Image reading device and program

Abstract

To provide an image processing device that makes it easier to utilize optically read image data.SOLUTION: An image processing device includes an outer edge specification unit that specifies the outer edge of an original document with respect to image data optically read from the original document, and a filling unit that fills in the image data outside the outer edge specified by the outer edge specification unit with a predetermined color. Preferably, the outer edge specification unit includes an end point specification unit that specifies positions of end points that constitute the original document end on the basis of a gradation change in the image data, and an outer edge determination unit that determines the outer edge of the original document by connecting a plurality of end points specified by the end point specification unit.SELECTED DRAWING: Figure 5

Description

[Technical field]

[0001] The present invention relates to an image processing device, an image processing method, and a program. [Background technology]

[0002] For example, Patent Document 1 discloses an image processing device comprising a scanning unit that scans an object to be scanned and outputs image data, an edge extraction unit that extracts edges of the image data, a non-edge area detection unit that detects non-edge areas of the image data based on the edges, a provisional determination unit that provisionally determines whether the non-edge area is an image area of ​​the object to be scanned or an image area of ​​the background based on predetermined conditions, and a correction unit that corrects the provisional determination of the non-edge area based on the arrangement of the provisionally determined non-edge area.

[0003] Patent Document 2 discloses an inspection device that includes an image reading unit that scans paper and a background member that is positioned opposite the image reading unit across the paper and can switch between a number of background colors, and that further includes a color discrimination unit that discriminates the color of the corners of the paper, a background color selection unit that selects, from the multiple background colors, a background color that has a relatively high contrast with the color of the corners of the paper, and a corner fold detection unit that detects folds in the corners of the paper based on the results of the image reading unit scanning the paper using the background member of the selected background color.

[0004] Patent Document 3 discloses an image editing system comprising an image reading means for decomposing an image of a form on which characters or figures are written in a specific pattern color on a predetermined background color into pixels and reading the image together with a background having a background color different from the background color, a defect search means for comparing the shape and dimensions of the image of the form read by the image reading means with preregistered form format information to search for whether there are any missing parts in the periphery of the image, a defect correction means for correcting any missing parts in the image by filling in the missing parts with the background color of the form, and a pattern adding means for adding a predetermined additional pattern of characters or figures to a predetermined position of the image of the form that has been read normally and the image of the form that has been corrected by the defect correction means.

[0005] Furthermore, Patent Document 4 discloses an image reading device comprising a first light source which irradiates light onto one side of a document, a second light source which irradiates light onto the other side of the document, reading means which receives light irradiated from the first light source and reflected by one side of the document to read a scanned image of the document and receives light irradiated from the second light source to read a transmitted image of the document, and control means which detects the state of the document based on the transmitted image. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Patent Publication No. 2017-163411 [Patent Document 2] Patent Publication No. 2020-057902 [Patent Document 3] Patent Publication No. 2001-144946 [Patent Document 4] Patent No. 4933402 Summary of the Invention [Problem to be solved by the invention]

[0007] An object of the present invention is to provide an image processing device that makes it easier to utilize optically read image data. [Means for solving the problem]

[0008] The image processing device of the present invention has an outer edge identification unit that identifies the outer edge of a document based on image data optically read from the document, and a filling unit that fills in the image data outside the outer edge identified by the outer edge identification unit with a predetermined color.

[0009] Preferably, the filled-in portion is filled outside the outer edge with a color different from the background color of the document used when the image data was optically read.

[0010] Preferably, the filling portion fills the outside of the outer edge with a color associated with software that passes on the image data.

[0011] Preferably, the outer edge identification unit includes an end point identification unit that identifies the positions of end points that constitute the document edge based on gradation changes in the image data, and an outer edge determination unit that determines the outer edge of the document by connecting the multiple end points identified by the end point identification unit.

[0012] Preferably, the outer edge identification unit further includes a noise correction unit that corrects the position of the end point based on other end points present in the vicinity, and the outer edge determination unit determines the outer edge of the document based on the positions of the end points corrected by the noise correction unit.

[0013] Preferably, the noise correction section performs the correction process depending on which side of the document the noise corresponds to.

[0014] In addition, the image processing method of the present invention includes an outer edge identification step for identifying the outer edge of a document with respect to image data optically read from the document, and a filling step for filling in the image data outside the outer edge identified by the outer edge identification step with a predetermined color.

[0015] In addition, the program of the present invention causes a computer to execute an outer edge identification step of identifying the outer edge of a document for image data optically read from the document, and a filling step of filling in the image data outside the outer edge identified by the outer edge identification step with a predetermined color. Effect of the Invention

[0016] It is possible to provide an image processing device that makes it easier to use optically read image data. [Brief description of the drawings]

[0017] [Figure 1] 1 is a diagram illustrating an example of the overall configuration of an image processing system 1. [Diagram 2] 11A and 11B are diagrams illustrating the relationship between cropping of a scanned image and document edges. [Diagram 3] FIG. 1 is a diagram illustrating an overview of image processing. [Figure 4] 2 is a diagram illustrating an example of a hardware configuration of an image processing device 2. FIG. [Diagram 5] 2 is a diagram illustrating an example of a functional configuration of an image processing device 2. FIG. [Figure 6] FIG. 2 is a diagram for explaining the outer edge specifying section 300 in more detail. [Figure 7] 1 is a flowchart illustrating the overall operation (S10) of the image processing device 2. [Figure 8] 13 is a diagram illustrating an example of a table that associates file output destinations with background colors. [Figure 9] 13 is a diagram for explaining the processing of an end point identification unit 302. FIG. [Figure 10] 10 is a diagram illustrating an example of noise at an end point identified by the end point identification unit 302. FIG. [Figure 11] 13 is a diagram for explaining the processing of the filling unit 320. FIG. [Figure 12] 10A and 10B are diagrams illustrating the detection result of the document edge and explaining the problem thereof; [Figure 13] FIG. 1 is a diagram for explaining an outline of a discriminant analysis method. [Figure 14]13A to 13C are diagrams illustrating variations of background areas filled by a filling unit 320. [Figure 15] 11 is a flowchart illustrating noise correction processing (S200) in a straight line region. [Figure 16] FIG. 13 is a diagram illustrating a local angle formed by three adjacent endpoints. [Figure 17] 13 is a flowchart illustrating a noise correction process (S240) in a non-linear region. [Figure 18] FIG. 13 is a diagram illustrating a local exception. [Figure 19] FIG. 13 is a diagram illustrating a shine edge and a shadow edge. [Figure 20] FIG. 11 compares a scanned image scanned with a white backing and an image filled with a black background color. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0018] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a diagram illustrating an example of the overall configuration of an image processing system 1. As shown in FIG. As illustrated in FIG. 1, an image processing system 1 includes an image processing device 2 and a scanner device 4, which are connected to each other via a cable 7. The image processing device 2 is, for example, a computer terminal, and performs image processing on the image data received from the scanner device 4. The scanner device 4 is an image reading device that optically reads image data from an original (image display medium) and transmits the read image data to the image processing device 2 via a cable 7, for example. The cable 7 is, for example, a USB cable. In this example, a specific example is described in which image data is transmitted from the scanner device 4 to the image processing device 2 via a cable 7, but this is not limited to this. For example, image data may be transmitted from the scanner device 4 to the image processing device 2 via wireless communication, or the image processing device 2 may be built into the scanner device 4.

[0019] In the above configuration, a typical hardware configuration of the scanner device 4 includes a "backing" that is placed underneath the passing paper (original) when an image is captured. This backing often differs depending on the product, as there are advantages and disadvantages depending on the color of the backing (background color). Of course, if you use the document cropping function, these differences are less noticeable (in a neat rectangle, the backing part is not visible), but cropping is not always assumed. In this case, the "backing part" in the scanned image looks like Figure 2(A). In other words, the presence of the backing is not necessarily invisible to the user, and there are cases where the "visibility of the backing" is directly related to work.

[0020] Furthermore, when reading white paper on a white backing, there is generally little difference in tone between the original and the backing, and factors such as shadows and shine caused by the illumination of the light source mean that it is often difficult to accurately determine each and every edge point of the original (which is often searched for based on tone differences). Therefore, as shown in FIG. 2B, CROPP assumes a "rectangle" and achieves a pseudo "high-precision rectangle detection" by creating approximate lines and rectangles. However, the functionality required of the scanner device 4 is not necessarily limited to being able to simply extract rectangular shapes, but there is also a demand for extraction that accurately follows the edges of a non-rectangular document (removal of the backing portion), as well as detection of broken or chipped documents. These functions are realized in the limited environment of a white paper document on a black backing, but are not realized for a white paper document on a white backing.

[0021] Therefore, in the image processing system 1 of this embodiment, the image processing device 2 cuts out only the document area from the image data read by the scanner device 4, creates a background image (black image) of the same size as the image read by the scanner device 4 (scanned image), and combines these two images, as shown in Fig. 3. This allows the image processing device 2 to generate a scanned image in which the background area is filled in black, even if the scanner device 4 performs scanning with a white backing.

[0022] FIG. 4 is a diagram illustrating an example of a hardware configuration of the image processing device 2. As shown in FIG. As illustrated in FIG. 4, the image processing device 2 includes a CPU 200, a memory 202, a HDD 204, a network interface 206 (network IF 206), a display device 208, and an input device 210, which are connected to each other via a bus 212. The CPU 200 is, for example, a central processing unit. The memory 202 is, for example, a volatile memory, and functions as a main storage device. The HDD 204 is, for example, a hard disk drive device, and serves as a non-volatile storage device for storing computer programs (for example, the image processing program 3 in FIG. 5) and other data files. The network IF 206 is an interface for wired or wireless communication, and realizes communication with the scanner device 4, for example. The display device 208 is, for example, a liquid crystal display. The input device 210 is, for example, a keyboard and a mouse.

[0023] FIG. 5 is a diagram illustrating an example of the functional configuration of the image processing device 2. 5, an image processing program 3 is installed and operates in the image processing device 2 of this example. The image processing program 3 is stored in a recording medium such as a CD-ROM, and is installed in the image processing device 2 via the recording medium. The image processing program 3 includes an outer edge specifying unit 300, a background color selecting unit 310, a filling unit 320, and a file output unit 330. Note that a part or the whole of the image processing program 3 may be realized by hardware such as an ASIC, or may be realized by borrowing some of the functions of an OS (Operating System).

[0024] In the image processing program 3, the outer edge specifying unit 300 specifies the outer edge of the document from the image data optically read from the document by the scanner device 4. More specifically, as shown in Fig. 6, the outer edge identification unit 300 includes an end point identification unit 302, a noise correction unit 304, and an outer edge determination unit 306. The end point identification unit 302 identifies the positions of end points (document edge points) that constitute the document edges based on tone changes in the image data. The noise correction unit 304 corrects the positions of the document end points identified by the end point identification unit 302 based on other end points that exist in the vicinity. The outer edge determination unit 306 connects the multiple end points identified by the end point identification unit 302 (or the end points corrected by the noise correction unit 304) to determine the outer edge of the document.

[0025] The background color selection unit 310 selects a background color to be used by the filling unit 320. For example, the background color selection unit 310 selects a color different from the backing of the scanner device 4 as the background color to be used by the filling unit 320. The background color selection unit 310 may also select a color associated with the software that transfers the image data as the background color to be used by the filling unit 320. The background color selection unit 310 in this example selects the background color to be used by the filling unit 320 depending on the output destination to which the image data is output by the file output unit 330.

[0026] The filling unit 320 fills in the image data read by the scanner device 4 outside the outer edge specified by the outer edge specification unit 300 with a predetermined color. For example, the filling unit 320 fills in the image data read by the scanner device 4 outside the outer edge specified by the outer edge specification unit 300 with the color selected by the background color selection unit 310. The filling unit 320 in this example cuts out only the document area from the scanned image according to the outer edge of the document specified by the outer edge specification unit 300, creates a background image (an image filled in the color selected by the background color selection unit 310) of the same size as the scanned image, and superimposes the image of the cut-out document area on the background image.

[0027] The file output unit 330 outputs, to a predetermined output destination, the image data in which the outer side of the outer edge is filled with a predetermined color by the filling unit 320. For example, the file output unit 330 outputs the image data in which the outer side of the outer edge is filled with a predetermined color by the filling unit 320 to other image processing software (image processing application) in the image processing device 2, an external Web service, or the like.

[0028] FIG. 7 is a flowchart illustrating the overall operation (S10) of the image processing device 2. As illustrated in FIG. 7, in step 100 (S100), the image processing program 3 (FIG. 5) of the image processing device 2 acquires image data (scanned image) read by the scanner device 4. In step 105 (S105), the end point identification unit 302 (FIG. 6) of the image processing program 3 searches for points outside the acquired scanned image where the gradation change or color change is greater than or equal to a reference value, and identifies the end points of the document.

[0029] In step 20 (S20), the noise correction unit 304 (FIG. 6) corrects the positions of the end points identified by the end point identification unit 302. Details will be described later with reference to FIGS.

[0030] In step 115 (S115), the outer edge determining unit 306 connects the end points identified by the end point identifying unit 302 or the end points corrected by the noise correcting unit 304 to determine the outer edge of the document in the scanned image. In step 120 (S120), the background color selection unit 310 (FIG. 5) refers to the table illustrated in FIG. 8 and selects a background color associated with the output destination of the image data by the file output unit 330.

[0031] In step 125 (S125), the filling unit 320 cuts out an image of the document area from the scanned image in accordance with the outer edge of the document determined by the outer edge determination unit 306, and superimposes the image of the cut-out document area on a background image of the background color selected by the background color selection unit 310, thereby generating an image in which the outside of the document is filled in with the default background color. In step 130 (S130), the file output unit 330 outputs the image data in which the outside of the document has been filled by the filling unit 320 to a default output destination.

[0032] FIG. 9 is a diagram for explaining the processing of the endpoint identification unit 302. As shown in FIG. As shown in Fig. 9(A), the end point identification unit 302 detects the boundary position (multiple end points) between the document and the scanner background (backing) based on the amount of change in gradation. Furthermore, as shown in Fig. 9(B), the end point identification unit 302 approximates the document end points (sampling points of the document edge) with a group of likely straight lines using a Hough transform or the like, and calculates and determines the document rectangle (four straight lines). The end point identification unit 302 also cuts out the image using the determined four straight lines, and performs skew correction as shown in Fig. 9(C). At this time, a margin of a determined size may be left outside the cut-out image.

[0033] FIG. 10 is a diagram illustrating noise at the end points identified by the end point identification unit 302. As shown in FIG. As shown in FIG. 10, the end points (sampling points of the document edge) identified by the end point identification unit 302 contain noise samples due to adverse effects on boundary detection accuracy, etc. Since these noises become a cause of image quality degradation during the filling process by the filling unit 320, a process for removing these noises is necessary. Specifically, the noise correction unit 304 treats as noise, and targets for correction, end points that are a certain distance or more away from the straight line (dotted line) in FIG. 10 and have adjacent end points near the straight line.

[0034] FIG. 11 is a diagram for explaining the processing of the filling unit 320. In FIG. As shown in FIG. 11A, the outer edge determination unit 306 connects the end points identified by the end point identification unit 302 or the end points corrected by the noise correction unit 304 with straight lines to determine the outer edge (periphery) of the document. As shown in Fig. 11(B), the filling unit 320 fills the outside of the outer edge determined by the outer edge determination unit 306 with a default color (the color selected by the background color selection unit 310). Furthermore, the filling unit 320 cuts the document area into a rectangle (cuts the outer margins) to generate the output image data shown in Fig. 11(C). As shown in Fig. 11(C), even if the backing of the scanner device 4 is white, the output image data is an image in which the background color is repainted to black and the color of damaged parts such as tears is repainted.

[0035] FIG. 12 is a diagram illustrating the detection result of the document edge and explaining the problem associated therewith. During scanning by the scanner device 4, the document physically shakes as it is transported, and the light source and reflection also cause blurring, so edge detection (end point identification) by the end point identification unit 302 cannot necessarily be performed with high accuracy and fluctuates within a certain range. This phenomenon is particularly noticeable when the backing is white. Note that when the backing is black and the document paper is white, or vice versa, the gradation difference is very large, so accurate edge detection is not expected. For example, edge points (end points) of an actual document often contain blur as shown in Fig. 12(A). Here, we would like to scan the outline of the document using these edge points (end points), but if the edge points are simply connected as in Fig. 12(B), the document edge will fluctuate slightly, and even if the actual chipping can be reflected, the original document shape will no longer be linear, and the outline will tend to be more distorted than the actual object. Furthermore, when cropping is done simply according to an approximate straight line, as in the case of general cropping, it naturally follows the shape of the original, as shown in FIG. 12(C), but it tends to ignore actual defects. In this embodiment, a contour line having a shape as shown in FIG. 12(D) is required, which has the advantages of both FIGS. 12(B) and (C).

[0036] The outer edge specifying unit 300 of this example overlaps FIG. 12(B) and FIG. 12(C) to search for an area where there is a deviation of "a certain amount or more" (the deviation area of ​​FIG. 12(E)). The outer edge specifying unit 300 adopts FIG. 12(B) for the deviation area and adopts FIG. 12(C) for the other areas, thereby determining the outer edge of FIG. 12(F) which is close to FIG. 12(D). In addition, the threshold value for "whether there is a deviation of a certain amount or more" may be, for example, a fixed value (e.g., 2 mm), or may be dynamically calculated by a discriminant analysis method by tallying up all the deviation values ​​in FIG. 12(B) and FIG. 12(C). The dynamic discriminant analysis method is performed by, for example, setting a threshold value based on the occurrence frequency of the deviation values ​​shown in FIG. 13 and comparing with the threshold value.

[0037] FIG. 14 is a diagram illustrating variations of the background area filled by the filling unit 320. In FIG. In this example, the filling unit 320 fills the background area with black when the backing of the scanner device 4 is white, but is not limited to this, and for example, as variation 1, the part corresponding to the backing may be dynamically switched according to the background color of the document. Specifically, the filling unit 320 fills with a color far from the background color (black background for white paper, white background for black paper) or a color close to the background color (white background for white paper, red background for red paper). Moreover, as variation 2, the filled-in portion 320 may be filled in with an extremely distinctive color such as red, or as variation 3, a background image having a pattern such as dots may be used. Furthermore, the filling section 320 may display the outline of the document as a fourth variation to assist the user in visual confirmation, or may have a transparency (alpha channel, etc.) attribute as a fifth variation.

[0038] Next, the noise correction process (S20) in Fig. 7 will be described in more detail. The noise correction process (S20) includes noise correction process (S200) in a linear region and noise correction process (S240) in a non-linear region (such as a broken portion). First, the noise correction process in the straight line region (S200) will be described. Fig. 15 is a flowchart illustrating the noise correction process in the straight line region (S200). FIG. 16 is a diagram illustrating a local angle formed by three adjacent end points. As illustrated in FIG. 15, the noise correction unit 304 (FIG. 6) collects edge points (end points identified by the end point identification unit 302) in straight line regions (top, bottom, left, and right sides) (S202), and calculates a straight line equation for each side based on the collected edge points (end points) using a Hough transform or the least squares method (S204). Next, the noise correction unit 304 calculates the distance from the line for each edge point (S206), and determines whether the calculated distance is greater than a reference value (S208). If the calculated distance is less than or equal to the reference value (S208: No), the noise correction unit 304 determines that edge point (end point) to be a stable edge (S210), and if the calculated distance is greater than the reference value (S208: Yes), the noise correction unit 304 determines that edge point (end point) to be a noise candidate (S212).

[0039] As illustrated in FIG. 16, the noise correction unit 304 calculates the local angle between the edge point of the noise candidate and the edge points on both sides of the edge point (S214), and determines whether the calculated local angle is smaller than a reference angle (S216). If the calculated local angle is smaller than the reference angle (S216: Yes), the noise correction unit 304 designates this edge point of the noise candidate as a noise edge (S218); if the calculated local angle is equal to or larger than the reference angle (S216: No), the noise correction unit 304 removes it from the noise candidate and moves on to processing the next edge point. The noise correction unit 304 replaces the coordinate values ​​of the edge points determined to be noise edges with the coordinate values ​​of points on a straight line (vertical intersections) (S220). Note that in this example, the coordinate values ​​are replaced with the coordinate values ​​of points on a straight line, but the edge points determined to be noise edges may simply be deleted. The noise correction unit 304 performs the above process for each edge point.

[0040] Next, the noise correction process (S240) in non-linear areas (torn areas, etc.) will be described. Non-linear areas are areas that correspond to corners of a document or bends or tears in the middle of a side, and since jaggies tend to occur in these areas, smoothing of coordinate values ​​is performed as a correction process. FIG. 17 is a flowchart illustrating the noise correction process (S240) in the non-linear region. FIG. 18 is a diagram illustrating a localized gouge, and FIG. 19 is a diagram for explaining a shine edge and a shadow edge. As illustrated in FIG. 17, the noise correction unit 304 (FIG. 6) performs a loop process for each side of the document (S242). The noise correction unit 304 determines whether each side is a shiny side (S244), and if it is determined that the side is not a shiny side (S244: No), the noise correction unit 304 proceeds to the process of the remaining sides (S246), and if it is determined that the side is a shiny side (S244: Yes), the noise correction unit 304 performs the process of S248 and subsequent steps. Here, the shiny side is a side where the shadow of the document edge is not visible from the optical sensor of the scanner device 4 as shown in FIG. 19, and the shadow side is a side where the shadow of the document edge is visible from the optical sensor of the scanner device 4. Whether the side of the document is a shiny side or a shadow side is determined by the positional relationship between the light source and the optical sensor of the scanner device 4 and the document. In other words, which of the four sides of the document is a shiny side is set in advance for each model of the scanner device 4. It is also possible to determine whether the side is a shiny side based on the degree of variation of the end points (edge ​​points) in the straight line region.

[0041] When the processing target is a shiny side, the noise correction unit 304 extracts a group of samples away from a straight line (S248). The group of samples is a group of multiple consecutive edge points that are located close to each other and are away from a straight line. The noise correction unit 304 performs a loop process for each extracted sample group (S250). That is, if there are no unprocessed sample groups (S252: Yes), the noise correction unit 304 exits the loop process, and if there are unprocessed sample groups (S252: No), the noise correction unit 304 performs the processes from S254 onwards.

[0042] The noise correction unit 304 determines a start point and an end point from among the samples to be processed (S254). The noise correction unit 304 determines whether or not a set of the start point and the end point exists (S256), and if a set of the start point and the end point does not exist, proceeds to processing of S250, and if a set of the start point and the end point exists, proceeds to processing of S258.

[0043] The noise correction unit 304 performs loop processing on midpoints, which are end points between the start point and the end point, so as to smooth the coordinate values ​​between the start point and the end point (S258). That is, the noise correction unit 304 judges whether each midpoint is out of alignment with the inside of the document (S260), and rewrites the coordinates of the midpoint that is out of alignment with the inside of the document to the coordinates on the line connecting the start point and the end point, as shown in FIG. 18 (S262). In this example, out of alignment means that an end point (edge ​​point) between the start point and the end point is inward of the document by a reference angle or more than the line connecting the start point and the end point. The out of alignment is judged by, for example, whether the angle of the line segment formed by the three points of the start point, the midpoint, and the end point is equal to or less than a reference angle. The noise correction unit 304 may be configured to add other sides (other sides on the side where the shadow is not visible) as shiny sides based on the degree of the skew angle when calculating the straight lines of the four-sided rectangle (when the skew angle is large).The noise correction unit 304 may also add partial edges on the side where the shadow is not visible as shiny areas based on the degree of tearing at the tearing portion.

[0044] As described above, according to the image processing device 2 of this embodiment, image data of an appropriate background color can be obtained without being restricted by the color of the backing of the scanner device 4. For example, the extraction accuracy of the white document edge in the white backing is improved. Therefore, the outline of the document can be accurately extracted as a non-rectangular outline, and chips and tears in the document can be faithfully reproduced. Also, colors outside the original can be freely replaced according to the purpose (application, etc.), making it possible to process images to suit the characteristics of downstream applications such as authenticity determination. For example, as shown in Fig. 20, as can be seen by comparing the scanned image (left side of Fig. 20) from the scanner device 4 with the image filled with a black background (right side of Fig. 20), missing or folded parts of the original can be made more noticeable, making it less likely that mistakes will occur in downstream processing. Furthermore, even if the hardware of the scanner device 4 is an inexpensive configuration with only a white backing, a black backing can be realized, and cost effectiveness can be expected. In addition, since it can faithfully extract defects such as chips and tears in rectangular documents, it is possible to repair them (by filling them in). It can also extract the outline of non-rectangular documents (such as round documents). Even if there are unstable factors such as shine or tonal fluctuations caused by backing paper, the document outline can be faithfully extracted.

[0045] Although the embodiment of the present invention has been described, the above embodiment is presented as an example and is not intended to limit the scope of the invention. The above embodiment can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the gist of the invention. The above embodiment and its modifications are included in the scope of the invention and its equivalents described in the claims, as well as in the scope and gist of the invention. [Explanation of symbols]

[0046] 1. Image Processing System 2. Image Processing Device 3. Image Processing Programs 4 Scanner device

Claims

1. an image generating unit that generates an image in which the outside of the outer edge of the document is a predetermined color; The outer side of the outer edge of the document includes chipped or torn areas of the document. Image processing device.

2. The default color is black. The image processing device according to claim 1 .

3. The image generating unit generates an image in which the outside of the outer edge is the default color, using a color associated with software that transfers the image. The image processing device according to claim 1 .

4. Further comprising an outer edge specifying unit for specifying an outer edge of the document; The outer edge specifying portion is an end point identifying unit that identifies the positions of end points that constitute the document end; an outer edge determination unit that connects the multiple end points identified by the end point identification unit to determine an outer edge of a document; The end point identification unit identifies an approximation line from a plurality of end points; The outer edge specification unit specifies a line connecting the end points as an outer edge for a region in which the end points deviate from the approximation line by a threshold value or more. The image processing device according to claim 1 .

5. The image generating unit superimposes an image of the original cut out at the outer edge from a scanned image of the original and an image of the predetermined color. The image processing device according to claim 1 .

6. the outer edge specification unit further includes a noise correction unit that corrects the position of the endpoint based on another endpoint existing in the vicinity of the endpoint specified by the endpoint specification unit, The outer edge determination unit determines the outer edge of the document based on the positions of the end points corrected by the noise correction unit. The image processing device according to claim 4.

7. The noise correction unit determines whether or not to delete the endpoint based on other endpoints existing in the vicinity of the endpoint identified by the endpoint identification unit. The image processing device according to claim 6.

8. An outer edge specifying unit that specifies an outer edge of a document from image data optically read from the document; an image generating unit that generates an image in which the outside of the outer edge of the document is a predetermined color; having The outer side of the outer edge of the document includes chipped or torn areas of the document. Image reading device.

9. causing a computer to execute a step of generating an image in which the outside of the outer edge of the document is a predetermined color; The outer side of the outer edge of the document includes chipped or torn areas of the document. program.

Images