JP6693184B2 - Image processing apparatus and image processing method - Google Patents

Description

The present invention relates to an image processing apparatus, and an image processing method.

  Patent Document 1 discloses "an image processing apparatus, an image processing method, and a program capable of appropriately obtaining a document area regardless of whether there is one document or a plurality of documents." ..

JP, 2010-288188, A

  A scan target such as a document may include lines such as ruled lines. It may be difficult to divide the image data of the scan target including such a line into a part of the scan target and a background part (for example, a cover for pressing the document) depending on the position placed on the scanning table. is there.

  It should be noted that Patent Document 1 does not describe separation of the original portion and the background portion of the image data depending on how the original is placed.

  Therefore, the present invention provides a technique capable of appropriately dividing image data into a scan target portion and a background portion.

  A first aspect of the present invention for solving the above-mentioned problems is to scan a scanning object, a scanning unit that outputs image data, an edge extraction unit that extracts an edge of the image data, and based on the edge. A non-edge area detection unit that detects a non-edge area of the image data, and the non-edge area is an image area of the scan target object or a background image area, based on a predetermined condition. An image processing unit comprising: a provisional determination unit that makes a provisional determination; and a correction unit that corrects the provisional determination of the non-edge region based on the provisionally determined arrangement of the non-edge region. According to the first aspect, the image processing apparatus can appropriately divide the image data into the scan target object portion and the background portion.

  The predetermined condition is the following two conditions, the first condition is that the non-edge region is in contact with two or more data edges of the image data, and the second condition is the non-edge. The region may not include the edge. Accordingly, the image processing apparatus can appropriately provisionally determine whether the non-edge area is the image area of the scan target object or the background image area.

  When the non-edge region does not satisfy one of the first condition and the second condition, the provisional determination unit provisionally determines the non-edge region as an image region of the scan target, When the edge region satisfies both the first condition and the second condition, the non-edge region may be provisionally determined as the image region of the background. Accordingly, the image processing apparatus can appropriately provisionally determine whether the non-edge area is the image area of the scan target object or the background image area.

  In the correction unit, the first non-edge area and the second non-edge area that are tentatively determined to be the image area of the background are adjacent to each other, and the first non-edge area is tentatively the image area of the scan target. If the second non-edge area is adjacent to the determined non-edge area and is not in contact with the non-edge area tentatively determined to be the image area of the scan target, the first non-edge area is It may be characterized in that it is corrected with the image area of the scan object. Thereby, the image processing apparatus can appropriately correct the temporary determination.

  The correction unit further includes a storage unit that stores the non-edge region arrangement combination provisionally determined by the provisional determination unit and the corrected non-edge region arrangement combination in association with each other. The provisional determination of the non-edge region may be corrected with reference to a section. As a result, the image processing apparatus can correct the provisionally determined non-edge regions having various arrangements.

  A conversion unit that converts the provisionally determined layout of the non-edge region into a simplified layout, a layout combination after layout conversion of the non-edge region that is provisionally determined by the provisional determination unit, and the corrected layout. A storage unit that stores the arrangement combination of the non-edge regions in association with each other, wherein the correction unit corrects the provisional determination of the non-edge regions by referring to the storage unit. Good. As a result, the image processing device can reduce the storage capacity of the storage unit.

  The edge extraction unit calculates a luminance dispersion difference between a luminance dispersion of a first pixel of the image data and a luminance dispersion of a second pixel around the first pixel, and based on the luminance dispersion difference, The edge of the image data may be extracted. Thereby, the image processing apparatus can appropriately extract the edge.

  A second aspect of the present invention for solving the above problem is to extract image data of a scan object scanned by an image processing apparatus from a receiving unit that receives the image data from the image processing apparatus, and to extract edges of the image data. An edge extraction unit, a non-edge region detection unit that detects a non-edge region of the image data based on the edge, and the non-edge region is an image region of the scan target object or a background image region. Or a provisional determination unit that makes a provisional determination based on a predetermined condition, and a correction unit that corrects the provisional determination of the non-edge region based on the provisionally determined arrangement of the non-edge region. Is an information processing device. According to the second aspect, the information processing apparatus can appropriately divide the image data into the scan target object portion and the background portion.

  A third aspect of the present invention for solving the above-mentioned problems is an image processing method of an image processing device, comprising a step of scanning a scan target object and outputting image data, and extracting an edge of the image data. And a step of detecting a non-edge region of the image data based on the edge, the non-edge region being an image region of the scan target object or a background image region, or a predetermined An image processing method comprising: a tentative determination based on a condition; and a step of correcting the tentative determination of the non-edge region based on the provisionally determined placement of the non-edge region. According to the third aspect, the image processing apparatus can appropriately divide the image data into the scan target portion and the background portion.

  A fourth aspect of the present invention for solving the above-mentioned problems is an image processing method for an information processing apparatus, the step of receiving image data of a scan target scanned by the image processing apparatus from the image processing apparatus. A step of extracting an edge of the image data, a step of detecting a non-edge area of the image data based on the edge, and the non-edge area being an image area of the scan target or a background Of the image area, the step of temporarily determining based on a predetermined condition, and the step of correcting the temporary determination of the non-edge area based on the provisionally determined arrangement of the non-edge area. This is a characteristic image processing method. According to the fourth aspect, the information processing apparatus can appropriately divide the image data into the scan target object portion and the background portion.

It is a schematic perspective view of the image processing apparatus according to the first embodiment. It is the 1 of the figure explaining the method of dividing image data into a manuscript part and a background part. It is the 2 of the figure explaining the method of dividing image data into a manuscript part and a background part. It is a figure showing an example of functional block composition of an image processing device. It is a figure explaining the example of calculation of the brightness | luminance dispersion difference of an edge extraction part. It is a figure explaining the example of determination of an edge. It is a figure explaining the example of detection of the non-edge area of a non-edge area detection part. It is a figure explaining a noticed pixel and a reference pixel. It is a figure explaining the non-edge area detection operation example of a non-edge area detection part. FIG. 8 is a diagram illustrating an example of determining the presence or absence of an edge area in a non-edge area. FIG. 6 is a diagram illustrating a document area and a background area after correction of image data. It is a figure explaining the adjacency determination example of a non-edge area. It is the 1st of the figure explaining the example which cuts out a manuscript. FIG. 7B is a second diagram illustrating an example of cutting out a document. FIG. 11 is a third diagram illustrating an example of cutting out a document. 6 is a flowchart showing an operation example of the image processing apparatus. It is a figure showing an example of functional block composition of an information processor. It is the figure which showed the functional block structural example of the image processing apparatus which concerns on 2nd Embodiment. FIG. 6 is a diagram showing a layout example of a non-edge region. It is a figure showing the judgment conditions for judging in which position of the layout the non-edge region is arranged. It is the figure which showed the data structural example of a memory | storage part. It is the 1st of the figure explaining the example of operation of an image processing device. It is the 2 of the figure explaining the example of operation of an image processing device. It is the third of the diagram illustrating the operation example of the image processing apparatus. It is the figure which showed the functional block structural example of the image processing apparatus which concerns on 3rd Embodiment. It is a figure explaining layout conversion. FIG. 9 is a diagram showing layout conversion conditions. It is the 1 of the figure which showed the data structural example of a memory | storage part. It is the 2 of the figure which showed the data structural example of a memory | storage part. It is the 1st of the figure explaining the example of operation of an image processing device. It is the 2 of the figure explaining the example of operation of an image processing device. It is the third of the diagram illustrating the operation example of the image processing apparatus. 6 is a flowchart showing an operation example of the image processing apparatus.

  Embodiments of the present invention will be described below with reference to the drawings.

[First Embodiment]
FIG. 1 is a schematic perspective view of an image processing device 1 according to the first embodiment. The image processing apparatus 1 illustrated in FIG. 1 is, for example, a scanner, and includes a base 2, a cover 3, and an input device 4.

  On the upper part of the table 2, a scan target whose image data is read by the image processing apparatus 1 is placed. The scan target is, for example, a document. In the following, the scan target will be described as a document.

  The cover 3 can be opened and closed with respect to the upper portion of the base 2. The cover 3 is closed with respect to the upper part of the table 2 to hold the document placed on the upper part of the table 2.

  The input device 4 is, for example, a key device or a touch panel. The input device 4 receives a predetermined operation from the user.

  In the image processing apparatus 1, when a document is placed on the table 2 and the cover 3 is closed and the input device 4 receives a predetermined operation from the user, the image data of the document placed on the table 2 is read. ..

  When the image processing device 1 reads the image data, the image processing device 1 extracts the edge of the read image data (see, for example, (A) in FIG. 7). As described in detail below, the image processing apparatus 1 detects a non-edge region of the read image data based on the extracted edge (see, for example, L1 to L3 in FIG. 7B), and detects the detected non-edge region. It is tentatively determined based on a predetermined condition whether the edge area is a document portion or a background portion (for example, a portion of the lower surface of the cover 3 scanned with the document).

  Then, the image processing apparatus 1 corrects the provisional determination of the non-edge region based on the provisionally determined arrangement of the non-edge region. For example, the image processing apparatus 1 determines, based on the arrangement of the provisionally determined non-edge regions L1 to L3 (see (B) of FIG. 7) such as “original-background-background” or “background-original-background”. The non-edge area tentatively determined to be the background portion is corrected as the original portion, and the non-edge area tentatively determined to be the original portion is corrected as the background portion.

  The image processing apparatus 1 cuts out the original portion from the read image data based on the area-corrected original portion and background portion.

  The image processing apparatus 1 transmits the image data of the cut out document portion to an information processing apparatus (not shown) such as a personal computer or a tablet terminal connected to the image processing apparatus 1, for example. Alternatively, the image processing apparatus 1 stores the cut-out image data of the original portion in an external memory (not shown) such as a USB (Universal Serial Bus) memory. As a result, the user can acquire the image data of only the document portion.

  Here, the separation of the original portion and the background portion by a method different from that of the image processing apparatus 1 will be described. An image processing apparatus that separates a document portion and a background portion by another method determines a non-edge region that is in contact with an edge of image data as a background portion, and divides the image data into a document portion and a background portion. An image processing apparatus according to another method has a base, a cover, and an input device, like the image processing apparatus 1.

  FIG. 2 is a first diagram illustrating a method of dividing image data into a document portion and a background portion. FIG. 2A shows an example of image data of a document scanned by the image processing apparatus. The scan range 11 shown in FIG. 2A corresponds to, for example, the edge of the glass surface of the table of the image processing apparatus from which light is emitted. That is, the scan range 11 is the edge of the image data.

  The background 12 corresponds to the cover scanned with the original. The document 13 is placed on the top of the table and corresponds to the scanned document. The ruled lines 14 correspond to the ruled lines drawn on the original 13. The character 15 corresponds to the character drawn on the original 13.

  FIG. 2B shows an example of edge data obtained by edge-processing the image data of FIG. The white portion in FIG. 2B is an edge area, and the black portion is a non-edge area (non-edge area). As shown by the white line in FIG. 2B, the boundary between the background 12 and the document 13, the ruled line 14, and the character 15 appear as edges.

  The image processing apparatus determines that a non-edge region in contact with an edge of image data is a background. For example, the image processing apparatus determines a non-edge region outside the large white line rectangle shown in FIG. 2B as the background. Then, the image processing apparatus determines the other area (inside the large white square shown in FIG. 2B) as the original.

  As a result, the image processing apparatus can divide the image data into a portion of the background 12 (a portion painted black) and a portion of the document 13 (a portion painted white), as shown in FIG. 2C. it can.

  FIG. 3 is a second diagram for explaining a method of dividing the image data into a document portion and a background portion. In FIG. 3, a scan range 21, a background 22, a document 23, ruled lines 24, and characters 25 are shown.

  FIG. 3A shows an example of image data when the document 23 is placed on a table in contact with or beyond the scan range 21 and scanned. FIG. 3B shows an example of edge data obtained by edge-processing the image data of FIG.

  The image processing apparatus determines the non-edge region in contact with the edge of the image data as the background, as in the description of FIG. However, as shown in FIG. 3B, the three non-edge regions 26a to 26c are all in contact with the edges of the image data. Therefore, the image processing apparatus determines that the original 23 is also the background 22, and it is difficult to properly divide the image data into the background 22 and the original 23. Further, even if the non-edge area 26a including the edge of the character "A" is determined to be the original, the non-edge area 26b does not include the edge and is not determined to be the original.

  That is, as shown in FIG. 2, the image processing apparatus can properly separate the original portion and the background portion if the original 13 is placed on the table so that the original 13 fits within the scan range 11. As shown in FIG. 3, when the document 23 is placed on the table in contact with or beyond the scan range 21 and scanned, the document part and the background part cannot be properly separated.

  Therefore, the image processing apparatus 1 of FIG. 1 appropriately separates the original portion and the background portion of the image data even when the original 23 is placed on the table 2 in contact with or beyond the scan range 21.

  FIG. 4 is a diagram showing a functional block configuration example of the image processing apparatus 1. As shown in FIG. 4, the image processing device 1 includes a scan unit 31, a control unit 32, a communication unit 33, an input unit 34, and a storage unit 35.

  The scanning unit 31 includes, for example, an image sensor (not shown), optically scans a document placed on the table 2, and outputs image data of the document. The scanning unit 31 outputs, for example, color (RGB) image data. The image data output from the scanning unit 31 is stored in the storage unit 35.

  The control unit 32 integrally controls the operation of the image processing device 1. As will be described in detail below, the control unit 32 divides the image data of the document scanned by the scanning unit 31 (image data stored in the storage unit 35) into a background portion and a document portion, and cuts out the document portion.

  The communication unit 33 communicates with an information processing device such as a personal computer or a tablet terminal. The communication unit 33 communicates with the information processing device, for example, by wire or wirelessly.

  The input unit 34 receives operation data from the user from the input device 4.

  The storage unit 35 stores programs and data for the control unit 32 to perform calculation processing and control processing. The storage unit 35 also stores a program, data, and the like for the control unit 32 to realize a predetermined application function. Various programs, data, and the like may be stored in advance in a non-volatile recording medium, or the control unit 32 may receive them from the server via the communication network and store them in the storage unit 35. The storage unit 35 includes, for example, various IC (Integrated Circuit) memories such as ROM (Read Only Memory), flash ROM, and RAM (Random Access Memory), hard disks, and recording media such as memory cards.

  The control unit 32 includes an edge extraction unit 41, a non-edge region detection unit 42, a temporary determination unit 43, a correction unit 44, and a cutout unit 45. The functions of the respective units of the control unit 32 are realized by, for example, a CPU (Central Processing Unit) that executes a program stored in the storage unit 35. Each unit of the control unit 32 may realize its function by a custom IC (Integrated Circuit) such as an ASIC (Application Specific Integrated Circuit), or may realize its function by a CPU and an ASIC. ..

  The edge extraction unit 41 extracts the edge of the image data of the document scanned by the scanning unit 31. For example, it is assumed that the scanning unit 31 outputs image data as shown in FIG. In this case, the edge extraction unit 41 outputs edge data as shown in FIG.

  The edge extracting operation of the edge extracting unit 41 will be specifically described. The edge extraction unit 41 converts the RGB image data of the document scanned by the scanning unit 31 into luminance image data. The edge extraction unit 41 converts the RGB image data into luminance image data by the following equation (1).

  Next, the edge extraction unit 41 uses the image data of the brightness calculated by the equation (1) to calculate the brightness average of the pixel of interest and the surrounding pixels. Then, the edge extraction unit 41 calculates the luminance dispersion of the target pixel using the calculated luminance average, and calculates the luminance dispersion difference between the target pixel and surrounding pixels.

  FIG. 5 is a diagram illustrating an example of calculation of the luminance dispersion difference of the edge extraction unit 41. The squares shown in FIG. 5 indicate pixels. “P0” indicates the target pixel. “P1 to P8” indicate pixels around the target pixel (pixels adjacent to the target pixel). Hereinafter, the pixels around the pixel of interest may be referred to as “surrounding pixels”.

The edge extraction unit 41 calculates the average luminance of the target pixel P0 and the surrounding pixels P1 to P8. Specifically, the edge extraction unit 41 calculates the luminance average “Y _u ” of the target pixel P0 and the surrounding pixels P1 to P8 by the following equation (2).

Here, “x _j , y _i ” in the equation (2) indicates the coordinates of the pixel. “Y (x _j , y _i )” indicates the brightness of the pixel at the coordinates “x _j , y _i ” (the brightness calculated by equation (1)). That is, the edge extraction unit 41 calculates the average luminance “Y _u ” of the pixels in the thick frame shown in FIG.

The edge extraction unit 41, after calculating the brightness average by the equation (2), calculates the brightness dispersion of the pixel of interest P0 using the calculated brightness average. Specifically, the edge extraction unit 41 calculates the luminance variance “Y _s ” of the pixel of interest P0 by the following equation (3).

The edge extraction unit 41 calculates the luminance variance “Y _s ” for all the pixels of the image data. For example, the edge extraction unit 41 moves the target pixel P0 like a raster scan, and calculates the luminance variance “Y _s ” of all the pixels of the image data.

  After calculating the luminance dispersion of all pixels, the edge extraction unit 41 calculates the luminance dispersion difference between the luminance dispersion of the target pixel and the luminance dispersion of the surrounding pixels. For example, the edge extraction unit 41 calculates the difference (luminance dispersion difference) between the luminance dispersion of the pixel of interest P0 shown in FIG. 5 and the luminance dispersion of each of the surrounding pixels P1 to P8. More specifically, the edge extraction unit 41 calculates the brightness dispersion difference between the peripheral pixel P1 and the target pixel P0, calculates the brightness dispersion difference between the peripheral pixel P2 and the target pixel P0, and so on. , The luminance dispersion difference between the surrounding pixel P8 and the target pixel P0 is calculated.

  After calculating the brightness dispersion difference between the target pixel and the surrounding pixels, the edge extraction unit 41 determines the edge of the image data based on the calculated brightness dispersion difference. For example, the edge extraction unit 41 has at least one absolute value of the luminance dispersion difference between the pixel of interest P0 and the surrounding pixels P1 to P8 and a predetermined threshold value (hereinafter, referred to as “luminance dispersion difference threshold value”). If it exceeds the certain value, the target pixel P0 is determined to be an edge pixel.

  The edge extraction unit 41 calculates the luminance dispersion difference for all the pixels of the image data. For example, the edge extraction unit 41 moves the target pixel P0 like a raster scan, and calculates the luminance dispersion difference for all the pixels of the image data. Then, the edge extraction unit 41 determines edges for all pixels of the image data.

  FIG. 6 is a diagram illustrating an example of edge determination. FIG. 6 shows image data of a document scanned by the scanning unit 31. 6, the same parts as those in FIG. 3 are designated by the same reference numerals. Each of the luminance dispersion differences D0 to D3 shown in FIG. 6 indicates the luminance dispersion difference (absolute value) of the pixel indicated by the arrow head of each of the arrows A0 to A3.

  As shown in FIG. 6, the luminance dispersion difference D1 of the pixel indicated by the arrow A1 exceeds the luminance dispersion difference threshold, and the pixel indicated by the arrow A1 is determined to be an edge. The luminance dispersion difference D3 of the pixel indicated by the arrow A3 exceeds the luminance dispersion difference threshold, and the pixel indicated by the arrow A3 is determined to be an edge. Although not shown in FIG. 6, the luminance dispersion difference of the character 25 exceeds the luminance dispersion difference threshold and is determined as an edge. As a result, the edge of the image data shown in FIG. 6 becomes as shown in (A) of FIG. 7 below, for example.

  Returning to the explanation of FIG. The non-edge area detection unit 42 detects the non-edge area of the image data based on the edge (edge data) of the image data extracted by the edge extraction unit 41. For example, the non-edge area detection unit 42 detects an area closed by an edge and an edge of image data or an area closed by an edge as a non-edge area.

  FIG. 7 is a diagram illustrating an example of detection of a non-edge area by the non-edge area detection unit 42. In FIG. 7A, the edge data of the image data shown in FIG. 6 is shown. The non-edge area detection unit 42 detects the non-edge areas L1 to L3 ((B) in FIG. 7) of the image data shown in FIG. 6 based on the edge data shown in (A) in FIG.

  An operation example of the non-edge area detection unit 42 will be specifically described. First, the definition of the pixel of interest and the reference pixel will be described.

  FIG. 8 is a diagram illustrating a target pixel and a reference pixel. “Note” shown in FIG. 8 indicates a target pixel, and “reference” indicates a reference pixel. When the target pixel is an edge pixel, the reference pixels are a pixel adjacent to the left of the target pixel, a pixel on the upper left, a pixel on the upper, and a pixel on the upper right, as shown in FIG. 8A. When the pixel of interest is a non-edge pixel, the reference pixels are the pixel to the left of and adjacent to the pixel of interest, as shown in FIG. 8B.

  The non-edge area detection unit 42 allocates label values to the target pixel. When there is no reference pixel having the same edge value as the target pixel (a value indicating whether it is an edge or a non-edge), the non-edge region detection unit 42 sets “1” to the label value assigned last to the target pixel. The added label value is assigned (the label value assigned first is "1"). However, when the reference pixel having the same edge value as the target pixel exists, the non-edge region detection unit 42 assigns the smallest label value of the reference pixel to the target pixel. In addition, when the smallest label value of the reference pixel is assigned to the target pixel, the non-edge region detection unit 42 corrects the label value of the reference pixel having the same edge value as the target pixel to the smallest label value.

  FIG. 9 is a diagram illustrating an example of the non-edge area detection operation of the non-edge area detection unit 42. The cells shown in FIG. 9 correspond to the pixels of the image data. The hatched cells correspond to non-edge pixels, and the unhatched cells correspond to edge pixels. The underlined label value pixel shown in FIG. 9 indicates the pixel of interest.

  As shown in FIG. 9A, the non-edge area detection unit 42 sequentially assigns label values from the upper left pixel to the right pixel. When the reference pixel having the same edge value as the target pixel does not exist, the non-edge region detection unit 42 assigns the label value obtained by adding “1” to the label value assigned last to the target pixel. In addition, when there is a reference pixel having the same edge value as the target pixel, the non-edge area detection unit 42 assigns the smallest label value of the reference pixel to the target pixel.

  For example, the pixel below the pixel at the upper left corner of FIG. This pixel of interest is a non-edge pixel. In this case, the reference pixel is only the upper left pixel based on the definition of the reference pixel in FIG. The pixel at the upper left corner is a non-edge pixel having the same edge value as the pixel of interest, and the label value is "1", so the label value of the pixel of interest is "1".

  It is assumed that the pixel of interest has moved to the right of the label value “7” shown in FIG. Since this target pixel is an edge pixel, the label value is “2” as shown in FIG. 9B, based on the definition of the reference pixel in FIG. Further, the reference pixel to the left of the pixel of interest having the label value “2” is the same edge pixel as the pixel of interest, and the non-edge region detection unit 42 sets the corrected label value “7” of the reference pixel to that of FIG. As shown in (B), the correction label value is changed to "2".

  When the non-edge area detection unit 42 repeats the above operation, the label value shown in FIG. 9C is assigned to the pixel. When the label value is assigned to all the pixels, the non-edge area detection unit 42 corrects the label value to the corrected label value. For example, the non-edge area detection unit 42 corrects the label value “7” to the corrected label value “2”. As a result, the label value assigned to the pixel becomes as shown in FIG.

  In this way, the non-edge area detection unit 42 assigns the label value to the edge area and the non-edge area of the image data. As a result, the non-edge area detection unit 42 can detect the non-edge area of the image data by distinguishing the label value. For example, the non-edge area detection unit 42 may assign the non-edge area to which the label value “1” is assigned, the non-edge area to which the label value “3” is assigned, and the non-edge area to which the label value “5” is assigned. And can be detected.

  The non-edge area detection operation of the non-edge area detection unit 42 is not limited to the above operation example. A general image data processing technique can be used to detect the non-edge region of the image data.

  Returning to the explanation of FIG. When the non-edge region detected by the non-edge region detection unit 42 satisfies both of the following two conditions, the tentative determination unit 43 tentatively determines the non-edge region as a background portion (background region). On the other hand, when the non-edge area detected by the non-edge area detection unit 42 does not satisfy at least one of the following two conditions, the tentative determination unit 43 temporarily determines the non-edge area as the original portion (original area). judge.

First tentative determination condition: Must be in contact with data edges of two or more sides of image data Second tentative determination condition: No edge is included

  For example, FIG. 7B shows the non-edge regions L1 to L3 detected by the non-edge region detection unit 42 as described above. Since the non-edge region L1 includes a character edge, the non-edge region L1 does not satisfy the second provisional determination condition. Therefore, the provisional determination unit 43 provisionally determines the non-edge area L1 as the original area. On the other hand, the non-edge regions L2 and L3 satisfy the first provisional determination condition and the second provisional determination condition. Therefore, the provisional determination unit 43 provisionally determines the non-edge regions L2 and L3 as background regions.

  The non-edge regions L1 and L3 are in contact with the data ends of the three sides. The non-edge region L2 is in contact with the data ends of the two sides.

  When the original is hollowed out, the non-edge region of the hollowed-out portion does not satisfy the first provisional determination condition, and thus is provisionally determined as the original region. However, as will be described later, since the original area is cut out in a rectangular shape, there is no problem even if the cutout portion is provisionally determined as the original.

  Further, as the non-edge region which is in contact with only one side of the image data, it is conceivable that the document has a missing edge. In this case, since the non-edge area does not satisfy the first provisional determination condition, it is provisionally determined as the original area. However, similarly to the above, since the original area is cut out in a rectangular shape, there is no problem even if the missing portion is provisionally determined as the original.

  The provisional determination unit 43 determines whether the non-edge region includes an edge region (whether or not the second provisional determination condition is satisfied) based on the label value calculated by the non-edge region detection unit 42. Can be determined.

  FIG. 10 is a diagram illustrating an example of determining the presence / absence of an edge area in a non-edge area. FIG. 10 shows an example of the label value shown in FIG. 9 (D). If the pixel of interest is an edge, the provisional determination unit 43 refers to the label values of the eight pixels surrounding the pixel of interest. The tentative determination unit 43 determines that the target pixel is included in the non-edge region when the surrounding eight pixels are non-edge and the label value is one type.

  For example, the pixel having the label value “6” shown in FIG. 10 is an edge. When this is the target pixel, the surrounding 8 pixels are non-edges with one type of label value (label value “1”). Therefore, in this case, the provisional determination unit 43 determines that the non-edge region with the label value “1” includes the edge region with the label value “6”. That is, the provisional determination unit 43 can determine the second determination condition as described above.

  In the above description, the provisional determination unit 43 refers to the label values of the eight pixels around the pixel of interest, but the present invention is not limited to this. For example, the provisional determination unit 43 may refer to the label values of n × n surrounding pixels. Further, the non-edge area detection unit 42 may execute the determination as to whether or not the non-edge area includes an edge area. That is, the provisional determination unit 43 may determine the second provisional determination condition using the determination result of the non-edge region detection unit 42.

  Returning to the explanation of FIG. The correction unit 44 corrects the provisional determination result of the non-edge region based on the arrangement of the non-edge region provisionally determined by the provisional determination unit 43. For example, it is assumed that two non-edge regions, which have been provisionally determined to be the background region, are adjacent to each other. It is assumed that one of them is adjacent to the document region and the non-edge region provisionally determined, and the other is not adjacent to the document region non-edge region provisionally determined. In this case, the correction unit 44 corrects the non-edge area tentatively determined to be the background area, which is adjacent to the non-edge area tentatively determined to be the document area, to the original area. Further, when the two adjacent background areas are not adjacent to the original area, the correction unit 44 corrects both background areas to the original area.

  For example, in FIG. 7B, the non-edge regions L2 and L3 that are provisionally determined to be the background region are adjacent to each other. Of the adjacent non-edge areas L2 and L3, the non-edge area L2 is adjacent to the non-edge area L1 of the original area. Therefore, the correction unit 44 corrects the temporary determination of the non-edge area L2 to the original area.

  FIG. 11 is a diagram illustrating the original area and the background area after the correction of the image data. An area L1 'shown in FIG. 11 shows an original area of image data. Area L3 'indicates the background area of the image data. The non-edge area L2 shown in FIG. 7B is corrected to the original area as shown in FIG.

  The correction unit 44 can determine whether or not the non-edge regions are adjacent to each other based on the label value calculated by the non-edge region detection unit 42.

  FIG. 12 is a diagram illustrating an example of determining the adjacency of a non-edge region. FIG. 12 shows an example of the label value shown in FIG. 9 (D). When the target pixel is an edge, the correction unit 44 refers to, for example, label values of pixels in a 5 × 5 area around the target pixel. The correction unit 44 determines that non-edge regions in the surrounding 5 × 5 area are adjacent to each other when the non-edge pixels are included in the surrounding 5 × 5 area and there are two or more label values. judge.

  In the case of the example in FIG. 12, the non-edge area with the label value “1” and the non-edge area with the label value “3” are included in the surrounding 5 × 5 area indicated by the thick line. Therefore, the correction unit 44 determines that the non-edge area with the label value “1” and the non-edge area with the label value “3” are adjacent to each other.

  In the above description, the correction unit 44 refers to the label value of 5 × 5 pixels around the target pixel, but the present invention is not limited to this. For example, the correction unit 44 may refer to the label values of n × n surrounding pixels. Further, the non-edge area detection unit 42 may execute the determination as to whether or not the non-edge areas are adjacent to each other. That is, the correction unit 44 may use the determination result of the non-edge region detection unit 42 to determine whether or not the non-edge regions are adjacent to each other.

  Returning to the explanation of FIG. The cutout unit 45 cuts out a document portion from the image data output from the scanning unit 11, based on the document region and the background region of the image data corrected by the correction unit 44.

  FIG. 13 is a first diagram illustrating an example of cutting out a document. FIG. 13 shows a document area (white portion) and a background area (black portion) of the image data corrected by the correction unit 44.

  The cutout unit 45 searches the document area from the right end of the image data toward the left end (toward the −x axis direction). For example, the clipping unit 45 searches for the document area in the −x-axis direction, as shown by the arrow in FIG. The cutout unit 45 stores the coordinates (x, y) of the document area initially searched (discovered) in the storage unit 35 as the left end coordinate of the document.

  The cutout unit 45 searches the first original area in the −x axis direction while incrementing the y coordinate by one.

  FIG. 14 is a second diagram illustrating an example of cutting out a document. The black circles shown in FIG. 14 indicate the coordinates of the original area first searched by the clipping unit 45 in the −x-axis direction. Although only 17 coordinates are shown in FIG. 14, there are actually many more.

  The clipping unit 45 determines that the median value of the coordinates of the first document area searched in the −x axis direction is the right edge of the document.

  FIG. 15 is a third diagram illustrating an example of cutting out a document. As illustrated in FIG. 15, the cutout unit 45 rearranges the coordinates of the document area searched in the −x-axis direction in descending order based on the x-coordinate value. Then, the clipping unit 45 determines the coordinate of the median value of the rearranged x coordinate values as the right edge of the document. For example, the clipping unit 45 determines that the x coordinate value of the coordinate of the arrow A11 shown in FIG. 15 is the coordinate value of the right end of the document.

  The clipping unit 45 searches for the first original area in the same manner as in FIGS. 13 and 14 from the upper end of the image data toward the lower end (whether toward the + y axis direction). Then, similarly to FIG. 15, the clipping unit 45 determines the y coordinate value of the median value of the first document area searched for as the coordinate value of the upper end of the document.

  In addition, the cutout unit 45 searches for the first original area in the same manner as in FIGS. 13 and 14 from the lower end of the image data toward the upper end (toward the −y axis direction). Then, similarly to FIG. 15, the clipping unit 45 determines the y-coordinate value of the median value of the first document area searched for as the coordinate value of the lower end of the document.

  In addition, the clipping unit 45 searches for the first original area in the same manner as in FIGS. 13 and 14 from the left end of the image data toward the right end (whether toward the + x axis direction). Then, the clipping unit 45 determines the x-coordinate value of the median value of the searched edge pixels as the coordinate value of the left end of the document, similarly to FIG.

  Then, the cutout unit 45 cuts out the document portion from the image data output from the scanning unit 11, using the coordinate values (rectangular regions) indicating the upper, lower, left, and right edges of the document determined as described above. As a result, for example, the white portion (original portion) shown in FIG. 13 is cut out.

  FIG. 16 is a flowchart showing an operation example of the image processing apparatus 1. For example, when the input device 4 receives a predetermined operation from the user, the image processing device 1 executes the process of the flowchart shown in FIG.

  First, the scanning unit 11 scans a document placed on the table 2 and outputs image data of the document (step S1).

  Next, the edge extraction unit 41 extracts an edge of the image data output in step S1 (step S2).

  Next, the non-edge area detection unit 42 detects the non-edge area of the image data based on the edge extracted in step S2 (step S3).

  Next, the provisional determination unit 43 determines whether the non-edge region detected in step S3 is a document region or a background region, based on the first provisional determination condition and the second provisional determination condition. Tentative determination is made (step S4).

  Next, the correction unit 44 corrects the non-edge area temporarily determined in step S4 based on the provisionally determined arrangement of the original area and the background area (step S5).

  Next, the cutout unit 45 cuts out a document portion from the image data output in step S1 based on the document region and the background region corrected in step S5 (step S6). Then, the image processing apparatus 1 ends the process of the flowchart.

  In this way, the scanning unit 11 scans the document placed on the table 2 and outputs the image data of the document. The edge extraction unit 41 extracts an edge of the image data output from the scanning unit 11, and the non-edge region detection unit 42 detects a non-edge region of the image data based on the extracted edge. Then, the provisional determination unit 43 provisionally determines, based on the first provisional determination condition and the second provisional determination condition, whether the detected non-edge region is the original region or the background region, and the correction unit 44. Corrects the provisional determination of the non-edge region based on the provisionally determined placement of the non-edge region. As a result, the image processing apparatus 1 can appropriately divide the image data into the original portion and the background portion. For example, the image processing apparatus 1 can appropriately divide the image data into a document portion and a background portion even if the document is placed on the table 2 in contact with or beyond the scanning range.

  The function of the control unit 32 illustrated in FIG. 4 may be included in an information processing device such as a personal computer or a tablet terminal.

  FIG. 17 is a diagram showing a functional block configuration example of the information processing device 50. As shown in FIG. 17, the information processing device 50 includes a control unit 32a, an I / F unit 51, and a storage unit 52.

  The control unit 32a, the edge extraction unit 41a, the non-edge region detection unit 42a, the temporary determination unit 43a, the correction unit 44a, and the cutout unit 45a illustrated in FIG. It has the same functions as the edge area detection unit 42, the temporary determination unit 43, the correction unit 44, and the cutout unit 45, and a description thereof will be omitted.

  The I / F unit 51 (corresponding to the receiving unit of the present invention) receives the image data of the document scanned by the scanning unit 11 of the image processing apparatus 1 by wire or wirelessly. The received image data is stored in the storage unit 52.

  The storage unit 52 stores an application program that realizes the function of the control unit 32a. The storage unit 52 is a storage device such as an HDD (Hard Disk Drive) or an SSD (Solid State Drive).

  The function of the control unit 32a is realized by the CPU that executes the application program stored in the storage unit 52. The control unit 32a determines the original area and the background area of the non-edge area for the image data received by the I / F unit 51 (the image data stored in the storage unit 52) in the same manner as described above. Temporarily judge and correct. Then, the control unit 32a cuts out a document portion from the image data based on the document region and the background region of the corrected image data.

  In this way, the information processing device 50 may divide the image data into the original area and the background area and cut out the original portion.

[Second Embodiment]
In the second embodiment, the image processing apparatus includes a storage unit that stores the temporarily determined non-edge area arrangement combination and the corrected non-edge area arrangement combination in association with each other. The image processing apparatus refers to the storage unit and corrects the temporarily determined original area and background area.

  FIG. 18 is a diagram showing a functional block configuration example of the image processing apparatus 1 according to the second embodiment. 18, the same parts as those in FIG. 4 are designated by the same reference numerals. Below, a part different from FIG. 4 will be described.

  As shown in FIG. 18, the image processing device 1 includes a temporary determination unit 61, a correction unit 62, and a storage unit 63. The provisional determination unit 61 has the same function as the provisional determination unit 43, but also determines which position of the image data the provisionally determined non-edge region is located. Here, a possible layout of the non-edge region will be described.

  FIG. 19 is a diagram showing a layout example of the non-edge region. If the non-edge region is in contact with two or more data edges of the image data according to the first provisional determination condition, the non-edge region may be provisionally determined as the background region. On the other hand, the non-edge area may be tentatively determined to be the original area unless it contacts two or more data edges of the image data. Therefore, as the layouts of the non-edge areas that are tentatively determined to be the background area and the original area, three layouts A to C can be considered, as shown in FIGS. "2A", "2B" ... "3A", "3B" ... "4" shown in FIG. 19 indicate the position of the layout area, and the number indicates the number of sides in contact with the side of the layout. There is.

  FIG. 20 is a diagram showing determination conditions for determining at which position in the layout the non-edge region is arranged. In the leftmost column of the table shown in FIG. 20, the areas of the layouts A to C shown in FIG. 19 are shown. The position of the data end of the image data is shown in the uppermost column of the table shown in FIG. 20, "1" indicates that the leftmost column area is in contact with the position of the data edge shown in the uppermost column, and "0" indicates the leftmost column area is the data edge shown in the uppermost column. It means that it is not in contact with the position of. The US mark indicates that either "0" or "1" may be used.

  For example, the non-edge region detected by the non-edge region detection unit 42 contacts the upper left end point, the upper side, and the left side of the image data, and the upper right end point, the lower left end point, the lower right end point, the lower side, and the right side Suppose you don't touch them. In this case, the tentative determination unit 61 determines that the non-edge region is the non-edge region located at the position "2A" in the layout (image data) based on the determination condition of FIG.

  Returning to the explanation of FIG. The correction unit 62 refers to the storage unit 63 and corrects the temporarily determined non-edge region.

  In the storage unit 63, the arrangement combination of the non-edge areas temporarily determined by the provisional determination unit 61 and the arrangement combination of the non-edge areas after the correction are stored in association with each other. The storage unit 63 also stores the same information as the storage unit 35 described with reference to FIG.

  FIG. 21 is a diagram showing a data configuration example of the storage unit 63. As shown in FIG. 21, the storage unit 63 has a provisional determination column and a post-correction column. Each of the tentative determination column and the corrected column has a column of the area of the layout shown in FIG.

  Although the layout area column is partially omitted in FIG. 21, there are 15 in each of the provisional determination and the corrected area (see the leftmost column in FIG. 20). Further, the layout area columns are arranged in order of “2A”, “2B”, “2C”, “2D” ... “3C”, “3D”, “4”, but due to space limitations, “2A”, "2B", "2C", "2D" ... could not be described in order. Therefore, the columns of the two areas of the layout after the provisional determination and the correction shown in FIG. 21 are misaligned.

The provisional determination field stores the arrangement combination of the document area and the background area of the non-edge area that is provisionally determined by the provisional determination unit 61. For example, as shown in the data D1 in FIG. 21, when the area of “2A” does not exist in the provisional determination field (NA indicates that there is no non-edge area), “2I” indicates the background area. An arrangement combination of the original area and the background area is stored, such as in the case of temporary determination. Note that there are three layout area fields in the provisional determination field: a case in which there is no non-edge area, a provisional determination in the original area, and a provisional determination in the background area. There are 3 ¹⁵ combinations of all the arrangements of the original area and the background area.

  The post-correction column stores the layout combination of the post-correction non-edge region of the original area and the background area, which corresponds to the provisional determination field. For example, as shown in the data D1 in FIG. 21, it can be seen that the non-edge area tentatively determined to be the “2I” background area is corrected to the original area from the corrected “2I” column. In the column after correction, after the temporarily determined background areas are adjacent to each other, and if one of them is adjacent to the original area, the one of the background areas is corrected to the original area. The arrangement combination of the non-edge regions is stored.

  The operation of the image processing apparatus 1 of FIG. 18 will be described below.

  FIG. 22 is a first diagram illustrating an example of the operation of the image processing apparatus 1. FIG. 22A shows an example of image data of a document scanned by the scanning unit 31. In FIG. 22A, a scan range 71a, a background 72a, a document 73a, ruled lines 74a, and characters 75a are shown.

  In FIG. 22B, the edge (white portion) of the image data of FIG. 22A is shown. When the edge extraction unit 41 extracts the edge of the image data of FIG. 22A, the edge becomes as shown in FIG.

  FIG. 22C shows the non-edge regions L11 to L13 detected based on the edges of FIG. 22B. When the non-edge area detection unit 42 detects the non-edge area based on the edge in FIG. 22B, the non-edge areas L11 to L13 are detected as illustrated in FIG. 22C.

  22D shows the provisional determination of the non-edge regions L11 to L13 of FIG. 22C. When the provisional determination unit 61 provisionally determines the non-edge regions L11 to L13 in FIG. 22C based on the first provisional determination condition and the second provisional determination condition, as shown in FIG. First, the non-edge region L11 is provisionally determined as a document region (white-filled portion), and the non-edge regions L12 and L13 are provisionally determined as background regions (black-filled portion).

  Further, the provisional determination unit 61 determines at which position in the image data the provisionally determined non-edge regions L11 to L13 are arranged. For example, the non-edge region L11 is in contact with the upper left end point, the lower left end point, the upper side, the lower side, and the left side of the image data, and is not in contact with the upper right end point, the lower right point, and the right side. Therefore, the tentative determination unit 61 determines that the non-edge region L11 is arranged in the region “3A” based on the determination conditions shown in FIG. Similarly, the provisional determination unit 61 determines that the non-edge area L12 is located in "2I" and the non-edge area L13 is located in "3B". That is, the provisional determination unit 61 provisionally determines the area of “3A” as the original area, the area of “2I” as the background area, and the area of “3B” as the background area. Note that the arrangement of the non-edge regions in FIG. 22D corresponds to the layout B in FIG. 19B.

  FIG. 22E shows the image after correction of the provisional determination of FIG. The correction unit 62 corrects the tentative determination of (D) of FIG. 22 with reference to the storage unit 63. The provisional determination in (D) of FIG. 22 corresponds to the provisional determination column of the data D1 shown in FIG. 21, and the area of “2I” is corrected from the background area to the original area.

  FIG. 23 is the second part of the diagram illustrating the operation example of the image processing apparatus 1. FIG. 23A shows an example of image data of a document scanned by the scanning unit 31. In FIG. 23A, a scan range 71b, a background 72b, a document 73b, and characters 74b are shown. The dotted line rectangle shown in FIG. 23A corresponds to the document placed on the table 2.

  In FIG. 23B, the edge (white portion) of the image data of FIG. 23A is shown. When the edge extracting unit 41 extracts the edge of the image data of FIG. 23A, the edge becomes as shown in FIG. 23B.

  In FIG. 23C, the non-edge regions L21 and L22 detected based on the edges in FIG. 23B are shown. When the non-edge area detection unit 42 detects the non-edge area based on the edge in FIG. 23B, the non-edge areas L21 and L22 are detected as illustrated in FIG. 23C.

  FIG. 23D shows the provisional determination of the non-edge regions L21 and L22 of FIG. 23C. When the provisional determination unit 61 provisionally determines the non-edge regions L21 and L22 in FIG. 23C based on the first provisional determination condition and the second provisional determination condition, as shown in FIG. First, the non-edge area L21 is provisionally determined as a background area (black-painted portion), and the non-edge area L22 is provisionally determined as a document area (white-painted portion).

  Further, the provisional determination unit 61 determines at which position in the image data the non-edge regions L21 and L22 that have been provisionally determined are arranged. For example, the non-edge area L21 is in contact with the upper left corner, the upper edge, and the left edge of the image data, and is not in contact with the upper right corner, the lower left edge, the lower right edge, the lower edge, and the right edge. Therefore, the tentative determination unit 61 determines that the non-edge region L21 is located at “2A” based on the determination conditions shown in FIG. Similarly, the provisional determination unit 61 determines that the non-edge region L22 is located at "4". That is, the provisional determination unit 61 provisionally determines the area “2A” as the background area and the area “4” as the original area. Note that the arrangement of the non-edge regions in FIG. 23D corresponds to the layout A in FIG. 19A.

  FIG. 23 (E) shows the state after correction of the provisional determination of FIG. 23 (D). The correction unit 62 corrects the tentative determination of (D) of FIG. 23 with reference to the storage unit 63. The provisional determination in FIG. 23D corresponds to the provisional determination column of the data D2 shown in FIG. 21, and the area of “2A” is corrected from the background area to the original area.

  FIG. 24 is a third part of the diagram illustrating the operation example of the image processing apparatus 1. FIG. 24A shows an example of image data of a document scanned by the scanning unit 31. In FIG. 24A, a scan range 71c, backgrounds 72c and 73c, a document 74c, characters 75c, and an underline 76c are shown. It is assumed that the character 75c and the underline 76c are connected. The square of the dotted line shown in FIG. 24A corresponds to the document placed on the table 2.

  In FIG. 24B, the edge (white portion) of the image data of FIG. 24A is shown. When the edge extracting unit 41 extracts the edge of the image data of FIG. 24A, the edge becomes as shown in FIG.

  In FIG. 24C, the non-edge regions L31 to L34 detected based on the edges in FIG. 24B are shown. When the non-edge area detecting unit 42 detects the non-edge area based on the edge of FIG. 24B, the non-edge areas L31 to L34 are detected as shown in FIG.

  FIG. 24D shows the provisional determination of the non-edge regions L31 to L34 of FIG. 24C. When the provisional determination unit 61 provisionally determines the non-edge regions L31 to L34 in FIG. 24C based on the first provisional determination condition and the second provisional determination condition, as shown in FIG. In addition, the non-edge regions L31 to L34 are provisionally determined to be background regions (blacked portions).

  Further, the provisional determination unit 61 determines at which position in the image data the provisionally determined non-edge regions L31 to L34 are arranged. For example, the non-edge region L33 is in contact with the upper right end point, the upper side, the left side, and the right side of the image data, and is not in contact with the upper left end point, the lower left end point, the lower right end point, and the lower side. Therefore, the provisional determination unit 61 determines that the non-edge region L33 is located at “3C” based on the determination conditions shown in FIG. Similarly, the provisional determination unit 61 determines that the non-edge area L31 is located in “2A”, the non-edge area L32 is located in “2C”, and the non-edge area L34 is “3D”. It is determined to be placed in ". That is, the provisional determination unit 61 determines the areas of “2A”, “2C”, “3C”, and “3D” as the background area. Note that the arrangement of the non-edge regions in FIG. 24D corresponds to the layout C in FIG. 19C.

  FIG. 24E shows the corrected image of FIG. 24D. The correction unit 62 corrects the tentative determination of (D) of FIG. 24 with reference to the storage unit 63. The provisional determination of (D) of FIG. 24 corresponds to the provisional determination column of the data D3 shown in FIG. 21, and the areas of “2A”, “2C”, “3C”, and “3D” are changed from the background area to the original area. Will be corrected.

  The flowchart of the image processing apparatus 1 of FIG. 18 is the same as the flowchart shown in FIG. 16, but the processes of steps S4 and S5 are different. The image processing apparatus 1 of FIG. 18 makes a provisional determination of the non-edge region and determines at which position of the image data the non-edge region is located in step S4 of FIG. Then, in step S5, the image processing apparatus 1 refers to the storage unit 63 and corrects the temporary determination.

  In this way, the storage unit 63 stores the arrangement combination of the non-edge areas provisionally determined by the provisional determination unit 61 and the arrangement combination of the non-edge areas after correction in association with each other. Then, the correction unit 62 refers to the storage unit 63 and corrects the tentative determination of the non-edge region. As a result, the image processing apparatus 1 can appropriately divide the image data into the original portion and the background portion.

[Third Embodiment]
In the third embodiment, the image processing apparatus converts the provisionally determined layout of the non-edge region into a simplified layout. Then, the image processing device includes a storage unit that stores the layout combination after layout conversion of the non-edge region temporarily determined by the temporary determination unit and the layout combination after correction of the non-edge region in association with each other. The temporarily determined original area and background area are corrected with reference to the copy.

  FIG. 25 is a diagram showing a functional block configuration example of the image processing apparatus 1 according to the third embodiment. 25, the same components as those in FIG. 18 are designated by the same reference numerals. Below, the part different from FIG. 18 will be described.

  As shown in FIG. 25, the image processing device 1 includes a conversion unit 81, a correction unit 82, and a storage unit 83. The conversion unit 81 converts the layout of the non-edge region provisionally determined by the provisional determination unit 61 into a simplified layout.

  FIG. 26 is a diagram for explaining layout conversion. Focusing on the arrangement of adjacent non-edge regions, the layout shown in FIG. 19 can be simplified as shown in FIG. Therefore, the conversion unit 81 converts each of the layouts A to C shown in FIG. 19 into each of the layouts a to c shown in FIG. 26 according to the conversion conditions described below.

  FIG. 27 is a diagram showing layout conversion conditions. For example, as shown in the first line of "Conversion conditions for layout A" in FIG. 27, the conditions for converting 2a of layout a in FIG. 26 into NA are "2A" to "2D" in layout A in FIG. This is the case where all of the areas of “” are “NA”. Further, for example, as shown in the sixth line of “Conversion condition of layout B” in FIG. 27, the area “3a” in layout b in FIG. 26 follows the provisional determination of “3A” in layout B in FIG. 19. Further, for example, as shown in the second line of “Conversion condition of layout C” in FIG. 27, the condition for converting the area “2a” of the layout c of FIG. 26 into the original area is the layout C of FIG. In “2A” and “2B”, the number of background areas is one or less.

  Returning to the explanation of FIG. The correction unit 82 refers to the storage unit 83 and corrects the temporarily determined non-edge region.

  The storage unit 83 stores, in advance, the layout combination of the non-edge regions after the layout conversion, which is provisionally determined by the provisional determination unit 61, and the corrected layout combination of the non-edge regions in association with each other. The storage unit 83 also stores the same information as the storage unit 35 described with reference to FIG.

  FIG. 28 is a first part of a diagram showing a data configuration example of the storage unit 83. As shown in FIG. 28, the storage unit 83 has a column for provisional determination and a column after correction. Each of the tentative determination column and the corrected column has a column of the layout area shown in FIG.

  The provisional determination field stores the layout combination after layout conversion of the document area and the background area of the non-edge area that are provisionally determined by the provisional determination unit 61. The post-correction column stores the layout combination of the post-correction non-edge region of the original area and the background area, which corresponds to the provisional determination field. The data D12 and D21 shown in FIG. 28 are corrected once, and then a part (3a ', 3b') thereof is further corrected by the data D31. Further, the data D13 and D22 are corrected once, and then a part (3a ', 3b') thereof is further corrected by the data D32.

  Further, the storage unit 83 further stores in advance information for returning the layout shown in FIG. 26 to the layout shown in FIG.

  FIG. 29 is the second part of the diagram showing the data configuration example of the storage unit 83. As shown in FIG. 29, in the storage unit 83, the area of the simplified layout and the area of the layout before the simplification are stored in association with each other.

  The operation of the image processing apparatus 1 of FIG. 25 will be described below.

  FIG. 30 is a first diagram illustrating an example of the operation of the image processing apparatus 1. 30A to 30D are the same as the tentative determination and layout determination of the non-edge region of FIGS. 22A to 22D, and the description thereof will be omitted.

  In the conversion unit 81, the provisional determination unit 61 determines that the area of “3A” is the original area, the area of “2I” is the background area, and the area of “3B” is the background area. 3B: background "or the like), the layout conversion is performed based on the layout conversion condition shown in FIG. In the case of the example of FIG. 30D, “3A: original”, “2I: background”, and “3B: background” are converted into the layout b of FIG. 2e: background ”,“ 3b: background ”.

  When the layout conversion is performed by the conversion unit 81, the correction unit 82 refers to the storage unit 83 in FIG. 28 and corrects the temporary determination. In the above example, the correction unit 82 corrects “3a: original”, “2e: original”, “3b: background” based on the data D12, D21, D31 of FIG. Then, the conversion unit 81 refers to the storage unit 83 in FIG. 29 and restores the original layout (layout B in FIG. 19).

  As a result, the temporary determinations “3A: original”, “2I: background”, and “3B: background” shown in FIG. 30D are “3A: original”, as shown in FIG. The original is corrected to “2I: original” and “3B: background”. That is, the non-edge area L11 is corrected to the original, the non-edge area L12 is corrected to the original, and the non-edge area L13 is corrected to the background.

  FIG. 31 is the second part of the diagram illustrating the operation example of the image processing apparatus 1. 31A to 31D are the same as the tentative determination and the layout determination of the non-edge region in FIGS. 23A to 23D, and the description thereof will be omitted.

  When the tentative determination unit 61 determines that the area “2A” is the background area and the area “4” is the original area, the conversion unit 81 performs layout conversion based on the layout conversion condition shown in FIG. 27. I do. In the case of the example of FIG. 31D, “2A: background” and “4: original” are converted into the layout a of FIG. 26A and converted into “2a: original” and “4: original”. To be done.

  When the layout conversion is performed by the conversion unit 81, the correction unit 82 refers to the storage unit 83 in FIG. 28 and corrects the temporary determination. In the above example, the correction unit 82 corrects “2a: original” and “4: original” based on the data D11 of FIG. Then, the conversion unit 81 refers to the storage unit 83 in FIG. 29 and restores the original layout (layout A in FIG. 19).

  As a result, the provisional determinations “2A: background” and “4 originals” of the non-edge area shown in FIG. 31D are “2A: original” and “4: original” as shown in FIG. Will be corrected. That is, the non-edge areas L21 and L22 are corrected to the original.

  FIG. 32 is the third part of the diagram illustrating the operation example of the image processing apparatus 1. 32A to 32D are the same as the tentative determination and the layout determination of the non-edge regions of FIGS. 24A to 24D, and the description thereof will be omitted.

  In the conversion unit 81, the provisional determination unit 61 determines that the area of “2A” is the background area, the area of “2C” is the background area, the area of “3C” is the background area, and the “3D”. If it is determined that the area is a background area, layout conversion is performed based on the layout conversion conditions shown in FIG. In the case of the example of FIG. 32D, “2A: background”, “2C: background”, “3C: background”, and “3D: background” are converted into the layout c of FIG. 2a: original ”,“ 2c: original ”,“ 3a: background ”,“ 3b: background ”.

  When the layout conversion is performed by the conversion unit 81, the correction unit 82 refers to the storage unit 83 in FIG. 28 and corrects the temporary determination. In the above example, the correction unit 82 corrects “2a: original”, “2c: original”, “3a: original”, “3b: original” based on the data D13, D22, D32 of FIG. Then, the conversion unit 81 refers to the storage unit 83 in FIG. 29 and restores the original layout (layout C in FIG. 19).

  As a result, the temporary determinations “2A: background”, “2C: background”, “3C: background”, and “3D: background” of the non-edge region shown in (D) of FIG. 32 are shown in (E) of FIG. As described above, “2A: original”, “2A: original”, “3A: original”, and “3A: original” are corrected. That is, the non-edge regions L31 to L34 are corrected as the original.

  FIG. 33 is a flowchart showing an operation example of the image processing apparatus 1. 33, the same processes as those in FIG. 16 are designated by the same reference numerals. Hereinafter, processing different from that in FIG. 16 will be described.

  In step S11, the provisional determination unit 61 determines whether the non-edge region detected in step S3 is a document region or a background region based on the first provisional determination condition and the second provisional determination condition. At the same time as the tentative determination, the position of the image data where the tentatively determined non-edge region is located is determined.

  Next, the conversion unit 81 performs layout conversion of the non-edge region temporarily determined in step S11 (step S12).

  Next, the correction unit 82 refers to the storage unit 83, corrects the provisionally determined non-edge region whose layout has been converted in step S12, and restores the original layout (step S13).

  In this way, the conversion unit 81 converts the provisionally determined layout of the non-edge region into a simplified layout. The storage unit 83 stores the layout combination after layout conversion of the non-edge area temporarily determined by the temporary determination unit 61 and the layout combination after correction of the non-edge area in association with each other. Then, the correction unit 82 refers to the storage unit 83 and corrects the temporary determination of the non-edge region. As a result, the image processing apparatus 1 can reduce the storage capacity of the storage unit 83.

  Although the present invention has been described above using the embodiments, the functional configurations of the image processing apparatus 1 and the information processing apparatus 50 are classified according to the main processing content in order to facilitate understanding of the configurations. .. The present invention is not limited by the classification method and names of the constituent elements. The configurations of the image processing device 1 and the information processing device 50 can be classified into more components according to the processing content. Also, one component can be classified so as to perform more processing. The processing of each component may be executed by one piece of hardware or a plurality of pieces of hardware.

  Further, the technical scope of the present invention is not limited to the scope described in the above embodiment. It is apparent to those skilled in the art that various modifications and improvements can be added to the above-described embodiment. For example, the embodiments may be combined. It is also apparent from the scope of the claims that the embodiments added with such changes or improvements can be included in the technical scope of the present invention. The present invention can also be provided as a program that realizes the functions of the image processing apparatus 1 and the information processing apparatus 50, and a storage medium that stores the program.

  DESCRIPTION OF SYMBOLS 1 ... Image processing device, 2 ... Stand, 3 ... Cover, 4 ... Input device, 11 ... Scan range, 12 ... Background, 13 ... Original, 14 ... Ruled line, 15 ... Character, 31 ... Scan section, 32 ... Control section, 33 ... Communication unit, 34 ... Input unit, 35 ... Storage unit, 41 ... Edge extraction unit, 42 ... Non-edge region detection unit, 43 ... Temporary determination unit, 44 ... Correction unit, 45 ... Cutout unit, 50 ... Information processing device , 51 ... I / F section, 52 ... Storage section, 61 ... Temporary determination section, 62 ... Correction section, 63 ... Storage section, 71a ... Scan range, 72a ... Background, 73a ... Original, 74a ... Ruled line, 75a ... Character, 81 ... conversion unit, 82 ... correction unit, 83 ... storage unit.

Claims (10)

An acquisition unit that acquires image data generated by scanning an object to be scanned ,
An edge extraction unit that extracts an edge of the image data,
A non-edge region detection unit that detects a non-edge region of the image data based on the edge;
The non-edge region is an image region of the scan target object or an image region of the background, a provisional determination unit for provisionally determining based on a predetermined condition,
A correction unit that corrects the provisional determination of the non-edge region based on the provisionally determined placement of the non-edge region;
An image processing apparatus comprising: The image processing apparatus according to claim 1, wherein
The predetermined conditions are the following two conditions,
The first condition is that the non-edge region is in contact with data edges of two or more sides of the image data,
The second condition is the fact thus edge non-edge region is not uncommon circumference,
An image processing device characterized by. The image processing apparatus according to claim 2, wherein
The provisional determination unit,
When the non-edge region does not satisfy at least one of the first condition and the second condition, the non-edge region is provisionally determined as an image region of the scan target,
If the non-edge region satisfies both the first condition and the second condition, the non-edge region is tentatively determined as the image region of the background,
An image processing device characterized by the above. The image processing apparatus according to any one of claims 1 to 3,
In the correction unit, the first non-edge area and the second non-edge area that are tentatively determined to be the image area of the background are adjacent to each other, and the first non-edge area is tentatively the image area of the scan target. If the second non-edge region is adjacent to the determined non-edge region and is not in contact with the non-edge region tentatively determined to be the image region of the scan target, the first non-edge region is temporarily Correct the determination to the image area of the scan target,
Image processing apparatus characterized by. The image processing apparatus according to any one of claims 1 to 3,
A storage unit that stores the combination of the arrangement of the non-edge regions provisionally determined by the provisional determination unit and the arrangement combination of the corrected non-edge regions in association with each other;
The correction unit corrects the temporary determination of the non-edge region with reference to the storage unit,
An image processing device characterized by the above. The image processing apparatus according to any one of claims 1 to 3,
A conversion unit for converting the provisionally determined layout of the non-edge region into a simplified layout;
A storage unit that stores the layout combination after layout conversion of the non-edge region provisionally determined by the provisional determination unit, and the alignment combination of the corrected non-edge region in association with each other;
The correction unit corrects the temporary determination of the non-edge region with reference to the storage unit,
An image processing device characterized by the above. The image processing apparatus according to any one of claims 1 to 6,
The edge extraction unit calculates a luminance dispersion difference between a luminance dispersion of a first pixel of the image data and a luminance dispersion of a second pixel around the first pixel, and based on the luminance dispersion difference, , Extracting edges of the image data,
An image processing device characterized by the above.   An acquisition unit that acquires image data generated by scanning an object to be scanned,
  An edge extraction unit that extracts an edge of the image data,
  For the image data, a determination unit that determines the image area of the scan target and the image area of the background,
  Have
  The determination unit determines whether the non-edge region of the image data is an image region of the scan target object or a background image region based on a predetermined condition including a condition for the arrangement,
  An image processing device characterized by the above. An acquisition step of acquiring image data generated by scanning the scan target object;
An extraction step of extracting an edge of the image data,
A detection step of detecting a non-edge region of the image data based on the edge;
The non-edge region is an image region of the scan target object or a background image region, a provisional determination step of provisionally determining based on a predetermined condition,
A correction step of correcting the provisional determination of the non-edge region based on the provisionally determined arrangement of the non-edge region;
An image processing method comprising:   An acquisition step of acquiring image data generated by scanning the scan target object;
  An extraction step of extracting an edge of the image data,
  For the image data, a determination step of determining the image area of the scan target and the image area of the background,
  Have
  The determination step determines whether the non-edge region of the image data is an image region of the scan target object or a background image region based on a predetermined condition including a condition for the arrangement,
  An image processing method characterized by the above.

Images