JP2019022010A - Document reading device, control method of document reading device, and program - Google Patents

Abstract

To exclude the shadow effect exactly even when the thickness of a document is not uniquely determined at the time of scanning a document and cut out an image such that missing information contained in the document does not occur in the resultant.SOLUTION: A scanning apparatus 100 captures image data by capturing an image capturing area 112 on which a document is placed with an imaging sensor 108 and acquires image data (S702), recognizes a rectangle formed by a plurality of line segments matching the document size that is a candidate for the document end portion included in the acquired image data within a predetermined error (S703), selects a line segment in which an angle formed by a normal vector of each line segment and the shadow generation direction information is 90 degrees or more from among the line segments constituting each of the rectangles as a line segment hard to be affected by the shadow on the basis of shadow generation direction information stored in an HDD 106 (S801 in FIG. 8), and cuts out a region of a designated size as a reference with the selected line segment (S802 in FIG. 8) (S704 to S705).SELECTED DRAWING: Figure 7

Description

  The present invention relates to an original reading apparatus, an original reading apparatus control method, and a program.

  There is a scanning device called an overhead scanner. The overhead scanner has a scanning function of photographing a document placed on a plane such as a desk with an imaging sensor and providing a user with an image of the document as a product. In such a scanning apparatus, when a document is photographed by the imaging sensor, an area other than the document is included in the imaging. In order for a user to obtain an image (product) including only a document, a scanning device or an information processing device connected to the scanning device is often provided with a document region cutout function. The document area cut-out function performs document area recognition processing for discriminating between a document area included in imaging and an area that is not, that is, an edge of the document, and an image including only the document based on the recognition result of the document edge Cut out.

  However, due to the influence of various factors such as shadows around the document and document deflection, an error occurs in the document edge recognition result by the document region recognition process. Thereby, for example, even if the user scans a plurality of documents of the same size, in many cases, the size of each product is different.

  In addition, the scanning device has a standard size cutout function so that the sizes of the deliverables are neatly arranged when the user scans a document of a specific size. In the standard size cutout function, the user inputs the size of the document in advance, and the image of the document area is cut out according to the recognition result of the document edge by the document area recognition process and the input document size. In the standard size cutout function, if there is a discrepancy in the recognition result of the document edge by the document area recognition process due to the shadow around the document, the result will be the standard size when cut out at the standard size. Misalignment is included in the cut out image. When the cut image includes a shift, an event such as missing information included in the document may occur.

  Patent Document 1 proposes a technique for eliminating a shift due to the influence of shadows in a standard size cutout in a scanning device that copies a CD (Compact Disc) label. In the technique of Patent Document 1, the center of the CD is calculated from the area ratio of the CD as a document included in the scanned image and the shadow image generated around the original, and the image of the CD area is cut out from the center in a circular shape. As a result, a product that excludes the influence of the shadow generated around the CD can be obtained. In the scanning apparatus of Patent Document 1, it is necessary to set information (profile) of the area ratio between the CD and the shadow image generated around it in advance for each place where the original CD is placed.

JP 2007-104504 A

  The scan target of Patent Document 1 is a CD, and its thickness is uniquely determined. In Patent Document 1, if an object having the same circular shape as a CD having a thickness different from that of a CD is used as a scan target, the shadow area is not accurately excluded because it does not match the registered area ratio of the profile. I can't do it.

  In a scanning apparatus such as an overhead scanner, the thickness of the original varies depending on the original, that is, the thickness of the original to be scanned is not uniquely determined. Therefore, the technique of Patent Document 1 cannot be applied.

  The present invention has been made to solve the above problems. An object of the present invention is to accurately exclude shadow effects during scanning even when the thickness of the document is not uniquely determined when scanning a document so that missing information contained in the document does not occur in the product. It is to provide a mechanism that can cut out an image of a predetermined size.

  The present invention relates to a document reading apparatus that reads image data from a document by a photographing unit, a storage unit that stores direction information indicating a direction in which a shadow is generated when a document is photographed, and a photographing region where the document is placed. A first acquisition unit that captures image data by capturing the image, and a figure having a predetermined shape formed by a plurality of lines that are candidates for the document edge included in the image data acquired by the first acquisition unit Recognizing means; and selecting means for selecting a reference line from lines constituting each figure recognized as a document edge candidate by the recognizing means based on the direction information stored in the storage means; Cutting-out means for cutting out a region of a predetermined size from the image data on the basis of the line selected by the selection means.

  According to the present invention, when scanning a document, even if the thickness of the document is not uniquely determined, the influence of shadow is accurately excluded at the time of scanning so that the missing information contained in the document does not occur in the product. An image of a predetermined size can be cut out.

External view of the scanning apparatus shown in this embodiment Hardware configuration diagram of the scanning apparatus of this embodiment The figure explaining the outline | summary of the shadow influence grasp processing of a present Example The figure which shows an example of the information handled by the shadow influence grasping process of Example 1. The flowchart which shows an example of the shadow influence grasping | ascertainment process of Example 1. FIG. 10 is a diagram illustrating an example of information handled in the document scanning process according to the first embodiment. 6 is a flowchart illustrating an example of document scanning processing according to the first exemplary embodiment. Flowchart illustrating an example of the cutout process according to the first embodiment. The figure which shows an example of the information handled by the shadow influence grasping process of Example 2. FIG. 10 is a diagram illustrating an example of information handled in the document scanning process according to the second embodiment. Flowchart illustrating an example of the cutout process according to the second embodiment. The figure which shows an example of the information handled by the shadow influence grasping | ascertainment process of Example 3. Flowchart illustrating an example of the shadow effect grasping process according to the third embodiment. FIG. 10 is a diagram illustrating an example of information handled in document scan processing according to the third embodiment. Flowchart illustrating an example of the cutout process according to the third embodiment. Flowchart illustrating an example of a shadow effect grasping process according to the fourth embodiment.

  Hereinafter, embodiments for carrying out the present invention will be described with reference to the drawings.

FIG. 1 is a diagram showing an example of the appearance of a scanning apparatus showing an embodiment of a document reading apparatus according to the present invention.
As shown in FIG. 1, the scanning apparatus 100 according to the present embodiment includes an imaging sensor 108 fixed on a desk 111 with a support 110 and an operation panel 107. An imaging area 112 indicates an area where the imaging sensor 108 can perform imaging. A document 113 indicates a document such as paper or a card. If the document is smaller than the shooting area 112, the user can place it on an arbitrary place on the desk 111. The scanning apparatus 100 is assumed to be placed in a room, and the direction of light entering the imaging region 112 from the room illumination 114 changes depending on the place where the scanning apparatus 100 is placed.

FIG. 2 is a block diagram illustrating an example of a hardware configuration of the scanning apparatus 100.
As shown in FIG. 2, the scanning apparatus 100 includes a controller 120, an operation panel 107, and an image sensor 108.

  In the controller 120, the internal bus 109 transmits and receives electrical signals so that information can be transmitted between the memory controller 103 and the I / O controller 105. The memory controller 103 performs overall control of memory access to the main memory 104. The I / O controller 105 transmits and receives information to and from the processing unit connected via the HDD 106, the operation panel 107, the image sensor 108, and the internal bus 109.

  The graphic controller 102 controls projection by the operation panel 107. A hard disk drive (HDD) 106 stores a boot loader program and a scanning device control program. Other storage devices such as a solid state drive (SSD) may be provided instead of or in combination with the HDD. The main memory 104 is a volatile memory. Since the main memory 104 can be accessed at high speed, information stored in the HDD 106 and information used temporarily are stored here.

  When the scanning device 100 is turned on, the CPU 101 reads and executes the boot loader program stored in the HDD 106, reads the scanning device control program stored in the HDD 106, and stores it in the main memory 104. Then, the CPU 101 executes a scanning device control program stored in the main memory 104, and realizes each function of the scanning device 100. In addition, the CPU 101 executes the scanning device control program to analyze the image captured by the imaging sensor 108, or to project and display information to the user on the operation panel 107. Is executed. The above is the description of the configuration of the controller 120.

  The imaging sensor 108 converts the captured image into a digital signal of RGB luminance information and generates image data (hereinafter referred to as “captured image”). The image sensor 108 transmits and receives information to and from the I / O controller 105. The operation panel 107 has a structure in which a display panel made of liquid crystal or the like and a touch panel made of a capacitive sensor or the like are stacked on the display panel. The user can input information by touching the touch panel while browsing the information displayed on the display panel. The operation panel 107 transmits / receives information to / from the I / O controller 105 and the graphic controller 102.

  Hereinafter, a scan process that is executed in the scanning apparatus 100 to grasp the direction of the shadow of the three-dimensional object placed in the imaging region caused by the illumination light of the room, and a scan process that scans the document by specifying the document size in advance The specific flow of will be described. In this embodiment, a license will be used as an example of a document.

FIG. 3 is a diagram for explaining the outline of the process (shadow influence grasping process) for grasping the shadow direction of the three-dimensional object according to the present embodiment.
The shadow effect grasping process is executed at least once when there is a change in at least the place where the scanning apparatus 100 is placed or the lighting configuration of the room. In a room where natural light enters from a window or the like, the state of light changes, so that the shadow effect grasping process may be executed at regular time intervals or before the scanning process. In this case, a message may be displayed on the operation panel 107 or the like to prompt the user to execute the shadow effect grasping process.

  As shown in FIG. 3, when the user places a rod-shaped three-dimensional object 301 having a specific color such as blue on the photographing region 112, a shadow 302 of the three-dimensional object 301 is generated by the light of the room illumination 114. Note that the color of the three-dimensional object 301 is preferably a color that can be easily distinguished from the imaging region 112 and the shadow. Using this shadow 302, shadow influence grasping processing is performed. This will be described in detail below with reference to FIGS.

FIG. 4 is a diagram illustrating an example of information handled in the shadow influence grasping process according to the first embodiment.
FIG. 5 is a flowchart illustrating an example of the shadow effect grasping process according to the first embodiment. S501 to S504 in the figure are processing steps executed by the scanning apparatus 100. The processing of the flowcharts shown in FIG. 5 and FIGS. 7 and 8 to be described later is realized by the CPU 101 of the scanning apparatus 100 reading out and executing the program stored in the HDD 106. This shadow influence grasping process may be started by a serviceman operation or may be started by another user operation.

  When the shadow influence grasping process is started, the scanning apparatus 100 captures the three-dimensional object 301 with the image sensor 108 in S501, acquires a captured image (400 in the example of FIG. 4), and advances the process to S502. The imaging sensor 108 incorporates an optical distortion correction mechanism such as a shift lens, and even if the imaging region 112 is imaged obliquely, a captured image 400 having no trapezoidal distortion is obtained as shown in FIG.

  In S502, the scanning apparatus 100 removes the specific color in the captured image acquired in S501 and advances the process to S503. Here, the specific color is the color of the three-dimensional object 301. By removing this specific color, the image of the three-dimensional object 301 can be removed from the captured image. Note that the shape of the three-dimensional object may be grasped using a distance sensor (not shown) and removed from the captured image acquired in S501.

In S503, the scanning apparatus 100 calculates a direction in which a shadow is generated when shooting a document from a shadow image (401 in the example of FIG. 4) included in the captured image from which the specific color is removed in S502, and proceeds to S504. . As indicated by an arrow 402 in FIG. 4B, the direction in which the shadow is generated when the document is shot is calculated as an angle θ _s from the reference direction of the shooting area.

In S504, the scanning apparatus 100 stores the shadow generation direction information calculated in S503 in the storage device (for example, the HDD 106) as shadow generation direction information, and ends the processing of this flowchart. For example, the shadow generation direction information is information as shown in FIG. In the example of FIG. 4, information indicating that the angle θ _s is 130 degrees is stored in the storage device as shadow occurrence direction information.

FIG. 6 is a diagram illustrating an example of information handled in the document scanning process according to the first embodiment.
FIG. 7 is a flowchart illustrating an example of a document scanning process according to the first embodiment. S701 to S705 in the figure are processing steps performed by the scanning apparatus 100.

  When the document scanning process is started, the scanning apparatus 100 accepts designation of the document size (size corresponding to the document) from the user via the operation panel 107 in S701, and advances the process to S702. In this embodiment, in order to scan the license, the user designates the document size by inputting from the operation panel 107 that the license is to be scanned. Note that the user may input the document size (vertical size, horizontal size) in a predetermined unit (for example, millimeter unit or inch unit).

  In S702, the scanning apparatus 100 captures a document with the image sensor 108, acquires a captured image (600 in the example of FIG. 6), and advances the process to S703. Since the license is thick, when it is photographed in an environment as shown in FIG. 3, as shown in FIG. 6A, a shadow 602 of the document is generated on the upper side and the left side of the document 601.

  In S703, the scanning apparatus 100 calculates a line segment included in the captured image acquired in S702, and among the calculated line segment, a line segment that matches the document size specified in S701 within a predetermined error. The rectangle candidate formed by is calculated, and the process proceeds to S704. In addition, the calculation (recognition) of the line segment included in the imaging is a widely performed technique such as a method of performing an edge detection process for detecting a sharp change in the brightness of the image and performing a Hough transform on the detection result. Therefore, the detailed explanation is omitted.

  A line segment as shown in FIG. 6B is calculated from the captured image 600 shown in FIG. Among the line segments, the line segments having the length constituting the license, for example, the line segments that coincide with each other within an error of 5% are Ev1 to Ev5 and Eh1 to Eh3 shown in FIG. There are a plurality of rectangle candidates formed by these line segments, such as those composed of (Ev1, Eh1, Ev5, Eh3) and those composed of (Ev2, Eh2, Ev5, Eh3).

In S704, the scanning apparatus 100 reads the shadow occurrence direction information (stored in S504 of FIG. 5) from the storage device (HDD 106), and advances the process to S705.
In S705, the scanning apparatus 100 executes a cut-out process (FIG. 8 in the first embodiment), and ends the process of this flowchart.

  FIG. 8 is a flowchart illustrating an example of the cutout process according to the first embodiment. S801 to S802 in the figure are processing steps performed by the scanning apparatus 100.

In S801, the scanning apparatus 100 calculates a normal vector of each side of each rectangle candidate calculated in S703 of FIG. 7, calculates an angle formed with the direction in which the shadow is generated, and advances the process to S802.
In the example of FIG. 6, the angle formed by the normal vector (vt, vb, vl, vr) of each side of the rectangle (upper side, lower side, left side, right side) and the direction in which the shadow is generated is shown in FIG. θ _ts , θ _bs , θ _ls , θ _rs . If this angle is 90 degrees or more, no shadow is produced on the side.

In S802, the scanning apparatus 100 uses a document-specified document size from the photographed image 600 based on the outermost side that is the side where the angle is 90 degrees or more among the sides of each rectangular candidate. The image is cut out, and the processing of this flowchart ends.
In the example of FIG. 6, the image is cut out from the photographed image 600 with the license size using the lower side Ev3 and the right side Ev5 as a reference.

  Through the above processing, it is possible to cut out a document size image based on a side where no shadow occurs. As a result, when the user scans a document with a scanning device called an overhead scanner, it is possible to obtain a product (read image data) that does not lack information contained in the document. In this embodiment, a license is used as an example of a manuscript. However, the manuscript is not limited to a license, and any manuscript may be used. The size of the manuscript is also limited to the license size. is not.

  Depending on the room in which the scanning device is used, there are a plurality of lights, and a plurality of shadows of a three-dimensional object may be generated. Depending on the direction in which the shadow is generated, there is a possibility that there is no side that is not affected by the shadow as described in the first embodiment. In the second embodiment, a specific flow of processing when there is no side that is not affected by the shadow will be described.

  FIG. 9 is a diagram illustrating an example of information handled in the shadow influence grasping process according to the second embodiment. The shadow effect grasping process in the second embodiment is the same as that in FIG. 5 in the first embodiment. Hereinafter, information handled in the shadow influence grasping process according to the second embodiment will be described with reference to FIGS.

In Example 2, when the scanning apparatus 100 captures the three-dimensional object 301 (FIG. 3) with the image sensor 108 in S501 of FIG. 5, for example, a captured image 900 illustrated in FIG. 9A is obtained.
Further, in S503 of FIG. 5, when the scanning apparatus 100 calculates the direction in which the shadow is generated from the shadow images 901 and 902 included in the captured image 900, as shown by arrows 903 and 904 in FIG. A shadow with an angle θ _{s1 and} a shadow with an angle θ _s2 are calculated from the reference direction.
In the second embodiment, the shadow occurrence direction information stored in the storage device (HDD 106) by the scanning apparatus 100 in S504 of FIG. 5 is information as shown in FIG. 9C. In the example of FIG. 9, information indicating that the angle θ _s1 is 130 degrees and the angle θ _s2 is −50 degrees is stored in the storage device as shadow generation direction information. That is, in the second embodiment, a plurality of pieces of information indicating the direction in which the shadow is generated can be stored as the shadow generation direction information.

FIG. 10 is a diagram illustrating an example of information handled in the document scanning process according to the second embodiment. The document scanning process of the second embodiment is the same as that of FIG. Hereinafter, information handled in the document scanning process according to the second embodiment will be described with reference to FIGS.
In the second embodiment, when the scanning apparatus 100 captures a document with the image sensor 108 in S702 of FIG. 7, the shadow of the document is applied to any of the top, bottom, left, and right sides of the document 1001 in the captured image 1000 as shown in FIG. 1002 is produced.

  FIG. 11 is a flowchart illustrating an example of the cutout process according to the second embodiment. S1101 to S1105 in the figure are processing steps performed by the scanning apparatus 100. Also, the processing of the flowchart shown in FIG. 11 is realized by the CPU 101 of the scanning apparatus 100 reading and executing a program stored in the HDD 106.

  When the clipping process according to the second exemplary embodiment is started, the scanning apparatus 100 calculates a normal vector of each side of each rectangular candidate calculated in S703 in FIG. Calculate and proceed to S1102.

  In S1102, the scanning apparatus 100 performs processing for extracting the sides where the angle calculated in S1101 is 90 degrees or more (that is, sides where no shadow is generated) as two sides constituting a rectangular corner, and performs processing in S1103. Proceed. In the example shown in FIG. 10, since the shadow can be shadowed in any direction of the four sides constituting the rectangle, there is no side having an angle of 90 degrees or more (that is, two sides constituting the corner of the rectangle). Become. In such a case, it is not possible to extract the two sides constituting the corner of the rectangle in S1102.

  In S1103, the scanning apparatus 100 determines whether or not the two sides constituting the corner of the rectangle have been extracted. The reason for carrying out this determination processing is that if two sides constituting a corner of a rectangle are determined, the two sides can be cut out as a reference.

In S1103, when the scanning apparatus 100 determines that the two sides constituting the corner of the rectangle have been extracted (Yes in S1103), the scanning apparatus 100 proceeds to S1105.
On the other hand, if it is determined in S1103 that the two sides constituting the corner of the rectangle cannot be extracted (No in S1103), the scanning apparatus 100 proceeds to S1104.

  In S1104, the scanning apparatus 100 determines from the outermost sides that the angle calculated in S1101 is less than 90 degrees among the sides forming the largest rectangle among the rectangle candidates calculated in S703 of FIG. Extract one inner edge. Then, the process proceeds to S1105. Hereinafter, a specific example will be described.

  In the example of FIG. 10, a line segment as shown in FIG. 10B is calculated from the imaging shown in FIG. Among the line segments, the line segments that coincide with the length of the line constituting the license, for example, within 5% error are Ev1 to Ev6 and Eh1 to Eh4 shown in FIG. In the processing of S1104, the largest rectangle of each rectangle candidate is a rectangle constituted by (Ev1, Eh1, Ev6, Eh4) shown in FIG. Among the sides forming the largest rectangle, the sides where the angle calculated in S1101 is less than 90 degrees, that is, the sides where shadows are generated, are all Ev1, Eh1, Ev6, and Eh4. Therefore, if one side inside is extracted for each of Ev1, Eh1, Ev6, and Eh4, Ev2, Eh2, Ev5, and Eh3 are obtained.

  In step S1105, the scanning apparatus 100 cuts out an image with a user-specified original size from the image pickup using the side extracted in step S1104 as a reference, and ends the processing of this flowchart.

As described above, even when there is no side where no shadow occurs due to multiple shadows, when the user scans a document with a scanning device called an overhead scanner, the product does not lack information contained in the document. (Read image data) can be obtained.
The second embodiment is particularly effective because the boundary between the document and the shadow can be accurately detected when the color of the document is easily distinguishable from the shadow density, for example. For example, the second embodiment is effective for an original such as a license whose original color is clearly different from the shadow density.

  In the above-described second embodiment, the case where a plurality of shadows are generated has been described. However, in many cases, the density differs depending on the generated shadow. Since a dark shadow is easily detected by the above-described edge detection process, there is a high possibility that an error will occur in document area recognition. However, if the shadow is thin, the possibility of causing an error decreases. In the third embodiment, the specific flow of processing in which the scanning apparatus 100 grasps the density of a plurality of shadows generated and minimizes the possibility of generating an error by preferentially considering the influence of the shadow from the dark shadow. Will be explained.

FIG. 12 is a diagram illustrating an example of information handled in the shadow influence grasping process according to the third embodiment.
FIG. 13 is a flowchart illustrating an example of the shadow influence grasping process according to the third embodiment. S1301 to S1304 in the figure are processing steps executed by the scanning apparatus 100. Further, the processing of the flowchart shown in FIG. 13 and FIG. 15 to be described later is realized by the CPU 101 of the scanning apparatus 100 reading out and executing the program stored in the HDD 106.

When the shadow influence grasping process of the third embodiment is started, the scanning apparatus 100 captures a three-dimensional object with the image sensor 108 in S1301, acquires a captured image, and advances the process to S1302.
For example, a photographed image 1200 shown in FIG. It can be seen that a dark shadow 1201 is generated on the left side of the object, and a light shadow 1202 is generated on the lower side.

In S1302, the scanning apparatus 100 removes the specific color from the captured image acquired in S1301, and advances the process to S1303.
In S1303, the scanning apparatus 100 calculates the direction and density of the shadow from the shadow image included in the captured image from which the specific color has been removed in S1302, and advances the process to S1304.
The definition of the density in the third embodiment is a numerical value in which the density is high when the shadow is dark and the density is low when the shadow is thin. For example, if the density is z and the luminance value l of the pixels included in the imaging is taken, It is expressed as Equation 1.

As indicated by arrows 1203 and 1204 in FIG. 12B, a dark shadow with an angle θ _{s1 and} a light shadow with an angle θ _s2 are calculated from the reference direction of the imaging region. With regard to the shadow density, for example, if the pixel constituting the shadow being imaged is less than 50%, it is not regarded as a shadow. If the density is less than 50% and less than 70%, the shadow is thin. Judge as a shadow.

In step S1304, the scanning apparatus 100 stores shadow generation direction and density information as shadow generation direction information in the storage device (HDD 106), and ends the process of this flowchart.
For example, the shadow occurrence direction information stored in the storage device by the scanning device 100 is information as shown in FIG. In the example of FIG. 12, information indicating that the angle θ _s1 is 180 degrees and the shadow is dark, and the angle θ _s2 is −60 degrees and the shadow is light is stored as shadow generation direction information. That is, in the third embodiment, a plurality of sets of information indicating the direction in which the shadow is generated and information indicating the density of the shadow can be stored as the shadow generation direction information.

FIG. 14 is a diagram illustrating an example of information handled in the document scanning process according to the third embodiment. The document scanning process of the third embodiment is the same as that of FIG. Hereinafter, information handled in the document scanning process according to the third embodiment will be described with reference to FIGS.
In the third embodiment, when the scanning apparatus 100 captures a document with the image sensor 108 in S702 of FIG. 7, as shown in FIG. 14A, the left side of the document 1401 in the captured image 1400 is dark due to the document. Shadow 1402 and a light shadow 1403 appear on the right and bottom sides.

  FIG. 15 is a flowchart illustrating an example of the cutout process according to the third embodiment. S1501 to S1506 in the figure are processing steps performed by the scanning apparatus 100.

When the clipping process of the third embodiment is started, the scanning apparatus 100 calculates a normal vector of each side of each rectangular candidate calculated in S703 of FIG. 7 in S1501, and a shadow having a density of a predetermined level or more is generated. The angle formed with the direction is calculated, and the process proceeds to S1502. In Example 3, a shadow having a density of 70% or more is regarded as a dark shadow. In the example of FIG. 14, the angles formed between the normal vector of each side of the rectangle and the direction in which the dark shadow is generated are θ _ts1 , θ _bs1 , θ _ls1 , and θ _rs1 shown in FIG.

  In S1502, the scanning apparatus 100 is a side where the angle calculated in S1501 is 90 degrees or more (that is, a side where a shadow with a density of a predetermined density or more does not occur) among the sides constituting each rectangle candidate, and The outermost side is extracted, and the process proceeds to S1503. In the example of FIG. 14, the upper side, the right side, and the lower side of the document are extracted.

  In S1503, the scanning apparatus 100 determines whether or not two or more rectangular parallel sides are extracted in S1502. If the scanning device 100 determines in S1502 that two or more sides that are parallel to the rectangle have not been extracted (No in S1503), the scanning device 100 proceeds to S1506.

On the other hand, if it is determined in S1503 that two or more sides that are parallel to the rectangle have been extracted in S1502 (Yes in S1503), the scanning apparatus 100 proceeds to S1504.
In the example of FIG. 14, since the upper side and the lower side of the document are parallel sides of the rectangle, the process proceeds to S1504.

  In S1504 to S1505, the scanning apparatus 100 narrows down the sides that can minimize the error due to the influence of the shadow. Details will be described below.

In S1504, the scanning apparatus 100 calculates a normal vector of each side that is parallel, calculates an angle between the normal vector and a direction in which a shadow having a density less than a predetermined value is generated, and advances the process to S1505. In this embodiment, a shadow having a density of 50% or more and less than 70% is regarded as a light shadow. In the example of FIG. 14, the angles formed by the parallel vectors, that is, the normal vectors of the upper and lower sides of the document, and the direction in which the light shadow is generated are θ _ts2 and θ _bs2 shown in FIG.

In S1505, the scanning apparatus 100 extracts a side where the angle calculated in S1504 is 90 degrees or more (that is, a side where a shadow with a density less than a predetermined value does not occur) from the parallel sides, and performs the process in S1506. Proceed. In the example of FIG. 14, since θ _ts2 is 90 degrees or more, the upper side of the document is extracted.

In step S1506, the scanning apparatus 100 cuts out an image with a user-specified original size from the image pickup using the extracted side as a reference. Here, the “extracted side” is the case of Yes in S1503, and when there are four parallel sides (two sets), the side extracted in S1504 for each group (ie, 2 sides). In addition, when there are two (one set) of parallel sides, the sides extracted in S1504 and the non-parallel sides extracted in S1502 are shown. In the case of No in S1503, the side extracted in S1502 is indicated. In the example of FIG. 4, clipping is performed based on the upper side extracted in S1505 and the right side extracted in S1502. That is, (1) in the third embodiment, a side that exists in a direction in which a shadow with a density of a predetermined level or higher (first shadow) and a shadow with a density of less than a predetermined level (second shadow) do not occur, and (2) the first shadow The sides that are the reference for extraction are extracted in the order of the sides that exist in the direction in which the second shadow does not occur and in the direction in which the second shadow occurs.
After the process of S1506, the scanning apparatus 100 ends the process of this flowchart.

  As described above, even under an environment where a plurality of shadows having different densities are generated, an error due to a shadow effect can be prevented as much as possible. For this reason, when a user scans an original with a scanning device called an overhead scanner, it is possible to obtain a product (read image data) that does not lack information contained in the original.

  In the shadow effect grasping process of the above-described first to third embodiments, as shown in FIG. 3, the user places a three-dimensional object on the shooting area, and the scanning device 100 detects the shadow generated by the three-dimensional object and the three-dimensional object. The shadow occurrence direction information was specified by photographing. In the fourth embodiment, as another shadow influence grasping process, a configuration in which shadow generation direction information is specified more simply by allowing a user to input a direction in which a shadow is generated on an operation panel will be described.

  FIG. 16 is a flowchart illustrating an example of the shadow effect grasping process according to the fourth embodiment. S1601 to S1603 in the figure are processing steps executed by the scanning apparatus 100. The processing of the flowchart shown in FIG. 16 is realized by the CPU 101 of the scanning apparatus 100 reading and executing a program stored in the HDD 106.

When the shadow influence grasping process of the fourth embodiment is started, the scanning apparatus 100 displays a screen for allowing the user to select a direction in which the shadow is generated on the operation panel 107 in S1601, and advances the process to S1602.
In S1602, the scanning apparatus 100 accepts a selection operation of the direction in which the shadow is generated by the user from the screen displayed in S1601, and advances the process to S1603.
In S1603, the scanning apparatus 100 stores the shadow generation direction received in S1602 in the storage device (HDD 106) as shadow generation direction information, and ends the processing of this flowchart.

  In addition, the shadow influence grasping process is configured such that, for example, when the scanning apparatus 100 has a flash mechanism attached to a fixed position, the direction in which the shadow caused by the flash is generated is preset by a manufacturer or the like. It may be.

  As described above, in each embodiment, the scanning device grasps the direction in which the shadow is generated before scanning the document, considers the direction in which the shadow is captured in advance when scanning the document, and uses the direction in which no shadow is generated as a reference. As shown in FIG. As a result, when scanning a document, even if the thickness of the document is not uniquely determined, shadow effects are accurately excluded during scanning, and a predetermined size of the document is prevented so that missing information contained in the document does not occur in the product. The image can be cut out. Therefore, it is possible to realize a cut-out function of a standard size that does not cause missing of information included in the document in the product.

  The document is not limited to a rectangle, and may be a triangle or a pentagon or more. In this case, it is assumed that the document shape is also specified in S701, and processing is performed using the specified document shape instead of the rectangle. Further, the document may be a document surrounded by one or a plurality of curves such as a circle or an ellipse. In the case of a document surrounded by a curve, for example, the scanning apparatus 100 recognizes a curve included in the photographed image, and within a predetermined error within a specified size and shape among figures formed by the curve. The matching figure is recognized as a document edge candidate. The normal vector of each side described above is, for example, provided with a plurality of points (for example, equidistant intervals) on a curve constituting a graphic recognized as the document edge candidate, and is used as the normal vector at that point. . Then, it is assumed that the original image is cut out with a specified shape and size with reference to a point where the angle between the normal vector and the direction indicated by the shadow generation direction information is 90 degrees or less. As a result, even in the case of a document such as a CD, a document image can be cut out with a predetermined size so that missing information contained in the document does not occur in the product.

In addition, the structure of the various data mentioned above and its content are not limited to this, You may be comprised with various structures and content according to a use and the objective.
Although one embodiment has been described above, the present invention can take an embodiment as, for example, a system, apparatus, method, program, or storage medium. Specifically, the present invention may be applied to a system composed of a plurality of devices, or may be applied to an apparatus composed of a single device.
Moreover, all the structures which combined said each Example are also contained in this invention.

(Other examples)
The present invention supplies a program that realizes one or more functions of the above-described embodiments to a system or apparatus via a network or a storage medium, and one or more processors in the computer of the system or apparatus read and execute the program This process can be realized. It can also be realized by a circuit (for example, ASIC) that realizes one or more functions.
Further, the present invention may be applied to a system composed of a plurality of devices or an apparatus composed of one device.
The present invention is not limited to the above embodiments, and various modifications (including organic combinations of the embodiments) are possible based on the spirit of the present invention, and these are excluded from the scope of the present invention. is not. That is, the present invention includes all the combinations of the above-described embodiments and modifications thereof.

100 scanning equipment
120 controller
107 Control panel
108 Imaging sensor

Claims (8)

A document reading device that reads image data from a document by photographing means,
Storage means for storing direction information indicating a direction in which a shadow is generated when a document is photographed;
First acquisition means for acquiring image data by imaging an imaging area where a document is placed by the imaging means;
Recognizing means for recognizing a figure having a predetermined shape formed by a plurality of lines that are candidates for the document edge included in the image data obtained by the first obtaining means;
Selection means for selecting a reference line from lines constituting each figure recognized as a document edge candidate by the recognition means based on the direction information stored in the storage means;
From the image data, a cutout unit that cuts out a region of a predetermined size with reference to the line selected by the selection unit;
A document reading apparatus comprising:   The selecting means selects a line having an angle of 90 degrees or more between a normal vector of each line constituting each figure and a direction indicated by the direction information as the reference line. The document reading device according to claim 1. Second acquisition means for acquiring image data by imaging the imaging area where the three-dimensional object is placed with the imaging means;
Identifying means for recognizing the shadow of the three-dimensional object included in the image data acquired by the second acquisition means and specifying the direction in which the shadow is generated;
The document reading apparatus according to claim 1, wherein the storage unit stores the direction specified by the specifying unit as the direction information. An input unit that receives an input of a direction in which the shadow is generated by the user;
The document reading apparatus according to claim 1, wherein the storage unit stores a direction input by the input unit as the direction information. The storage means can store a plurality of information indicating the direction in which the shadow occurs as the direction information,
When the selection unit cannot select the reference line based on the direction information, a line existing inside one of the outermost lines among the lines constituting each figure is used as a reference line. The document reading device according to claim 1, wherein the document reading device is selected as the document reading device. The storage means can store a plurality of sets of information indicating the direction in which the shadow occurs and information indicating the density of the shadow as the direction information,
When the direction information includes first information indicating a direction in which a shadow with a density greater than a predetermined value is generated and second information indicating a direction in which a shadow with a density less than a predetermined value is generated, A reference line candidate is selected from each line constituting each figure recognized by the recognition means based on one information, and the selected reference line candidate is selected based on the second information. The document reading apparatus according to claim 1, wherein the line is selected. A method of controlling a document reading device that reads image data from a document by a photographing means,
An acquisition step of acquiring image data by imaging the imaging area where the document is placed by the imaging means;
A recognition step for recognizing a figure having a predetermined shape formed by a plurality of lines that are candidates for document edge portions included in the image data acquired in the acquisition step;
A selection step of selecting a reference line from each line constituting each figure recognized in the recognition step, based on direction information indicating a direction in which a shadow is generated during photographing of the document stored in the storage unit;
A cutout step of cutting out a region of a predetermined size from the image data on the basis of the line selected in the selection step;
A method for controlling a document reading apparatus, comprising:   A program for causing a computer to execute the control method according to claim 7.

Images