JP7197830B2 - Image processing system and image processing program - Google Patents

Description

The present invention relates to an image processing system and an image processing program for extracting marks from marked images to which marks have been added.

2. Description of the Related Art Conventionally, there is known an image processing system that extracts a mark from a marked image to which a mark has been added (see, for example, Japanese Unexamined Patent Application Publication No. 2002-100003).

Marking based on, for example, at least one of hue, lightness and saturation from a marked image in which translucent marks are applied not only over light-colored parts such as white, but also over dark-colored characters such as black is extracted, the part of the mark that is given above the dark-colored characters is not extracted, and the mark may be interrupted.

The image processing system described in Patent Document 1 extracts an image of a marked portion and an image of characters as a description portion other than the mark from a marked image as an image to which a mark is added, and extracts an image of the marked portion. and the image of the character, and the image of the mark portion is expanded and shrunk by an amount corresponding to the image of the character, that is, the width of the character. and generates a hole-filled mark image as a logical product image of the logical sum image and the expanded and shrunk mark image.

The image processing system described in Patent Document 1 can fill in gaps in the mark by expanding and contracting the image of the mark portion by an amount corresponding to the width of the character.

Further, the image processing system described in Patent Document 1 provides a logical sum image as a logical sum image of a mark portion image and a character image, and a hole filling image as a logical product image of an expanded and shrunk mark image. Since a completed mark image is generated, when the marks are close to each other in the marked image, even if the gaps between the close marks are filled as discontinuities, the close marks will eventually be separated from each other. The gap between can be properly treated as a gap.

JP 2016-039525 A

However, in the conventional image processing system described in Patent Document 1, when marks are close to each other in a marked image, if a character exists in a gap between the marks that are close to each other, this character is included in the filled-in mark image.

SUMMARY OF THE INVENTION Accordingly, an object of the present invention is to provide an image processing system and an image processing program capable of improving the accuracy of extracting a mark from a marked image to which a mark has been added.

The image processing system of the present invention extracts an image of the marked part and a non-marked part image as an image of the written part other than the mark from the marked image as the marked image based on the color. and an image extracting section, wherein the filled mark image obtained by filling the discontinuity caused by the written portion other than the mark in the color extraction mark image as the image of the mark portion extracted by the image extracting portion is the color extraction mark. a hole-filled mark image generator for generating an image and the non-marked partial image using image expansion and contraction, image logical sum, and image logical product; and the hole-filled mark image. An unnecessary area-removed mark for generating an unnecessary area-removed mark image as an image of the mark part from which the unnecessary area has been removed by removing a continuous area that does not overlap at all with the area of the color extraction mark image. and an image generator.

With this configuration, the image processing system of the present invention removes, from the hole-filled mark image, continuous areas that do not overlap with the area of the color extraction mark image. Since the mark image with the unnecessary area removed is generated, when the marks are close to each other in the marked image, if there is a description part such as a ruled line or a character in the gap between the marks that are close to each other, this description It is possible to reduce the possibility that the part is included in the unnecessary area removed mark image. Therefore, the image processing system of the present invention can improve the accuracy of extracting marks from marked images.

In the image processing system of the present invention, the image extraction unit extracts a ruled line image as an image of a ruled line portion and a character image as an image of a character portion as the non-marked portion images, and generates the filled mark image. generates a first logical sum image as an image obtained by logically summing the color extraction mark image and one of the ruled line image and the character image; generating a first dilated and shrunk mark image by dilating and shrinking the color extraction mark image by an amount corresponding to the one image, the hole-filled mark image generating unit generating the first logical sum image; generating a first filled-in mark image as an image of a logical AND with the first expanded/shrinked mark image, and the filled-in mark image generation unit generates the first filled-in mark image and the ruled line image; and a second logical sum image as an image obtained by logical sum of the character image and the other image that is not the one image, and the filled-in mark image generation unit generates the first filled-in mark image is expanded and shrunk by an amount corresponding to the other image to generate a second expanded and contracted mark image, and the hole-filled mark image generation unit generates the second logical OR image and the second A second hole-filled mark image may be generated as the hole-filled mark image as a logical product image with the expanded and contracted mark image.

With this configuration, the image processing system of the present invention distinguishes between ruled lines and characters, which have different influences on marks, and thus can improve the accuracy of extracting marks from a marked image.

The image processing program of the present invention extracts an image of the marked portion and a non-marked portion image as an image of the written portion other than the mark from the marked image as the marked image based on the color. and an image extracting section, wherein the filled mark image obtained by filling the discontinuity caused by the written portion other than the mark in the color extraction mark image as the image of the mark portion extracted by the image extracting portion is the color extraction mark. a hole-filled mark image generator for generating an image and the non-marked partial image using image expansion and contraction, image logical sum, and image logical product; and the hole-filled mark image. An unnecessary area-removed mark for generating an unnecessary area-removed mark image as an image of the mark part from which the unnecessary area has been removed by removing a continuous area that does not overlap at all with the area of the color extraction mark image. It is characterized in that the image generating section is implemented by a computer.

With this configuration, the computer that executes the image processing program of the present invention removes the continuous area that does not overlap the area of the color extraction mark image from the hole-filled mark image. Therefore, when the marks are close to each other in the marked image, there are written parts such as ruled lines and characters in the gaps between the marks that are close to each other. In this case, it is possible to reduce the possibility that this written portion is included in the unnecessary area-removed mark image. Therefore, the computer executing the image processing program of the present invention can improve the accuracy of extracting the marks from the marked image.

The image processing system and image processing program of the present invention can improve the accuracy of extracting marks from marked images to which marks have been added.

1 is a block diagram of an MFP according to an embodiment of the invention; FIG. 2 is a flowchart of the operation of the MFP shown in FIG. 1; 3 is a diagram showing an example of a marked image acquired in the operation shown in FIG. 2; FIG. 4(a) is a diagram showing a color-extracted mark image extracted based on color from the marked image shown in FIG. 3; FIG. 4(b) shows a non-marked partial image extracted from the marked image shown in FIG. 3; FIG. 5A is a diagram showing a ruled line image extracted from the non-marked partial image shown in FIG. 4B; FIG. 4B is a diagram showing a character image extracted from the non-marked partial image shown in FIG. 4B; FIG. It is a figure which shows an example of the ruled line shown to Fig.5 (a). 5 is a diagram showing a first dilated and shrunk mark image obtained by dilating and shrinking the color extraction mark image shown in FIG. 4(a); FIG. (a) A diagram showing a first logical sum image, which is a logical sum image of the color extraction mark image and the ruled line image shown in FIG. 4(a). (b) A diagram showing a first filled-in mark image in which discontinuities caused by ruled lines in the color extraction mark image shown in FIG. 4(a) are filled. FIG. 9B is a diagram showing a second expanded and contracted mark image obtained by expanding and contracting the first hole-filled mark image shown in FIG. 8B; 8A is a diagram showing a second logical sum image, which is a logical sum image of the character image and the first hole-filled mark image shown in FIG. 8B; FIG. 4(b) is a diagram showing a second filled-in mark image in which discontinuities caused by characters in the color extraction mark image shown in FIG. 4(a) are filled. FIG. 11B is a diagram showing an unnecessary area-removed mark image generated from the second hole-filled mark image shown in FIG. 10B; 3 is a flowchart of the operation of the MFP shown in FIG. 1 in an example different from the example shown in FIG. 2;

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments of the present invention will be described with reference to the drawings.

First, the configuration of an MFP (Multifunction Peripheral) as an image processing system according to one embodiment of the present invention will be described.

FIG. 1 is a block diagram of MFP 10 according to this embodiment.

As shown in FIG. 1, the MFP 10 includes an operation unit 11, which is an input device such as a button for inputting various operations, and a display device such as an LCD (Liquid Crystal Display) for displaying various information. A unit 12, a printer 13 that is a printing device that prints an image on a recording medium such as paper, a scanner 14 that is a reading device that reads an image from a document, and communication with an external facsimile device (not shown) and a public telephone line. A fax communication unit 15, which is a fax device that performs fax communication via a line, communicates with an external device via a network such as a LAN (Local Area Network) or the Internet, or directly by wire or wirelessly without a network. a communication unit 16 that is a communication device that performs the above operations, a storage unit 17 that is a non-volatile storage device such as a semiconductor memory or HDD (Hard Disk Drive) that stores various information, and a control unit 18 that controls the entire MFP 10 It has

The storage unit 17 stores an image processing program 17a for extracting marks from marked images to which marks have been added. The image processing program 17a may be installed in the MFP 10 at the manufacturing stage of the MFP 10, may be additionally installed in the MFP 10 from an external storage medium such as a USB (Universal Serial Bus) memory, or may be installed from a network. It may be additionally installed in the MFP 10 .

The control unit 18 includes, for example, a CPU (Central Processing Unit), a ROM (Read Only Memory) that stores programs and various data, and a RAM (Random Access Memory) that serves as a work area for the CPU. It has The CPU executes programs stored in the storage unit 17 or ROM. The MFP 10 constitutes a computer.

The control unit 18 is an image extracting unit that extracts an image of the marked portion and a non-marked portion image as an image of the written portion other than the mark from the marked image based on the color by executing the image processing program 17a. 18a, a filled mark image obtained by filling discontinuities caused by description portions other than the mark in the color extraction mark image as an image of the mark portion extracted by the image extracting unit 18a, and extracting the color extraction mark image and the non-mark portion; a hole-filled mark image generation unit 18b that generates a hole-filled mark image generation unit 18b based on the image using image dilation and contraction, image logical sum, and image logical product; and an unnecessary area-removed mark image generation unit 18c for generating an unnecessary area-removed mark image as an image of the mark portion from which the unnecessary area has been removed by removing a continuous area that does not overlap with the area of .

Next, operations of the MFP 10 will be described.

FIG. 2 is a flow chart of the operation of the MFP 10. As shown in FIG.

When the control unit 18 is instructed via the operation unit 11 or the communication unit 16 to execute a specific process corresponding to the mark given to the document, the control unit 18 executes the image processing program 17a to perform the operation shown in FIG. to run. This document is marked with a marker such as a highlighter pen.

As shown in FIG. 2, for example, when a document is set in the scanner 14, the control unit 18 acquires a marked image by reading the document with the scanner 14 (S101). Note that a method other than the method of reading the document by the scanner 14 may be used as the method of acquiring the marked image.

FIG. 3 is a diagram showing an example of a marked image acquired in the operation shown in FIG.

A mark-applied image 30 shown in FIG. 3 is an image in which translucent marks 33 are applied not only on light-colored parts such as white but also on dark-colored ruled lines 31 and characters 32 such as black. In the following description, it is assumed that the marked image acquired in S101 is the marked image 30 shown in FIG.

FIG. 4(a) is a diagram showing a color-extracted mark image 41 extracted from the mark-applied image 30 based on the color. FIG. 4(b) shows a non-marked partial image 42 extracted from the marked image 30. As shown in FIG.

As shown in FIG. 2, after the process of S101, the image extraction unit 18a extracts the image shown in FIG. A color extraction mark image 41 and a non-mark partial image 42 shown in FIG. 4B are extracted (S102). Here, the color extraction mark image 41 is a portion of the marks 33 added to the mark-attached image 30, excluding portions overlapping description portions other than the marks 33, such as the ruled lines 31 and the characters 32. FIG. The non-marked partial image 42 includes ruled lines 31 and characters 32 .

FIG. 5A is a diagram showing an image 43 of the ruled line 31 portion (hereinafter referred to as "ruled line image") 43 extracted from the non-marked partial image 42. FIG. FIG. 5B is a diagram showing an image 44 of the character 32 portion (hereinafter referred to as "character image") 44 extracted from the non-marked portion image 42. As shown in FIG.

As shown in FIG. 2, after the process of S102, the image extraction unit 18a extracts the ruled line image 43 shown in FIG. 5A and the character image 44 shown in FIG. 5B from the non-marked partial image 42. Extract (S103). Any known method can be adopted as a method for extracting ruled lines from an image including ruled lines. Similarly, any known method can be adopted as a method for extracting characters from an image containing characters. For example, if the maximum size of a character that may be included in the markup image 30 is assumed to be a particular size, such as 72 points, then the lines that characters drawn at this maximum size may include. Since the maximum length of is specified, the image extraction unit 18a may extract a line longer than this maximum length as the ruled line 31. FIG.

FIG. 6 is a diagram showing an example of the ruled line 31. As shown in FIG.

As shown in FIG. 2, after the process of S103, the filled mark image generating unit 18b determines the amount of expansion of the color extraction mark image 41 for filling the discontinuity of the mark 33 caused by the ruled line 31. Next, the width 31a of the ruled line 31 in the extending direction of the mark 33 indicated by the arrow 30a is obtained for each ruled line 31 from the ruled line image 43 extracted in S103 (S104). Note that the ruled line 31 does not always extend in the direction indicated by the arrow 30b orthogonal to the direction indicated by the arrow 30a. For example, as shown in FIG. 6, the ruled line 31 may extend in a direction inclined from the direction indicated by the arrow 30b. Therefore, the hole-filled mark image generation unit 18b acquires from the ruled line image 43 the width 31b in the direction orthogonal to the extending direction of the ruled line 31 and the inclination 31c of the ruled line 31 with respect to the direction indicated by the arrow 30b, and from this information, , the width 31a of the ruled line 31 in the direction indicated by the arrow 30a.

FIG. 7 is a diagram showing an image 45 obtained by dilating and shrinking the color extraction mark image 41 (hereinafter referred to as "first dilated and shrunk mark image").

As shown in FIG. 2, after the process of S104, the filled-in mark image generation unit 18b adjusts the width 31a of the ruled line 31 in the direction indicated by the arrow 30a acquired in S104 to fill the discontinuity caused by the ruled line 31. The color extraction mark image 41 is expanded in the direction indicated by the arrow 30a by the corresponding amount (S105), and the image expanded in S105 is indicated by the arrow 30a by the same amount as the expansion amount in the direction indicated by the arrow 30a in S105. By contracting in the direction, the first expanded and contracted mark image 45 shown in FIG. 7 is generated (S106). Here, when the ruled line image 43 includes a plurality of ruled lines 31, the hole-filled mark image generation unit 18b adjusts the width of all the ruled lines 31 included in the ruled line image 43 to the maximum width in the direction indicated by the arrow 30a. The color extraction mark image 41 may be expanded or shrunk in the direction indicated by the arrow 30a by a corresponding amount, or each ruled line 31 may be expanded or shrunk by an amount corresponding to the width 31a of the ruled line 31 in the direction indicated by the arrow 30a. A portion of the color extraction mark image 41 corresponding to a portion in the vicinity of the ruled line 31 may be expanded and shrunk in the direction indicated by .

FIG. 8A is a diagram showing an image 46 of the logical sum of the color extraction mark image 41 and the ruled line image 43 (hereinafter referred to as "first logical sum image") 46. FIG. FIG. 8(b) is a diagram showing an image (hereinafter referred to as a "first filled mark image") 47 in which discontinuities caused by the ruled lines 31 in the color extraction mark image 41 are filled.

When the marks 33 are close to each other in the marked image 30 , the first dilated and shrunk mark image 45 has not only a discontinuity of the marks 33 caused by the ruled line 31 , but also the marks 33 are simply close to each other in the marked image 30 . There is a possibility that the part that was doing is also filled in. Therefore, the filled-in mark image generating unit 18b removes the filled-in portions other than the discontinuities of the marks 33 caused by the ruled lines 31 from the first expanded/shrinked mark image 45, as shown in FIG. After the processing, a first logical sum image 46 shown in FIG. 8A, which is a logical sum image of the color extraction mark image 41 extracted in S102 and the ruled line image 43 extracted in S103, is generated (S107 ), the first filled mark shown in FIG. An image 47 is generated (S108).

After the process of S108, the filled-in mark image generation unit 18b uses the arrow 30a and the height of the character 32 in the direction indicated by the arrow 30b are acquired for each character 32 from the character image 44 extracted in S103 (S109).

FIG. 9 is a diagram showing an image 48 obtained by expanding and contracting the first hole-filled mark image 47 (hereinafter referred to as "second expanded and contracted mark image").

As shown in FIG. 2, after the process of S109, the filled-in mark image generation unit 18b fills the discontinuity caused by the character 32 according to the width of the character 32 in the direction indicated by the arrow 30a acquired in S109. The first hole-filled mark image 47 is expanded in the direction indicated by the arrow 30a by an amount corresponding to the height of the character 32 in the direction indicated by the arrow 30b obtained in S109, respectively. (S110), the image expanded in S110 is expanded in the direction indicated by arrow 30a by the same amount as the amount of expansion in the direction indicated by arrow 30a in S110, and by the same amount as the expansion amount in the direction indicated by arrow 30b in S110. The second dilated and shrunk mark image 48 shown in FIG. 9 is generated by shrinking by an amount in the direction indicated by the arrow 30b (S111). Here, when the character image 44 includes a plurality of characters 32, the fill-in-the-blank mark image generation unit 18b reduces the width of all the characters 32 included in the character image 44 to the maximum width in the direction indicated by the arrow 30a. In the direction indicated by the arrow 30a by an amount corresponding to the maximum height of all the characters 32 included in the character image 44 in the direction indicated by the arrow 30b, in the direction indicated by the arrow 30b by an amount corresponding to the first The filled mark image 47 may be expanded and contracted, and each character 32 may be expanded and contracted in the direction indicated by the arrow 30a by an amount corresponding to the width of the character 32 in the direction indicated by the arrow 30a. A portion of the first hole-filled mark image 47 corresponding to a portion near the character 32 may be expanded and contracted in the direction indicated by the arrow 30b by an amount corresponding to the height in the direction indicated by . .

FIG. 10A is a diagram showing an image 49 of the logical sum of the character image 44 and the first fill-in mark image 47 (hereinafter referred to as "second logical sum image"). FIG. 10(b) is a diagram showing an image 50 in which discontinuities caused by characters 32 in the marked image 30 are filled (hereinafter referred to as "second filled mark image").

If the marks 33 are close together in the marked image 30 , the second dilated and shrunk mark image 48 will not only show the discontinuity of the marks 33 caused by the character 32 , but also simply the marks 33 close together in the marked image 30 . There is a possibility that the part that was doing is also filled in. Therefore, the filled-in mark image generation unit 18b removes the filled-in portion of the mark 33 other than the discontinuity caused by the characters 32 from the second dilated and shrunk mark image 48, as shown in FIG. After the processing, a second logical sum image 49 shown in FIG. 10(a), which is a logical sum image of the character image 44 extracted in S103 and the first hole-filled mark image 47 extracted in S108, is generated. (S112), and the second image shown in FIG. A filled mark image 50 is generated (S113).

FIG. 11 is a diagram showing an unnecessary area-removed mark image 60 generated from the second hole-filled mark image 50 .

If the marks 33 and the characters 32 are close to each other in the marked image 30 , there is a possibility that the characters 32 remain in the second filled-in mark image 50 . For example, in the second hole-filled mark image 50 shown in FIG. 10(b), the letter "T" remains. Therefore, as shown in FIG. 2, after the process of S113, the unnecessary area removed mark image generation unit 18c generates the color extracted mark image 41 extracted in S102 from the second hole-filled mark image 50 generated in S113. By removing continuous areas that do not overlap with the area of , an unnecessary area-removed mark image 60 shown in FIG. 11 is generated (S114).

After the processing of S114, the control unit 18 executes specific processing according to the mark 33 indicated by the unnecessary area removed mark image 60 generated in S114 (S115), and ends the operation shown in FIG. Here, the specific processing includes, for example, removal of the mark 33 from the marked image 30, redaction of the portion of the mark 33 in the marked image 30, and deletion of the characters 32 in the portion surrounded by the mark 33 in the marked image 30. extraction and so on.

As described above, the MFP 10 removes, from the second hole-filled mark image 50, a continuous area that does not overlap the area of the color extraction mark image 41 at all, thereby removing the unnecessary area from the mark 33 portion. Since the unnecessary area-removed mark image 60 is generated as an image of (S114), when the marks 33 are close to each other in the mark-added image 30, ruled lines and characters , etc. exists, the possibility that this described portion is included in the unnecessary area-removed mark image 60 can be reduced. Therefore, the MFP 10 can improve the accuracy of extracting the mark 33 from the marked image 30 .

Since the MFP 10 distinguishes between ruled lines and characters, which have different influences on the mark 33 (S104 to S113), the accuracy of extracting the mark 33 from the marked image 30 can be improved.

In the present embodiment, the MFP 10 determines the amount of expansion or contraction in the direction indicated by the arrow 30a in relation to the amount of expansion in S110 and the amount of contraction in S111 according to the width of the characters 32 in the direction indicated by the arrow 30a. , and the amount of expansion or contraction in the direction indicated by the arrow 30b is set according to the height of the character 32 in the direction indicated by the arrow 30b. However, regarding the amount of expansion in S110 and the amount of contraction in S111, the MFP 10 determines the width of the character 32 in the direction indicated by the arrow 30a and the width of the character 32 in the direction indicated by the arrow 30b regardless of the direction of expansion or contraction. The amount of expansion or contraction may be the amount corresponding to the larger one of the heights.

Unlike ruled lines, characters are considered to be less likely to be covered by a plurality of marks 33 . Therefore, unlike ruled lines, even if the image is expanded or shrunk by an amount greater than the appropriate amount, there is a high possibility that unnecessary portions will be appropriately removed by the processing of S112 to S114. be done. Therefore, the MFP 10 may perform the operation shown in FIG. 12 instead of the operation shown in FIG. The operation shown in FIG. 12 is the same as the operation shown in FIG. 2 in which the process of S129 is executed instead of the process of S109. In S129, the hole-filled mark image generation unit 18b sets the width of the character 32 in the direction indicated by the arrow 30a and the height of the character 32 in the direction indicated by the arrow 30b to 72 points, for example, which is included in the marked image 30. Get the width and height of the largest possible size character. Compared to the operation shown in FIG. 2, the operation shown in FIG. 12 eliminates the need to obtain the width and height of each character 32 from the character image 44, and always obtains fixed width and height values. Therefore, the processing load on the control unit 18 can be reduced.

In the present embodiment, the MFP 10 fills in the discontinuities of the marks 33 caused by ruled lines before filling in the discontinuities of the marks 33 caused by characters. However, the MFP 10 may fill in the discontinuities of the marks 33 caused by characters before filling in the discontinuities of the marks 33 caused by ruled lines.

In the present embodiment, the MFP 10 separately fills in gaps in the mark 33 caused by ruled lines and fills in gaps in the mark 33 caused by characters. However, the MFP 10 may fill in gaps in the marks 33 without distinguishing between ruled lines and characters.

In the present embodiment, the MFP 10 employs characters and ruled lines as description portions other than the marks 33 extracted from the marked image. However, the image processing system of the present invention may employ only one of characters and ruled lines as the description part other than the mark 33 extracted from the marked image. In addition, the image processing system of the present invention extracts description parts other than the marks 33 from the marked image, instead of at least one of characters and ruled lines, or in addition to characters and ruled lines, for example, figures, etc. A written portion that is neither characters nor ruled lines may be employed.

Although the image processing system of the present invention is an MFP in this embodiment, it may be an image forming apparatus other than an MFP such as a dedicated printer, or a computer other than an image forming apparatus such as a PC (Personal Computer). The image processing system of the present invention may be composed of one computer or may be composed of a plurality of computers.

10 MFP (image processing system, computer)
17a image processing program 18a image extractor 18b hole-filled mark image generator 18c unnecessary area-removed mark image generator 30 mark-added image 31 ruled line (description portion)
32 characters (description part)
33 mark 41 color extraction mark image 42 non-mark partial image 43 ruled line image 44 character image 45 first inflated and shrunk mark image 46 first logical sum image 47 first hole-filled mark image 48 second inflated and shrunk mark Image 49 Second OR image 50 Second hole-filled mark image (hole-filled mark image)
60 Unnecessary area removed mark image

Claims (3)

an image extracting unit for extracting an image of the marked portion and a non-marked partial image as an image of the written portion other than the mark, based on colors from the marked image as the marked image;
The color extraction mark image and the non-filled mark image obtained by filling discontinuities caused by portions other than the mark in the color extraction mark image as the image of the mark portion extracted by the image extraction unit. a hole-filled mark image generating unit for generating based on the mark partial image using image expansion and contraction, image logical sum, and image logical product;
An unnecessary area-removed mark image is generated as an image of the mark portion from which the unnecessary area has been removed by removing a continuous area that does not overlap at all with the area of the color extraction mark image from the hole-filled mark image. and an unnecessary area-removed mark image generator. The image extracting unit extracts, as the non-marked partial images, a ruled line image as an image of a ruled line portion and a character image as an image of a character portion,
The filled-in mark image generation unit generates a first logical sum image as an image of a logical sum of the color extraction mark image and one of the ruled line image and the character image,
The filled mark image generation unit generates a first expanded and contracted mark image by expanding and contracting the color extraction mark image by an amount corresponding to the one image,
The hole-filled mark image generation unit generates a first hole-filled mark image as an image of a logical product of the first logical sum image and the first expanded and contracted mark image,
The filled-in mark image generation unit generates a second logical sum as an image of a logical sum of the first filled-in mark image and the other image of the ruled line image and the character image that is not the one image. generate the image,
The hole-filled mark image generating unit generates a second expanded/shrinked mark image by expanding and contracting the first filled-in mark image by an amount corresponding to the other image,
The filled-in mark image generation unit generates a second filled-in mark image as an image of a logical product of the second logical sum image and the second dilated/contracted mark image as the filled-in mark image. 2. The image processing system according to claim 1, wherein: an image extracting unit for extracting an image of the marked portion and a non-marked partial image as an image of the written portion other than the mark, based on colors from the marked image as the marked image;
The color extraction mark image and the non-filled mark image obtained by filling discontinuities caused by portions other than the mark in the color extraction mark image as the image of the mark portion extracted by the image extraction unit. a hole-filled mark image generating unit for generating based on the mark partial image using image expansion and contraction, image logical sum, and image logical product;
An unnecessary area-removed mark image is generated as an image of the mark portion from which the unnecessary area has been removed by removing a continuous area that does not overlap at all with the area of the color extraction mark image from the hole-filled mark image. and an unnecessary area-removed mark image generation unit for generating an image processing program that causes a computer to realize.

Images