JP2015211321A - Reading apparatus and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a technique for improving accuracy of detecting edges of an original document.SOLUTION: A scanner 100 includes a reading head 21 for reading an image and acquires image data 51 by reading an original document. The scanner 100 further acquires edge data, which is information on edges of the original documents, from the acquired image data 51. The scanner 100 further determines whether an end part in a principle scanning direction of the original document exceeds a reading range in a longer direction of the reading head 21 on the basis of the acquired edge data.

Description

  The present invention relates to a reading device and a program for reading an image of a document. More specifically, the present invention relates to a technique for detecting an edge of a document serving as a boundary between a document and a background member in a read image.

  2. Description of the Related Art Conventionally, a technique for reading an image of a document and detecting an edge of the document serving as a boundary between the document and a background member among the read images is known. For example, Patent Document 1 discloses an image reading apparatus having a function of detecting an edge of a document among read images, and removes the shadow of the document in the background portion from the read image using the function. Technology is disclosed.

JP 2009-16904 A

  However, the conventional technique described above has the following problems. That is, when the document size is large or the document is inclined, the edge of the document may be outside the reading range. In this case, the middle of the document may be erroneously determined as the edge of the document.

  The present invention has been made to solve the above-described problems of the prior art. That is, the problem is to provide a technique for improving the detection accuracy of the edge of the document.

  A reading apparatus for solving this problem includes a reading unit that reads an image and a control unit, and the control unit reads the image by the reading unit that causes the reading unit to read the image. An edge acquisition process for acquiring edge data, which is edge information of the document, from the image data, and an edge in the main scanning direction of the document is based on the edge data acquired in the edge acquisition process. A determination process for determining whether or not the reading range is exceeded is performed.

  The reading device disclosed in this specification acquires edge data, which is information about the edge of a document, from the read image data. Then, based on the acquired edge data, it is determined whether or not the end of the document in the main scanning direction exceeds the reading range of the reading unit.

  In other words, the reading device disclosed in this specification, in addition to acquiring edge data, which is information on the edge of the document, determines whether the end of the document in the main scanning direction exceeds the reading range of the reading unit based on the edge data. Judgment of whether or not. Therefore, by using the result of the determination, an erroneous determination that determines the position of the edge that is not the end of the document as the position of the end of the document is compared with the case where the position of the end of the document is determined only by the edge data. Can be expected.

  Further, the control unit executes an extraction process for extracting a leading edge portion of the document based on the edge data acquired in the edge acquisition process, and the leading edge portion of the document is extracted in the determination process. If the leading edge of the document is within the range from the limit of the reading range of the reading unit to a specified distance, it may be determined that the reading range of the reading unit is exceeded. If the leading end corner of the document is not extracted, the end of the document may be outside the reading range. Also, when it is difficult to accurately determine the position of the end of the document from the acquired edge data even when the leading edge of the document is within the range from the limit of the reading range of the reading unit to the specified distance. There is. For this reason, it is preferable to determine that the reading range of the reading unit is exceeded.

  Further, the control unit executes an inclination acquisition process for acquiring an inclination of the document based on the edge data acquired in the edge acquisition process, and the determination process acquires the inclination acquisition process. Based on the inclination, it may be determined whether or not the end of the document in the main scanning direction exceeds the reading range of the reading unit. By taking into account the inclination of the original, a position outside the reading range can be detected from the middle of the original in the sub-scanning direction. Therefore, it can be determined whether or not the reading range of the reading unit is exceeded more accurately.

  In the edge acquisition process, the control unit acquires the edge data in the main scanning direction at a predetermined frequency in the sub-scanning direction, and is determined to exceed the reading range of the reading unit in the determination process. In this case, the frequency of acquiring the edge data in the boundary area that is an area within the predetermined range from the excess position that is the position exceeding the reading range may be higher than the frequency in the area other than the boundary area. . According to this configuration, it is possible to increase the accuracy of the edge in the boundary region, while it is possible to increase the edge acquisition speed in the region other than the boundary region.

  In the edge acquisition process, the control unit may acquire the edge data at a different frequency according to at least one of a characteristic and a state of the document acquired using the edge data. There is at least one of the characteristics and state of the document as information of the document to be acquired, and the required accuracy differs depending on the type. For this reason, it is preferable to achieve a balance between high accuracy and high speed according to the type of information of the document. For example, the shape of the document, the size of the document, and the position of the punch hole correspond to the characteristics of the document. The document state corresponds to, for example, document tilt and rotational skew. For example, high accuracy is required for the skew state of the document, but high accuracy is not required for the position of the punch hole or the document size.

  In addition, the control unit includes edge data of a specific side that is a side on the end side that is determined not to exceed the reading range in the determination process among the edge data acquired in the edge acquisition process. It is preferable to execute an information acquisition process for acquiring at least one of the characteristics and the state of the document. Using the edge data on the side of the edge that is determined not to exceed the reading range, information on the document that is at least one of the characteristics and state of the document is acquired, that is, the edge determined to have exceeded the reading range. By obtaining the document information without using the edge data on the side of the section side, the accuracy of the document information can be improved.

  In the information acquisition process, when a negative determination is made in the determination process, the inclination of the document is acquired using the edge data on the leading edge side of the document and the edge data on the specific side, and the determination process is performed. In the case where the affirmative determination is made in the above, it is preferable to include processing for obtaining the inclination of the document using the edge data on the leading end side of the document. If the side edge data can be used, the inclination of the document can be acquired with high accuracy.

  In addition, in the information acquisition process, the control unit uses the edge data of the specific side to acquire the rotation skew amount when a negative determination is made in the determination process, and an affirmative determination is made in the determination process. In this case, the edge data on the specific side may not be used for obtaining the rotation skew amount. If the side edge data can be used, the rotational skew amount can be acquired with high accuracy.

  Further, in the information acquisition process, when the determination is negative in the determination process, the control unit uses the edge data of the specific side to acquire the shape of the document, and is positively determined in the determination process. In this case, the edge data on the specific side may not be used for obtaining the shape of the document. If the side edge data can be used, the shape of the document can be acquired with high accuracy. In addition, the shape of the document includes shapes other than the rectangle in addition to the standard size of the document.

  In addition, in the information acquisition process, the control unit uses the edge data of the specific side to acquire the position of the punch hole and makes an affirmative determination in the determination process when a negative determination is made in the determination process. In this case, the edge data of the specific side may not be used for acquiring the punch hole position. If the edge data of the side can be used, the position of the punch hole can be acquired with high accuracy.

  A computer program for realizing the function of the reading device is also new and useful.

  According to the present invention, a technique for improving the edge detection accuracy of a document is realized.

It is sectional drawing which shows schematic structure of the scanner concerning Embodiment. It is a block diagram which shows the electric constitution of a scanner. It is explanatory drawing which shows arrangement | positioning of a document. It is a flowchart which shows the procedure of a reading process. It is explanatory drawing which shows the detected edge. It is a flowchart which shows the procedure of a sampling position determination process. It is explanatory drawing which shows an inclination angle. It is explanatory drawing which shows the example of sampling frequency. 6 is a flowchart illustrating a procedure of document size acquisition processing. It is a flowchart which shows the procedure of a rotation skew feeding acquisition process.

  DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments embodying a reading apparatus according to the present invention will be described below in detail with reference to the accompanying drawings. In this embodiment, the present invention is applied to a scanner having an image reading function.

  As shown in FIG. 1, the scanner 100 of this embodiment has a reading head 21, and reads an image of a document while relatively moving the document and the reading head 21. The reading head 21 irradiates light at the opposite location and outputs a signal based on the amount of received light. For example, the scanner 100 continuously acquires image data for one line based on the output signal of the reading head 21. The reading method of the scanner 100 of this embodiment may be a CIS method or a CCD method.

  In order to read a document as a surface, the scanner 100 of this embodiment has a document transport unit 22 that transports the document. The document conveyance unit 22 includes a plurality of conveyance rollers 26 and the like, and conveys the document along a conveyance path 25 indicated by a one-dot chain line in FIG. Then, the document conveying unit 22 conveys the document one by one at a predetermined speed, and passes the portion facing the reading head 21. Then, the scanner 100 reads the document being conveyed by the document conveying unit 22 with the reading head 21.

  Further, the document transport unit 22 is provided with a document tray 28 for storing a document before reading and a document discharge tray 29 for storing a document after reading. When the scanner 100 receives an instruction to execute reading, the scanner 100 draws one of the documents set on the document tray 28 into the conveyance path 25 and conveys it, and discharges the document after reading to the document discharge tray 29.

  The scanner 100 according to the present embodiment continuously acquires image data in the main scanning direction by the reading head 21. The scanner 100 acquires image data in a reading range that is a range corresponding to the length of the reading head 21 in the main scanning direction regardless of the arrangement of the originals. As shown in FIG. 1, a white plate 23 is disposed at a position facing the reading head 21 across the document transport path 25. Therefore, the scanner 100 reads the white plate 23 when there is no document at the position facing the reading head 21.

  The scanner 100 according to this embodiment acquires image data in the sub-scanning direction as the document is conveyed. That is, the image data is acquired continuously in the main scanning direction, and increases in the sub-scanning direction as the document is conveyed and read. In this embodiment, the reading head 21 is an example of a reading unit, and the length of the reading head 21 is an example of the limit of the reading range.

  Next, the electrical configuration of the scanner 100 will be described. As shown in FIG. 2, the scanner 100 includes a controller 30 including a CPU 31, a ROM 32, a RAM 33, an NVRAM (nonvolatile RAM) 34, and an ASIC 35. Further, the scanner 100 includes a reading head 21, a document conveying unit 22, a network interface 37, a USB interface 38, and an operation panel 40, which are electrically connected to the controller 30.

  The ROM 32 stores various control programs for controlling the scanner 100, various settings, initial values, and the like. The RAM 33 is used as a work area from which various control programs are read, or as a storage area for temporarily storing data. The CPU 31 controls each component of the scanner 100 while storing the processing result in the RAM 33 or the NVRAM 34 according to the control program read from the ROM 32.

  The CPU 31 is an example of a control unit. The controller 30 may be a control unit, and the ASIC 35 may be a control unit. Note that the controller 30 in FIG. 2 is a collective term for the hardware used for controlling the scanner 100, such as the CPU 31, and does not necessarily represent a single piece of hardware that actually exists in the scanner 100.

  The network interface 37 is hardware for communicating with a device connected via a network using a LAN cable or the like. The USB interface 38 is hardware for communicating with a device connected via a USB cable or the like. The operation panel 40 has a display, an input button, etc., displays a notification to the user, and accepts an input operation by the user.

  Next, a document reading operation in the scanner 100 of this embodiment will be described. As described above, the scanner 100 according to the present embodiment reads a document while conveying it. Then, using the image data of the read portion, the edge of the image included in the range is detected, and the position information of the detected edge is acquired as edge data.

  An edge is a position in the image data where the data value of the image data changes beyond a predetermined threshold. For example, the scanner 100 binarizes the luminance value of each point in the image data with a predetermined threshold, determines that the position where the value of the binarized data has changed from 0 to 1 is an edge, and determines the position of the edge Is acquired as edge data.

  When a sheet-like document is read by the scanner 100 of this embodiment, a shadow is generated at the boundary between the edge of the sheet and the white plate 23. For this reason, as shown in FIG. 3, the image data 51 often includes a shadow image 52 having a darker color than the content portion of the document at the portion corresponding to the edge of the sheet. In FIG. 3, the entire acquired image data 51 is surrounded by a broken line and indicates that an image of the document is included therein. For example, the image data 51 obtained by reading a rectangular document includes a shadow image 52 based on the four sides of the document. When the above-described edge detection operation is executed, the edge of the shadow image 52 is detected.

  Here, each side of the document is defined as shown in FIG. The portion of the document that first reaches the position of the reading head 21 is referred to as “upper side”. If the document is arranged parallel to the reading head 21, the shadow image 52 formed by the upper side appears in the read image data 51 aligned in the main scanning direction. Further, a portion of the original that finally reaches the position of the reading head 21 is defined as a “lower side”. Further, when the upper side is arranged on the upper side and the lower side is arranged on the lower side of the original reading surface, a position arranged on the right side is designated as a “right side” and a place arranged on the left side is designated as a “left side”. When the left side and the right side are not distinguished, they are collectively referred to as “side”.

  The position of each point in the image data 51 obtained by reading the document with the scanner 100 of this embodiment will be described using coordinates in the XY plane as shown in FIG. With the generation start position of the image data 51 as the origin, the longitudinal direction (main scanning direction) of the reading head 21 is the X axis, and the reading progress direction (sub scanning direction) is the Y axis. The width Wx of the image data 51 in the X-axis direction is a fixed length based on the length of the reading head 21. Wx is the maximum value of the X coordinate of the image data 51. The image data 51 increases in the Y-axis direction by one pixel as reading progresses.

  When the scanner 100 of this embodiment receives an instruction to start reading, the scanner 100 starts conveying a document, and then starts a reading operation at a predetermined timing. As shown in FIG. 3, the timing for starting reading is slightly before the timing at which the upper side of the document is predicted to reach a position facing the reading head 21. Note that the timing at which the leading edge of the document is predicted to reach the position facing the reading head 21 can be determined based on the elapsed time from the start of document conveyance and the document conveyance speed.

  As the reading progresses, the scanner 100 performs an edge detection operation on the image data 51 being generated. First, after acquiring a predetermined number of lines of image data 51 from the start of reading, the scanner 100 detects the edge of the shadow image 52 caused by the upper side of the document. For example, an edge is detected from the image data 51 in a predetermined range from the point Y = 0 in the sub-scanning direction. Then, the position of the upper side of the document is estimated based on the edge data that is the acquired edge position. Note that if the edge detection operation is executed for all points in the X-axis direction, the process takes too much time. Therefore, in the scanner 100 of this embodiment, an edge detection operation is executed for X positions at a predetermined interval. Further, in order to suppress erroneous determination of noise as the upper side, the upper side is determined only when the luminance value continuously exceeds the threshold at a predetermined number of points or more.

  As the reading further proceeds, the scanner 100 detects the edge of the shadow image 52 caused by the side of the document. In this case, as the reading progresses, detection is sequentially performed from the upper side of the document in the sub-scanning direction. The scanner 100 samples at a predetermined frequency in the Y-axis direction and acquires edge data in the X-axis direction. That is, an edge is detected for each position at a predetermined interval in the Y-axis direction in the image data 51, and the position of the side is estimated. For example, the position of the left side of the document is estimated by detecting in the X-axis direction from the point X = 0. Further, the position of the right side of the document is estimated by detecting in the direction opposite to the X-axis direction from the point of X = Wx. The sampling frequency may be the same on the left side and the right side or may be different.

  In the scanner 100, the vicinity of the end portion in the length direction of the reading head 21 is easily affected by noise, and an edge may be detected near the end portion in the main scanning direction at a location other than the side of the document. . Further, when the document is set at a position that protrudes in the length direction of the reading head 21, there is a possibility that the edge due to the side of the document cannot be detected properly. Therefore, in these cases, if the position of the side of the document is estimated by the above-described edge detection operation, there is a risk of erroneous determination.

  The scanner 100 according to the present embodiment determines that the detected edge position is not an appropriate edge position when the detected edge position is within a predetermined distance from both ends of the image data 51 in the X-axis direction. . Specifically, if the X coordinate of the detected edge position is close to 0 or close to Wx, it is determined that the edge position is not appropriate. In this case, the possibility that the end of the document in the main scanning direction exceeds the reading range of the reading head 21 cannot be denied, so it is determined that the end of the document in the main scanning direction exceeds the reading range of the reading head 21. To do. Then, the side position of the document is not estimated based on the position of the edge.

  Further, when the scanner 100 determines that the reading range is exceeded, the position information on both ends of the detected edge is not used in the process of acquiring document information to be described later. The specified distance may be the same or different between the left end side and the right end side in the X-axis direction of the image data 51.

  In addition, the scanner 100 similarly determines the leading edge corners, which are the positions of the left and right edges of the edge of the upper side of the document. When the position of the edge due to the tip corner cannot be obtained, or when the tip corner is within a predetermined distance from both ends in the X-axis direction of the image data 51, the position of the tip corner Do not use. For example, the position of the side of the document is not estimated based on the position of the leading edge. That is, when the position of the edge estimated to be the leading edge portion of the document is near the end portion in the main scanning direction, it is determined that the leading edge portion of the document exceeds the reading range.

  That is, the scanner 100 according to the present embodiment is the position information of the edge when it is determined that the edge is detected in a range that is not near the both ends in the X-axis direction in the reading range. The edge data is estimated to be the position of the side of the document, and is used when acquiring various information related to the position and shape of the document. In the following, it is assumed that the edge data is valid when it is determined that an edge is detected within the reading range. On the other hand, if it is determined that no edge is detected within the reading range, the edge data is invalid and the edge data is not used.

  Next, reading processing for performing the above-described document reading operation in the scanner 100 according to the present embodiment will be described with reference to a flowchart of FIG. The reading process is executed by the CPU 31 when the scanner 100 receives an instruction to start reading. Hereinafter, the position of each point in the image data 51 is expressed as “(X coordinate, Y coordinate)” using the coordinates in the XY plane described above.

  When the scanner 100 of the present embodiment starts the reading process, it first starts conveying the document (S101). When a predetermined time elapses from the start of document conveyance, the scanner 100 starts reading with the reading head 21 (S102). In step S102, the scanner 100 acquires image data for each line in the X-axis direction. Thereafter, the image data of each line is sequentially acquired in the Y-axis direction and acquired. The process of S102 is an example of a reading process.

  Further, based on the acquired image data 51, the scanner 100 starts detecting an edge that is estimated to be a shadow image 52 by the upper side of the document (S104). Specifically, at each position sampled in the X-axis direction, based on the image data 51 having a predetermined width in the Y-axis direction, a location where the luminance value exceeds a threshold value is detected as an edge, which is edge position information. Get edge data.

  Then, the scanner 100 determines whether an edge that can be determined to be a part of the upper side has been detected (S105). If it is determined that the upper side has not been detected (S105: NO), the scanner 100 proceeds further. If it is determined that the upper side has been detected (S105: YES), the scanner 100 further reads a predetermined distance from the detected position of the upper side (S106).

  Then, the scanner 100 acquires position information of the entire upper side of the document based on the edge positions detected from the image data 51 read from the reading start to the end of S106 (S108). For example, a continuous edge in the X-axis direction is detected from the position estimated as the upper side in S104 and set as the position of the continuous upper side. The process of S108 is an example of an edge acquisition process.

  Next, the scanner 100 acquires the position of the upper corner (S109). The position at which the edge detected in S108 is interrupted is estimated as the corner of the document, and an edge continuous in the Y-axis direction is detected from that position. If the edges are continuous in the Y-axis direction, it is determined that the position where the edge is interrupted by the upper side corresponds to the leading edge of the document. That is, in S109, the scanner 100 acquires the coordinates of the corners of the upper side of the document as shown in FIG. 5 from the position of the end of the edge by the upper side of the document. For example, a corner C (Xc, Yc) formed by the upper side and the left side and a corner D (Xd, Yd) formed by the upper side and the right side are acquired. The process of S109 is an example of an extraction process.

  Then, the scanner 100 determines whether or not the right and left corners C and D have been successfully acquired (S112). For example, if the end cannot be detected on the upper side, or if both ends of the upper side cannot be detected properly due to the upper side being wavy, the scanner 100 appropriately acquires the coordinates of the corners C and D. Can not. Accordingly, the scanner 100 determines that acquisition of the corners C and D has failed.

  Further, as described above, when the position of the detected edge is in the vicinity of both ends of the image data 51 in the X-axis direction, the scanner 100 has low reliability. That is, it is not preferable to determine the position of the side based on the edge data. Therefore, assuming that the edge data is invalid, the scanner 100 does not use the edge data for estimating the position of the side. In the following, as shown in FIG. 5, the range in which the edge data is invalid is defined as a specified distance Wl on the left side and a specified distance Wr on the right side. Wl and Wr may be the same distance or different.

  That is, in S112, the scanner 100 determines that if the acquired X-coordinate Xc of the corner C is within the range from 0 to a predetermined distance Wl, that is, if Xc <Wl. It is determined that acquisition of part C has failed. Further, the scanner 100 determines that the acquired X coordinate Xd of the corner D is within a range from the maximum value Wx of the X coordinate to a predetermined specified distance Wr, that is, when Xd> Wx−Wr. Determines that acquisition of the corner D has failed. The process of S112 is an example of a determination process.

  When the scanner 100 determines that the left and right corners C and D have been successfully acquired (S112: YES), the scanner 100 determines to use the upper side, the left side, and the right side (S113). That is, both the edge data on the left side and the edge data on the right side are valid and can be used for subsequent processing.

  On the other hand, when the scanner 100 determines that acquisition of at least one of the corners C and D is not successful (S112: NO), the scanner 100 determines whether acquisition of the corner C on the left side is successful (step S112). S115). If the scanner 100 determines that the left side corner C has been successfully acquired (S115: YES), the left side edge data is valid, and the upper side and the left side are used in the subsequent processing. Determine (S116).

  If the scanner 100 determines that the left side corner C has not been successfully acquired (S115: NO), the scanner 100 determines whether the right side corner D has been successfully acquired (S118). . If the scanner 100 determines that acquisition of the corner D on the right side is successful (S118: YES), the edge data on the right side is valid, and the upper side and the right side are used in the subsequent processing. Determine (S119).

  If the scanner 100 determines that acquisition of the corner D on the right side has not been successful (S118: NO), the scanner 100 determines to use only the upper side (S121). That is, the left edge data and the right edge data are invalid, and neither the left edge nor the right edge can be used. In this case, the scanner 100 does not execute processing for acquiring document information, which will be described later.

  Therefore, after S121, the scanner 100 determines whether reading has been completed up to the lower side of the document (S123). For example, when the end of the original passes through the position facing the reading head 21 and is conveyed for a predetermined time, it can be determined that the reading to the lower side of the original has been completed. If it is determined that the reading is not completed (S123: NO), the reading is further continued. If it is determined that the reading is finished (S123: YES), the reading process is finished. If there is a process that can be performed using only the information on the upper side after S121, the process may be executed.

  On the other hand, if it is determined in any one of S113, S116, and S119 that at least one of the left side and the right side is to be used, the scanner 100 first executes a sampling position determination process (S125). The flowchart of FIG. 6 shows the procedure of the sampling position determination process.

  In the case where the inclination of the document is large, for example, as shown in FIG. 7, there is a possibility that the edge data by one of the left and right sides becomes invalid from the middle. In the sampling position determination process, the coordinates of the excess position H, which is the position where the side edge data is predicted to change from valid to invalid, are acquired, and the position of the reading position in the sub-scanning direction is changed to the Y coordinate of the excess position H. When approaching, the sampling frequency is increased more than before. In this process, it is estimated that the document is rectangular.

  In the sampling position determination process, the scanner 100 first acquires the document inclination angle θ with respect to the X-axis direction of the upper side (S201). As shown in FIG. 7, the inclination angle θ may be an angle formed by a line segment connecting the corner C and the corner D and the X axis, or a straight line approximating each edge data of the upper side and the X axis. It may be an angle formed by The process of S201 is an example of an inclination acquisition process.

  Next, if the inclination angle θ> 0, the coordinates of the excess position H are acquired. For example, as shown in FIG. 7, if Yc <Yd, the edge data on the left side may become invalid. Therefore, the scanner 100 assumes a straight line perpendicular to the upper side from the corner C and assumes the left side of the document. Further, the scanner 100 acquires the coordinates (Wl, Yh) of the excess position H where the assumed left side and the straight line X = Wl are predicted to intersect (S203).

Specifically, the Y coordinate Yh of the excess position H can be expressed as (Equation 1) below using the inclination angle θ obtained in S201.
tan θ = (Xc−Wl) / (Yh−Yc)
Yh = Yc + (Xc−Wl) / tan θ (Formula 1)

  Next, the scanner 100 determines the sampling frequency for detecting the left side edge (S205). As described above, the scanner 100 performs edge detection at a predetermined sampling frequency in the Y-axis direction. Then, as shown in FIG. 8, the scanner 100 increases the sampling frequency in the region Q within the predetermined range from the Y coordinate Yh of the excess position H as compared to the region other than the region Q. The region Q is an example of a boundary region.

Specifically, the scanner 100 determines sampling frequencies Pa and Pb as shown in FIG. 8 according to the Y coordinate of the detection position. The predetermined range is, for example, 100 pixels.
Sampling frequency Pa = 50 (pixels) where Y <Yh-100
Sampling frequency Pb = 5 (pixels) where Y ≧ Yh−100

  Then, the scanner 100 determines a position for executing edge detection according to the sampling frequency determined in S205, and stores it in the RAM 33 (S207). Note that since the position that actually exceeds may be below the excess position H, the edge detection position may be determined to be slightly below the excess position H. After S207, the sampling position determination process ends.

  When the scanner 100 generates the image data 51 up to the determined edge detection position, the scanner 100 performs an edge detection operation. When the X coordinate of the detected edge position is equal to or less than W1, the scanner 100 does not perform edge detection on the left side thereafter. That is, when the edge data becomes invalid, the execution of the edge detection operation for the left side is stopped, and only reading of the original is continued. The scanner 100 does not use the left side after this point. Even if the left side is invalid, the edge detection on the right side continues if the edge data on the right side can be acquired appropriately.

  As determined by the sampling position determination process, since the sampling frequency is increased at a position close to the excess position H, it is unlikely that the detected edge data has been invalidated. Therefore, there is a low possibility of erroneously determining invalid edge data as valid.

  As a result, it is possible to accurately determine whether or not the side edge data is valid. That is, it is unlikely that edge data with low reliability is used as the position of the side. If there is a possibility that the edge data on the right side is invalid in the middle, the scanner 100 assumes a right side passing through the corner D in S203, and the assumed right side and a straight line of X = Wr are obtained. The point predicted to intersect is defined as the excess position H.

  If the length of the document in the conveyance direction is smaller than (Yh−Yc), the reading ends before the edge data becomes invalid. If the end of the document is reached before the Y coordinate of the reading position reaches (Yh-100), the scanner 100 does not change the sampling frequency as a result.

  Returning to the reading process of FIG. 4, since the sampling position determination process of S125 is completed, an original information acquisition process is executed (S127). The document information acquisition process is a process of acquiring information related to a document using edge data determined to be valid. The information of the original to be acquired is at least one of the characteristics and state of the original. For example, the size of the original, the inclination of the original, the rotation of the original, the shape of the original, and the position of the punch hole formed in the original is there. Details of the manuscript information acquisition process will be described later. The process of S127 is an example of an information acquisition process.

  The scanner 100 according to the present embodiment includes one or more processes as necessary among various document information acquisition processes when the edge data regarding not only the position of the upper side but also at least one of the left side and the right side is valid. Execute. If both the left and right sides are invalid, no processing is performed. The scanner 100 only needs to have one or more of various processes for acquiring document information. When there are a plurality of processes, all of them may be executed, or only the processes selected by the user's settings may be executed.

  When the document information acquisition process is completed, it is determined whether or not the document reading is completed (S123). If it is determined that the reading has not been completed (S123: NO), the reading is further continued. If it is determined that the reading is finished (S123: YES), the reading process is finished.

  Subsequently, the document information acquisition process executed in S127 will be described. First, an overview of processing for acquiring various types of information will be described. In the process of acquiring the size of the document, the scanner 100 according to the present embodiment acquires the minimum rectangular size including the entire document even if the document has a shape other than the rectangle. For example, when there is a part wider than the upper side in the middle of the document, the scanner 100 acquires the size of the document in the main scanning direction based on the width of the part. The scanner 100 executes a process of acquiring the size of the document when both the left and right edge data are valid.

  Further, the scanner 100 according to the present embodiment acquires an inclination angle formed between the edge of the original and the axial direction in the process of acquiring the inclination of the original. By using at least one position of the left side and the right side instead of detecting only by the upper side, the tilt angle can be acquired more precisely. For example, the scanner 100 corrects the image data 51 based on the acquired tilt angle. The scanner 100 executes processing for obtaining the inclination of the document when at least one edge data of the left side and the right side is valid.

  Further, the scanner 100 according to the present embodiment acquires whether or not the inclination angle formed between the left side or the right side of the document and the axial direction changes in the process of acquiring the rotation skew of the document. . If the inclination angle of the document increases as it is conveyed, it may cause jamming of the document. For example, when the scanner 100 detects rotation skew, it displays an error message and stops conveying the document. If at least one edge data of the left side and the right side is valid, the scanner 100 executes a process of acquiring the rotation skew of the document. However, if both edge data at the left and right sides are valid, a more reliable determination is possible.

  Further, in the process of acquiring the shape of the document, the scanner 100 according to the present embodiment acquires the shape of the document by connecting the positions of the left side and the right side detected as edge data. For example, the scanner 100 corrects the read image data 51 according to the shape of the document. When both edge data are valid, the scanner 100 executes processing for acquiring the shape of the document. If only either the left side or the right side is valid, the shape of the document that can be acquired is incomplete information.

  In the scanner 100 according to this embodiment, in the process of acquiring the position of the punch hole formed in the document, a circular image having a predetermined size is included at a position within a predetermined distance range from each side. Judged as a punch hole. For example, when the scanner 100 determines that there is a punch hole, the scanner 100 executes processing for converting the punch hole image into a surrounding ground color. When the edge data at at least one position of the left side and the right side is valid, the scanner 100 executes a process of acquiring the punch hole position with respect to the punch hole close to the valid side.

  Note that the sampling frequency of the edge detection operation for acquiring the edge data of the side is different depending on the type of document information acquired using the side. The sampling frequency Pa outside the vicinity of the excess position H described above is set to a frequency set in advance for each type of processing executed as the document information acquisition processing. For example, in the case of acquiring information whose side position is required with high accuracy, the sampling frequency is set higher than in other cases. As a result, it is expected that the accuracy of the information of the obtained document will be improved. Note that the sampling frequency Pb in the vicinity of the excess position H may be determined based on the sampling frequency Pa, or may be set to a predetermined value regardless of the type of information.

  In the following, as an example of the document information acquisition process, the document size acquisition process shown in FIG. 9 and the rotation skew acquisition process shown in FIG. 10 will be described with reference to the respective flowcharts. These processes are executed after executing any one of S113, S116, and S119 in the reading process. That is, at the start of this processing, the sides (left side or right side or both) that can be used other than the upper side are determined.

  First, the document size acquisition process will be described with reference to the flowchart of FIG. This process is executed when it is determined in S113 that the positions of both the left and right sides can be used. In the document size acquisition process, the scanner 100 first determines the initial values of the left and right positions of the document based on the upper side information, and determines the initial positions of the left side position and the right side position (S301). Specifically, the scanner 100 sets the left side position as the X coordinate Xc of the corner C and the right side position as the X coordinate Xd of the corner D.

  Next, the scanner 100 determines whether or not the edge data acquired by the document reading operation and the edge detection operation is outside the left side position or the right side position (S303). Specifically, if the X coordinate of the left edge data is smaller than the left side position or the X coordinate of the right edge data is larger than the right side position, the scanner 100 satisfies the edge data. Judged to be outside.

  When the scanner 100 determines that the edge data is outside (S303: YES), the X coordinate of the edge data determined to be outside is set as the corresponding left side position or right side position (S305). If the edge data determined to be outside is only on one side, the scanner 100 changes only the side position on one side. On the other hand, when the scanner 100 determines that the edge data is not outside (S303: NO), it skips S305.

  Then, the scanner 100 determines whether or not reading for one page is completed (S307). If the reading has not been completed (S307: NO), the scanner 100 determines whether the next acquired edge data is valid (S309). As described above, the side of the document may deviate from the reading range as the document is conveyed. After the deviation, the detected edge data becomes invalid, and the scanner 100 is not used even if the edge data is acquired.

  When the scanner 100 determines that the edge data is valid (S309: YES), the scanner 100 returns to S303 and compares the edge data with the side position. If the scanner 100 determines that the edge data is valid and the reading has been completed (S307: YES), the scanner 100 calculates the size of the document using the distance between the final left side position and the right side position as the width of the document (S311). The document size acquisition process is terminated.

  On the other hand, if the scanner 100 determines that the edge data has become invalid before the reading is completed (S309: NO), it means that acquisition of the document size has failed. Therefore, the scanner 100 determines that the document size is unknown (S313), and ends the document size acquisition process. If the reading of the original is not completed, the reading operation continues.

  Next, the rotation skew acquisition process will be described with reference to the flowchart of FIG. This process is executed when it is determined that at least one position of the left side and the right side can be used. In the rotation skew acquisition processing, first, the scanner 100 acquires the predicted positions of the left side and the right side based on the information on the upper side (S401). For example, using the tilt angle θ and the sampling position described above, the scanner 100 predicts edge data at a position where edge detection is performed next.

  After performing the edge detection, the scanner 100 compares the acquired edge data with the predicted position predicted in S401. Then, the scanner 100 determines whether or not the difference between the predicted position and the edge data is within a predetermined approximate range (S403). When it is determined that the predicted position is close to the edge data and the difference is within the approximate range (S403: YES), the scanner 100 determines whether reading of one page has been completed (S405).

  When the scanner 100 determines that the reading has been completed (S405: YES), the rotation skew feeding does not occur and the reading is finished. On the other hand, if the reading is not completed (S405: NO), the scanner 100 continues the reading and returns to S401 to acquire the next predicted position.

  On the other hand, if it is determined that the difference between the predicted position and the edge data exceeds the approximate range (S403: NO), there is a possibility of rotational skew. Therefore, for example, if the edge data on both sides is valid, the scanner 100 determines whether or not the difference between the two sides is comparable (S407).

  If the difference between the two sides is approximately the same (S407: YES), the scanner 100 determines that the rotation is skewed (S409). Then, for example, the scanner 100 displays an error message and stops the conveyance and reading of the document. On the other hand, if the difference between the two sides is not comparable (S407: NO), the scanner 100 determines that the document is an irregular document (S411). If it is determined to be indefinite, reading may be continued until reading is completed. After S409 and S411, the rotation skew acquisition process is terminated.

  If edge data on only one side is valid, if NO in S403, it may be determined that the rotation is skewed. Alternatively, the next edge data may be compared with the predicted position, and if the difference increases, it may be determined that the rotation is skewed. Depending on the determination result, it is determined whether or not to continue reading, and the rotational skew acquisition processing is terminated.

  As described above in detail, the scanner 100 of this embodiment reads an image of a document while conveying the document. Then, an edge detection operation is executed, and edge data that is edge position information is acquired from the read image data 51. Further, based on the acquired edge data, it is determined whether the left side or the right side of the document exceeds the reading range of the reading head 21. In other words, since all the edges detected by the edge detection operation are not used as the end of the document as it is and it is determined whether or not the edge data is highly reliable, an improvement in the detection accuracy of the document edge can be expected. .

  Note that this embodiment is merely an example, and does not limit the present invention. Therefore, the present invention can naturally be improved and modified in various ways without departing from the gist thereof. For example, the present invention is not limited to a scanner, and can be applied to any apparatus having an image reading function, such as a copying machine, a multifunction machine, and a FAX apparatus. Further, for example, the present invention can be applied to a configuration in which the reading head 21 is moved and read instead of conveying the document.

  The present invention can also be applied to an information processing apparatus such as a PC. For example, image data can be acquired from the scanner 100 by a PC, and an edge detection operation can be performed based on the acquired image data. Then, the edge data may be acquired by the PC, and it may be determined whether or not the edge data is valid. Furthermore, the determination result may be transmitted to the scanner 100 or may be used by a PC.

  Further, for example, at least one of the specified distance Wl on the left side and the specified distance Wr on the right side may be omitted. That is, Wl = 0 or Wr = 0 may be used.

  Further, the excess position H is calculated based on the coordinates of the corners C and D of the tip. However, when the distance between the detected edge position and the position of the specified distance from the limit of the reading range is less than the threshold value. , It may be determined that the excess position H has been approached. Further, although the sampling frequency is increased when the excess position H is approached, it does not have to be changed. Further, although sampling is performed at a different frequency according to the information of the original to be acquired, all may be set at a uniform frequency.

  For example, a timeout may be provided in S105. In other words, if the upper edge data cannot be acquired, the reading has failed, and an error message may be displayed to end the process.

  The processing disclosed in the embodiments may be executed by a single CPU, a plurality of CPUs, hardware such as an ASIC, or a combination thereof. Further, the processing disclosed in the embodiment can be realized in various modes such as a recording medium or a method recording a program for executing the processing.

21 Read head 31 CPU
100 scanner

Claims (11)

A reading unit for reading an image;
A control unit;
With
The controller is
A reading process for causing the reading unit to read an image;
Edge acquisition processing for acquiring edge data, which is edge information of the document, from the image data read by the reading processing;
A determination process for determining whether or not an end of the document in the main scanning direction exceeds a reading range of the reading unit based on the edge data acquired in the edge acquisition process;
The reading apparatus characterized by performing. The reading device according to claim 1,
The controller is
Based on the edge data acquired in the edge acquisition process, an extraction process for extracting the leading edge corner of the document is executed,
In the determination process, when the leading end corner of the document is not extracted or when the leading end corner of the document is within the range from the limit of the reading range of the reading unit to a specified distance, A reading apparatus characterized by determining that the range is exceeded. The reading device according to claim 1 or 2,
The controller is
Based on the edge data acquired in the edge acquisition process, an inclination acquisition process for acquiring the inclination of the document is executed,
In the determination process, based on the inclination acquired in the inclination acquisition process, it is determined whether or not the end portion of the document in the main scanning direction exceeds the reading range of the reading unit. . The reading apparatus according to any one of claims 1 to 3,
The controller is
In the edge acquisition process,
Acquiring the edge data in the main scanning direction at a predetermined frequency in the sub-scanning direction;
When it is determined in the determination process that the reading range of the reading unit is exceeded, the frequency of acquiring the edge data in a boundary region that is a region within a predetermined range from an excess position that is a position exceeding the reading range, A reading apparatus characterized in that the reading frequency is higher than the frequency in an area other than the boundary area. The reading device according to claim 4,
The controller is
In the edge acquisition process, the edge data is acquired at a different frequency according to at least one of a characteristic and a state of a document acquired using the edge data. In the reading device according to any one of claims 1 to 5,
The controller is
Of the edge data acquired in the edge acquisition process, using edge data of a specific side that is a side on the edge side determined not to exceed the reading range in the determination process, the characteristics of the document An information acquisition process for acquiring at least one of a state and a state is performed. The reading device according to claim 6,
The controller is
In the information acquisition process, when a negative determination is made in the determination process, the inclination of the document is acquired using the edge data on the leading edge side of the document and the edge data on the specific side, and the determination process A reading apparatus, comprising: processing for obtaining an inclination of a document by using edge data on a leading end side of the document when an affirmative determination is made. In the reading device according to claim 6 or 7,
The controller is
In the information acquisition process, when a negative determination is made in the determination process, the edge data of the specific side is used to acquire the rotation skew amount, and when the determination process is positive, the specific side A reading apparatus, wherein edge data of a side is not used for acquiring a rotational skew amount. In the reading device according to any one of claims 6 to 8,
The controller is
In the information acquisition process, when a negative determination is made in the determination process, the edge data of the specific side is used to acquire the shape of the document, and when the determination is positive, the specific side The reading apparatus is characterized in that the edge data is not used for obtaining the shape of the document. In the reading device according to any one of claims 6 to 9,
The controller is
In the information acquisition process, when a negative determination is made in the determination process, the edge data of the specific side is used to acquire the position of the punch hole, and when the determination process is positive, the specific side A reading apparatus, wherein edge data of a side is not used for acquiring a punch hole position. In the information processing device,
A document acquisition process for acquiring image data of the document;
Edge acquisition processing for acquiring edge data, which is edge information of the document, from the image data acquired in the document acquisition processing;
A determination process for determining whether or not an end of the document in the main scanning direction exceeds the reading range based on the edge data acquired in the edge acquisition process;
A program characterized by having executed.

Images