In this embodiment, the abnormality detection processing mode 11 of the control unit 6 performs an edge identification step on the binarized image 15 shown in state ST1 of Figure 3. The edge identification step is a process in which, in the abnormality determination region 13, one edge 24 and the other edge 25 are identified in the media width direction (X-axis direction), which intersects with the transport direction F of the media 2, based on the binarized image 15 obtained by binarizing the read image 14 included in the abnormality determination region 13.

[0033] <Edge Identification Step: For one edge> On the other hand, the identification of edge 24 is performed as follows: (1) For the binarized image 15 of state ST1 in Figure 3, the first color C1 is searched for by moving the confirmation position toward the center position CL in the rightward direction (-X direction) while checking the color of each pixel, starting from the corner position PL which is the downstream edge 132 of the anomaly detection region 13 and the position of one of the left-hand sides 133. If the first color C1 is found, the position where it was found is taken as the position of one of the edges 24. If the first color C1 is not found even after moving to the center position CL, it is assumed that one of the edges does not exist at that position. (2) Next, the search position is moved by one pixel towards the upstream edge 131 (-Y direction), and the same search as in (1) is performed. (3) Furthermore, the search position is moved by one pixel towards the upstream edge 131 (-Y direction), and the same search as in (1) is performed. (4) Repeat steps (2) and (3) over the entire length of one side 133 of the anomaly detection region 13 to determine the overall position of one edge 24 within the anomaly detection region 13. This allows the overall position of one edge 24 within the anomaly detection area 13 to be determined.

[0034] <Edge Identification Step: Regarding the Other Edge> The identification of the other edge 25 is performed as follows. (1) For the binarized image 15 of state ST1 in Figure 3, the first color C1 is searched for by moving the confirmation position toward the center position CL to the left (+X direction) while checking the color of each pixel, starting from corner position PR, which is the position of the downstream edge 132 of the anomaly detection region 13 and the other side edge 134 that forms the right edge. If the first color C1 is hit, the position where it was hit is taken as the position of the other edge 25. If the first color C1 is not hit even after moving to the center position CL, it is assumed that the other edge 25 does not exist at that position. (2) Next, the search position is moved by one pixel towards the upstream edge 131 (-Y direction), and the same search as in (1) is performed. (3) Furthermore, the search position is moved by one pixel towards the upstream edge 131 (-Y direction), and the same search as in (1) is performed. (4) Repeat steps (2) and (3) over the entire length of the other side 134 of the abnormality detection region 13 to determine the overall position of the other edge 25 within the abnormality detection region 13. This allows the overall position of the other edge 25 within the anomaly detection region 13 to be determined.

[0035] The edge 24 and edge 25 identified by the above edge identification step refer to the positions of one side edge 17 and the other side edge 18 in the width direction of the medium 2 when the medium 2 is transported without any abnormalities such as skew, as shown in Figure 3. However, as will be described later (Figures 6 and 8), if the degree of skew of the medium 2 increases, part or all of one of the edges 24 and the other edge 25 may become the leading edge 16 of the medium 2. In either case, by identifying the positions of one edge 24 and the other edge 25 within the abnormality detection area 13, it is possible to define the area inside the media area 20 and the area outside the media area 21. In Figure 3, the area between one edge 24 and the other edge 25 is the media area 20. That is, the position of the leading edge 16 is the straight line connecting the furthest downstream position of one edge 24 in the transport direction F and the furthest downstream position of the other edge 25. Therefore, the area enclosed by one edge 24, the other edge 25, and the leading edge 16 becomes the media area 20 within the abnormality detection area 13.

[0036] In this embodiment, the abnormality determination processing mode 11, as the first step of abnormality determination processing, processes the pixels of the read data included in the abnormality determination area 13 to a state in which pixels within the media area 20 and pixels outside the media area 21 can be distinguished. State ST2 in Figure 3 shows a state in which a first step has been performed on State ST1 in Figure 3, in which pixels 22 within the media area 20 and pixels 23 outside the media area 21 are distinguishable. In State ST1 in Figure 3, the media area 20 contains both the portion of pixels 22 of the first color C1 and the portion of pixels 22 of the second color C2. Specifically, the first step is configured to perform a homochromatization process on the binarized image 15 after the binarization process, in which all pixels 22 in the media region 20, which is the region between one edge 24 and the other edge 25 in the abnormality detection region 13, are changed to the first color C1. This homochromatization process changes the state from ST1 in Figure 3 to ST2 in Figure 3, making it possible to distinguish between pixels in the media region 20 and pixels 22 outside the media region 21.

[0037] In this embodiment, the abnormality detection processing mode 11 executes a second step after the first step. The second step is to obtain the maximum distance WL from the center position CL in the medium width direction (X-axis direction), which is the direction intersecting the transport direction F of the medium 2, to one edge 24, and the maximum distance WR from the center position CL to the other edge 25. The process of obtaining the maximum distances WL and WR will be described later. The abnormality detection processing mode 11 executes the third step following the second step. The third step compares a value R, which is the ratio or difference between the maximum distance WR and the maximum distance WL obtained in the second step, with a predetermined threshold R1. This comparison determines whether the transport state of the medium 2 is abnormal or not. That is, if the value R is greater than the threshold R1, it is determined to be "abnormal" and the transport of the medium 2 is stopped. If the value R is less than the threshold R1, it is determined to be "normal" and the transport of the medium 2 continues.

[0038] <Obtaining the maximum distance WL> The process of obtaining the maximum distance WL is performed in state ST2 in Figure 3. Starting from the center position CL, the second color C2 is searched for by moving the confirmation position to the left (+X direction) while checking the color pixel by pixel on the upstream edge 131 of the abnormality detection area 13 towards one of the left side edges 133. The position where the second color is hit is the position of one of the edges 24, and the maximum distance WL is obtained from this. In Figure 3, one of the edges 24 is the position of one of the side edges 17 of the medium 2, so the maximum distance WL is the distance from the center position CL to one of the side edges 17 of the medium 2. <Obtaining the maximum distance WR> The process of obtaining the maximum distance WR involves, in state ST as shown in Figure 3, moving from the center position CL toward the other side edge 134 to the right, checking the color pixel by pixel on the upstream edge 131 of the abnormality detection area 13, and moving the check position to the right (-X direction) to search for the second color C2. The position where the second color is hit is the position of the other edge 25, and the maximum distance WR is obtained from this. In Figure 3, the other edge 25 is the position of the other side edge 18 of the medium 2, so the maximum distance WR is the distance from the center position CL to the other side edge 18 of the medium 2.

[0039] [If the medium is thin paper] The reading unit 5 irradiates light from upstream to downstream or from downstream to upstream in the transport direction F of the medium 2. As a result, the leading edge 16 and trailing edge of the medium 2 are easily contrasted and the edges are easily extracted. However, the side edges 17 and 18 of the medium 2 are difficult to contrast and the edges are difficult to extract. This tendency is more likely to occur when the medium 2 is translucent, and depends on the degree of translucency. In this specification, "thin paper" refers to a medium 2 in which the above tendency is strong and the side edges 17 and 18 are difficult to extract. State ST1 in Figure 4 shows the state of the binarized image 15, which is the result of the reading image 14 within the abnormality detection region 13 being binarized. Since the medium 2 is thin paper, one edge 24 and the other edge 25 are not extracted.

[0040] In the abnormality detection processing mode 11, if the medium 2 is thin paper, the leading edge 16 of the medium 2 is identified in the binarized image 15 in the abnormality detection region 13, and further processing is performed to create one edge 24 and the other edge 25 at both ends 31 and 32 of the leading edge 16. The one edge 24 and the other edge 25 are created in a direction perpendicular to the longitudinal direction of the leading edge 16. ST2 in Figure 4 shows the state after the processing to create one edge 24 and the other edge 25 has been performed. That is, the parts of one edge 24 and the other edge 25 have been changed to the first color C1. The first step is to perform a homogeneity process in the abnormality detection region 13, which sets all pixels between the created edge 24 and the other edge 25 to the first color C1. This homogeneity process makes it possible to distinguish between pixels within the media region 20 and pixels outside the media region 21, as shown in ST3 in Figure 4.

[0041] Next, the control flow of the abnormality detection process in abnormality detection processing mode 11 will be explained based on Figure 5. The abnormality detection processing mode 11 executes control based on the abnormality detection processing program 12. First, in step S1, the reading unit 5 begins reading the medium 2 that is being transported along the transport path 3 in the transport direction F by the transport unit 4. Next, the process proceeds to step S2, where it is determined whether a predetermined time T has elapsed since the transported medium 2 passed the reading position 9 of the reading unit 5. If the predetermined time T has elapsed (Yes), the process proceeds to step S3. In step S3, the abnormality detection area 13 is set. Next, the process proceeds to step S4, where the reading image 14 read by the reading unit 5 within a predetermined time T is binarized to create a binarized image 15. The state ST1 in Figure 3 and the state ST1 in Figure 4 are the same as described above.

[0042] Next, in step S5, one edge 24 and the other edge 25 are identified. Specifically, the processing in the edge identification step described above is performed. The process proceeds to step S6 to determine whether one edge 24 and the other edge 25 have been identified. Here, for example, if the medium 2 is thin paper, it may be difficult to identify one edge 24 and the other edge 25 as described above. If the edges have been identified (Yes), the process proceeds to step S7. In step S7, a homogeneous colorization process is performed to make all pixels in the media region 20, which is the area between one edge 24 and the other edge 25, the first color C1. This homogeneous colorization process changes the state from ST1 in Figure 3 to ST2 in Figure 3, making it possible to distinguish between pixels in the media region 20 and pixels outside the media region 21. In other words, the first step described above is executed.

[0043] Next, the process proceeds to step S8, where, after the same-color processing, it is determined whether or not there are separated edges 30. The separated edges 30 will be explained later. If there are no separated edges 30 (No), the process proceeds to step S9. In step S9, the process of obtaining the maximum distance WL and maximum distance WR is executed, and the maximum distance WL and maximum distance WR are determined. In other words, the second step described above is executed. Next, the process proceeds to step S10, where the third step described above is executed, and a determination is made as to whether or not there is an abnormality in the transport state of the medium 2. In the case of Figure 3, the value R is less than the threshold R1, and it is determined to be "normal".

[0044] If an edge cannot be identified in step S6 (No), proceed to step S11. In step S11, a process is performed to identify the leading edge 16 of the medium 2 in the binarized image 15 in the abnormality detection region 13. That is, the position of the leading edge 16 is identified. Next, the process proceeds to step S12, where one edge 24 and the other edge 25 are created at both ends 31 and 32 of the tip edge 16. State ST2 in Figure 4 is this state as described above. Next, we proceed to step S7. The subsequent control flow is the same as described above, so it will be omitted. In the case of Figure 4, the value R becomes less than the threshold R1, and is judged as "normal".

[0045] <Example of abnormality detection 1> Next, we will explain the case where the abnormality detection processing mode 11 determines that the transport status of the medium 2 is "abnormal," based on Figures 6 to 8. First, let's explain an anomaly detection example 1 based on Figure 6. Figure 6 shows the case where medium 2 moves at an angle during transport. Specifically, it shows the case where medium 2 is transported at an angle to the left. State ST1 in Figure 6 is shown as a binarized image 15, which is obtained by binarizing the read image 14 within the abnormality detection region 13. That is, State ST1 in Figure 6 is a read image 14 that has been binarized in a state where the medium 2 is being transported at an angle to the left, and a portion of the right side of the leading edge 16 of the medium 2 has passed the downstream edge 132 of the abnormality detection region 13.

[0046] State ST2 in Figure 6 shows the state after the same-color processing of the first step (step S7 in Figure 5) has been performed on the binarized image 15. Figure 6 differs from Figure 3 only in that the medium 2 is transported at an angle; all other aspects are the same as in Figure 3, so the explanation of the same parts will be omitted. By performing the second and third steps described above on the image of state ST2 in Figure 6, it is determined whether or not there is an abnormality in the transport state of medium 2 (steps S9 and S10 in Figure 5). In the case of Figure 6, the maximum distance WR becomes larger than the maximum distance WL due to the skew, so the value R becomes greater than or equal to the threshold R1, and is determined to be "abnormal".

[0047] <Example of abnormality detection 2> Next, an example of abnormality detection 2 will be explained based on Figure 7. Figure 7 shows the case where a medium with a narrow width, 2, is transported diagonally to the left at a position slightly to the left of the transport path 3. When medium 2 is transported with the center position CL as the reference position described above, medium 2 is guided on both sides by edge guides (not shown), and transport usually begins with medium 2 positioned symmetrically with respect to the center position CL, so the situation in Figure 7 is unlikely to occur. However, if medium 2 is transported without using the edge guides, the situation in Figure 7 may occur. State ST1 in Figure 7 is shown as a binarized image 15 obtained by binarizing the read image 14, which is in a state where the medium 2 is transported diagonally to the left from a leftward position, and the leading edge 16 has passed the downstream edge 132 of the abnormality detection region 13.

[0048] State ST2 in Figure 7 shows the state in which the same-color processing of the first step (step S7 in Figure 5) has been performed on the binarized image 15. Figure 7 differs from Figure 3 only in that the narrow medium 2 is being transported at an angle and the leading edge 16 has passed the downstream edge 132 of the abnormality detection region 13. All other points are the same as in Figure 3, so the explanation of the same parts will be omitted. By performing the second and third steps described above on the image of state ST2 in Figure 7, it is determined whether or not there is an abnormality in the transport state of medium 2 (steps S9 and S10 in Figure 5). In the case of Figure 7, the maximum distance WR is considerably smaller than the maximum distance WL, so the value R becomes greater than or equal to the threshold R1, and is determined to be "abnormal".

[0049] <Example of abnormality detection 3> Next, we will explain an anomaly detection example 3 based on Figure 8. Figure 8 shows a case where a medium with a narrow width, 2, is transported at a greater oblique angle to the left than in the case of Figure 7, at a position closer to the left of the transport path 3. State ST1 in Figure 8 shows that the medium 2 is transported at a large diagonal angle to the left from a position slightly to the left, and a portion of the right side of the leading edge 16 has passed the downstream edge 132 of the abnormality detection region 13. Furthermore, a portion of the left side of the leading edge 16 of the medium 2 and one side edge 17 are outside one side edge 133 of the abnormality detection region 13. In other words, state ST1 in Figure 8 is shown as a binarized image 15, which is obtained by binarizing the read image 14 in this state.

[0050] State ST2 in Figure 8 shows the state in which the same-color processing of the first step (step S7 in Figure 5) has been performed on the binarized image 15. In this example, since one side edge 17 of the medium 2 is not within the abnormality detection region 13, the one edge 24 identified in the edge identification step described above is made up of a pair of a leading edge 16 and a separated edge 30 that corresponds to the leftmost part of the part corresponding to characters, figures, etc. 19. The separated edge 30 is located away from the downstream edge 132 of the abnormality detection region 13. As shown in state ST2 in Figure 8, in the same-color processing of step S7, not all pixels 22 within the medium region 20 become the first color C1. Therefore, if there is a separated edge 30 (step S8 in Figure 5), the second stage of homogeneity processing is performed. That is, the second stage of homogeneity processing is performed to make all pixels 22 between the separated edge 30 and one side 133 of the anomaly detection region 13 the first color C1. This corresponds to step S13 in the flowchart of Figure 5.

[0051] State ST3 in Figure 8 shows the state in which all pixels 22 in the media area 20 have been processed to the first color C1 by the second stage of homogeneity processing. By performing the second and third steps described above on the image of state ST3 in Figure 8, it is determined whether or not there is an abnormality in the transport state of medium 2 (steps S9 and S10 in Figure 5). In the case of Figure 8, the maximum distance WL is too large to be determined, and the maximum distance WR is small, so the value R is greater than or equal to the threshold R1, and it is determined to be "abnormal".

[0052] <Description of the effects of Embodiment 1> (1) In this embodiment, the read pixels 14 included in the abnormality determination area 13 are configured to be processed in a state in which pixels 22 inside the media area 20 and pixels 23 outside the media area 21 can be distinguished based on one edge 24 and the other edge 25. This processing makes it possible to accurately search for the position of the boundary line between the inside and outside parts of the media area, thereby making it possible to accurately identify the positions of one edge 24 and the other edge 25. As a result, the maximum distance WL from the center position CL to one edge 24 and the maximum distance WR from the center position CL to the other edge 25 in the media width direction (X-axis direction) can be accurately determined.

[0053] (2) In this embodiment, there is also an edge identification step that identifies one edge 24 and the other edge 25, and the first step is configured to define the media area 20 and the area outside the media area 21 within the abnormality determination area 13 based on the one edge 24 and the other edge 25 identified in the edge identification step. This makes it easy to define the media area 20 and the area outside the media area 21 within the abnormality determination area 13. (3) In addition, in this embodiment, the pixels within the abnormality detection region 13 are divided into a first color C1 portion and a second color C2 portion by binarization processing based on the brightness difference, which facilitates subsequent image processing.

[0054] (4) In this embodiment, the first step is to perform a homochromatization process in the abnormality determination region 13 after the binarization process in step S4, in which all pixels between one edge 24 and the other edge 25 are made the first color C1. This makes it easy to achieve a state in which pixels 22 inside the media region 20 and pixels 23 outside the media region 21 can be distinguished.

[0055] (5) In addition, in this embodiment, the pixels of the abnormality detection region 13 are configured to the second color C2 before the binarization process. This reduces the risk of missearch due to dust or vertical lines when searching for and identifying the positions of one edge 24 and the other edge 25.

[0056] (6) In addition, in this embodiment, if there is a separated edge 30 between one edge 24 and the other edge 25 that is located away from the downstream edge 132 of the abnormality determination region 13, a second stage of homogeneity processing is performed to make all pixels 22 between the separated edge 30 and one side edge 133 of the abnormality determination region 13 the first color C1. This makes it possible to handle transport orientations in which one of the left and right edges of the medium 2 is not within the abnormality determination region 13.

[0057] (7) In this embodiment, based on the positions 31 and 32 at both ends of the leading edge 16 of the medium 2, one edge 24 and the other edge 25 are created within the abnormality detection area 13, and all pixels 22 between the created edge 24 and the other edge 25 are set to the first color C1 in a homogeneous coloring process. This allows the medium 2 to be used even if it is thin paper, and the maximum distance WL and the maximum distance WR can be determined with high accuracy.

[0058] [Other embodiments] The image reading device 1, abnormality determination processing program 12, and abnormality determination method according to the present invention are based on having the configuration of the embodiments described above, but it is of course possible to make partial changes or omissions to the configuration without departing from the spirit of the present invention. For example, in the first step of the abnormality detection process, a same-coloring process (processing to make it identifiable) may be performed in the X direction between one edge 24 and the other edge 25, and then the same-coloring process may be performed from the leading edge 16 (downstream end) of the medium toward the upstream end (for example, the upstream end within the medium region 20, or the upstream edge 131). In other words, the same-coloring process is performed in both the X and Y directions. In this way, the abnormality detection process can be executed in a single process without branching based on the thickness of the medium 2, etc. Furthermore, when performing the same-color processing along the Y direction as described above, for example, if a portion of the leading edge 16 is not within the abnormality detection area 13, the downstream edge 132 portion corresponding to the portion of the leading edge 16 that extends beyond the abnormality detection area 13 may be treated as the leading edge 16 and processed accordingly. In this way, the same-color processing can be easily performed even if the medium 2 is not contained within the abnormality detection area 13 due to skew or the like. Furthermore, when performing the same-color processing along the Y direction as described above, the same-color processing may be performed from one edge 24 or the other edge 25 toward the upstream end in the -Y direction. In this way, the same-color processing can be easily performed even if the medium 2 is in a skewed or other condition. Furthermore, in the first step of the abnormality detection process, if, for example, one of the edges 24 and the other edge 25 is not present within the abnormality detection region 13, the side 133 or the other side 134 of the abnormality detection region 13 on the side where the aforementioned edge is not present may be treated as an edge, and the same-color processing may be performed in the X direction. In this way, the same-color processing can be easily performed even if the medium 2 is not contained within the abnormality detection region 13 due to skew or the like. The above embodiments may be combined as appropriate. [Explanation of Symbols]

[0059] 1...Image reading device, 2...Medium, 3...Transport path, 4...Transport unit, 5...Reading unit, 6...Control unit, 7...Front end of media, 8...Front end detection unit, 9...Reading position, 10...Document glass 11...Anomaly detection mode, 12...Anomaly detection processing program, 13...Anomaly detection area 131...Upstream side, 132...Downstream side, 133...One side, 134...The other side 14... Read image, 15... Binarized image, 16... Front edge of the medium, 17...One side edge, 18...The other side edge, 19...Text, shapes, etc. 20... Pixels within the media area, 21... Pixels outside the media area, 22... Pixels within the media area 23...Pixels outside the media area, 24...One edge, 25...The other edge, 30... Separated edges, 31, 32... Positions at both ends, C1... First color, C2... Second color CL...Center position, PL...Corner position, PR...Corner position

Claims

1. A transport unit that transports the medium along the transport path, A reading unit provided in the transport path for reading the image of the medium, The system includes a control unit that processes the read data read by the reading unit and controls the transport of the medium along the transport path, The control unit, An abnormality detection area is set for the leading edge region of the aforementioned medium. An abnormality determination process is performed on the abnormality determination area. The aforementioned abnormality determination process is: A first step involves processing the pixels of the read data included in the abnormality determination area in a way that allows for the identification of pixels within the media area and pixels outside the media area. Following the first step, a second step is taken to obtain the maximum distance WL from the center position in the media width direction, which is a direction intersecting the media transport direction, to one edge, and the maximum distance WR from the center position to the other edge. The third step involves comparing a value R, which is the ratio or difference between the maximum distance WR and the maximum distance WL, with a predetermined threshold R1. An image reading device characterized by the following:

2. An image reading device according to claim 1, The process includes an edge identification step in which the pixels of the read data included in the abnormality determination region are binarized to identify one edge and the other edge. The first step determines the area within the media region and the area outside the media region within the abnormality determination region based on the one edge and the other edge identified in the edge identification step. An image reading device characterized by the following:

3. An image reading device according to claim 2, The edge identification step involves performing a binarization process in the abnormality determination region, where pixels in the read data where the brightness difference is greater than or equal to a threshold R2 are designated as a first color, and pixels in the region where the brightness difference is less than the threshold R2 are designated as a second color. An image reading device characterized by the following:

4. An image reading device according to claim 3, The first step is to achieve identification by performing a homochromatization process in the abnormality determination region after the binarization process, in which all pixels between one edge and the other edge are made the first color. An image reading device characterized by the following:

5. An image reading device according to claim 3, Before the first step, set the pixels of the abnormality detection area to the second color. An image reading device characterized by the following:

6. An image reading device according to claim 4, After the same-color processing, if there is a separated edge between one edge and the other edge that is located away from the downstream edge in the transport direction of the abnormality determination region, a second stage of same-color processing is performed to make all pixels between the separated edge and one side of the abnormality determination region in a direction intersecting the transport direction the first color. An image reading device characterized by the following:

7. An image reading device according to claim 3, The aforementioned abnormality determination process is: In the binarized image within the abnormality detection region, the leading edge of the medium is identified. The one edge and the other edge are formed at both ends of the aforementioned tip edge. The first step is to achieve identification by performing a homogeneity process in the abnormality determination region, in which all pixels between the created edge and the other edge are made the first color. An image reading device characterized by the following:

8. A transport unit that transports the medium along the transport path, A reading unit provided in the transport path for reading the image of the medium, An abnormality determination program executed in an image reading device comprising a control unit that processes the reading data read by the reading unit and controls the transport of the medium along the transport path, An abnormality detection area is set for the leading edge region of the aforementioned medium, and an abnormality detection process is performed on the abnormality detection area. The aforementioned abnormality determination process is: A first step involves processing the pixels of the read data included in the abnormality determination area in a way that allows for the identification of pixels within the media area and pixels outside the media area. Following the first step, a second step is taken to obtain the maximum distance WL from the center position in the media width direction, which is a direction intersecting the media transport direction, to one edge, and the maximum distance WR from the center position to the other edge. The third step involves comparing a value R, which is the ratio or difference between the maximum distance WR and the maximum distance WL, with a predetermined threshold R1. An anomaly detection program characterized by the following:

9. A transport unit that transports the medium along the transport path, A reading unit provided in the transport path for reading the image of the medium, An abnormality detection method in an image reading device comprising a control unit that processes the reading data read by the reading unit and controls the transport of the medium along the transport path, An abnormality detection area is set for the leading edge region of the aforementioned medium, and an abnormality detection process is performed on the abnormality detection area. The aforementioned abnormality determination process is: The pixels of the read data included in the abnormality determination area are processed to a state in which pixels within the media area and pixels outside the media area can be distinguished. After the second step, the maximum distance WL from the center position in the media width direction, which is the direction intersecting the media transport direction, to one edge, and the maximum distance WR from the center position to the other edge are obtained. The value R, which is the ratio or difference between the maximum distance WR and the maximum distance WL, is compared with a predetermined threshold R1. An abnormality detection method characterized by the following.

Images