JP2011228792A - Image processing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image processing device capable of determining a character which a user desires to specify, even when the user has roughly marked thereon.SOLUTION: A network composite machine mounted with an image processing device comprises: a scanner for obtaining image data from a document; a marked area extracting portion 31 for extracting a marked area; a character described area extracting portion 331 for extracting a character described area, in which a relevant character is described, for each character contained in a character string; an overlapped character described area extracting portion 322 for extracting an overlapped character described area in which the character described area and the marked area are overlapped with respect to each other; a character calculating portion 333 for calculating each length of the overlapped character described area and the character described area corresponding to the overlapped character described area in a character string direction; and a character determining portion 334 for determining whether a character to be determined is the character specified by a marking on the document or not based on the length of the overlapped character described area in the character string direction with respect to the length of the character described area in the character string direction.

Description

  The present invention relates to an image processing apparatus.

  An image processing apparatus that reads a document on which characters are marked with a marker pen with a scanner to generate image data, and performs image processing such as deletion, color change, and enhancement on the marked characters is disclosed in Patent Document 1 below. Has been. The image processing apparatus disclosed in Patent Document 1 below first obtains a marking area filled with a marker pen and a rectangular area in which characters are described in order to specify an area to be subjected to image processing. The image processing apparatus identifies a rectangular area including the marking portion in the area, and sets this as a target for image processing.

JP 2001-230919 A

  By the way, when a user marks a character on a document with a marker pen or the like, the character adjacent to the character to be specified may be marked. In this case, in the image processing apparatus described in Patent Document 1, since the marking portion is located in the rectangular area where the character next to the character to be specified is described, the character that is not desired to be specified is specified as the image processing target. Will be.

  For this reason, when using the image processing apparatus, it is necessary for the user to mark only the character that the user wants to designate, and to perform marking precisely so that the non-designated character is not marked. However, in general, a thick pen is used for marking so that it is easy to fill the area where the character is written. Therefore, in order to prevent marking around the character to be specified, it is necessary to use a nerve. Need to be marked.

  The present invention has been made to solve the above-described problems, and provides an image processing apparatus that can specify a character that the user wants to specify even when the user has roughly marked. Objective.

  An image processing apparatus according to the present invention includes an acquisition unit that acquires image data of a document, a marking extraction unit that extracts a colored marking region on the document from the image data acquired by the acquisition unit, and the document processing unit. A character description area extracting unit that extracts a character description area in which the character is described for each character included in a plurality of described characters from image data, a character description area extracted by the character description area extraction unit, and a marking extraction About the overlapping character area extracting means for extracting the overlapping character area overlapping the marking area extracted by the means, the overlapping character area extracted by the overlapping character area extracting means, and the character description area corresponding to the overlapping character area, Character calculation means for calculating the length in the direction in which a plurality of characters are arranged, and character descriptions calculated by the character calculation means Based on the above length of the overlapping character region with respect to the length of the region, the character is characterized in that it comprises a character determination means for determining whether or not the specified characters by the marking on the document.

  According to the image processing apparatus of the present invention, the colored marking area on the document is extracted from the image data. In addition, for each character included in a plurality of characters described in the document, a character description region in which the character is described is extracted from the image data. Then, an overlapping character area where the character description area and the marking area overlap is extracted. Subsequently, the length in the direction in which the plurality of characters are arranged is calculated for the overlapping character region and the character description region corresponding to the overlapping character region. Then, based on the length of the overlapping character area with respect to the length of the character description area, it is determined whether the character is designated by marking. For this reason, even if the user has roughly marked the character and, for example, the marking has been applied to the character next to the character to be designated, or even if the character to be designated is not completely filled Based on the degree, the user can determine whether the character is desired to be designated by marking. Therefore, even if the user has roughly marked, it is possible to specify the character that the user wants to specify.

  In the image processing apparatus according to the present invention, a column description area extracting unit that extracts, from image data, a column description area in which a character string is described for each character string included in a plurality of character strings arranged and described in a document. A duplicate column region extraction unit that extracts a duplicate column region in which a column description region extracted by the column description region extraction unit and a marking region extracted by the marking extraction unit overlap, and an overlap extracted by the duplicate column region extraction unit A column calculation unit for calculating a length in a direction orthogonal to a direction in which a plurality of characters included in the character string are arranged, and a column calculation unit for the column region and the column description region corresponding to the overlapping column region; Whether or not the character string includes characters designated by marking on the manuscript based on the orthogonal length of the overlapping row region with respect to the calculated length of the row description region in the orthogonal direction It is preferable and a determining column determining means.

  According to this preferable configuration, for each character string included in the plurality of character strings described in the document, a column description area in which the character string is described is extracted from the image data. Then, an overlapping row area where the row description area and the marking area overlap is extracted. Subsequently, the length in the orthogonal direction orthogonal to the direction in which a plurality of characters included in the character string are arranged is calculated for the overlapping row region and the column description region corresponding to the overlapping row region. Then, it is determined whether or not the character string includes a character designated by the marking based on the length in the orthogonal direction of the overlapping column region with respect to the length in the orthogonal direction of the column description region. For this reason, even if the user roughly marks a character and, for example, the character string next to the character string to be specified has been marked, or even if the character string to be specified is not completely filled Whether or not the character string includes a character that the user wants to designate by marking can be determined based on the degree of the character. Therefore, even if the user has roughly marked, it is possible to specify a character string including the character that the user wants to specify.

  In the image processing apparatus according to the present invention, the character determination means includes a character designated by marking on the document for a plurality of characters included in the character string determined to include the character marked on the document by the column determination means. It is preferable to determine whether or not.

  According to this preferable configuration, the character to be determined is extracted in character string units, and then it is determined whether or not the character included in the extracted character string is a character specified by marking. Therefore, it is possible to efficiently specify the character designated by the marking as compared with the case where all characters are determined one by one.

  In the image processing apparatus according to the present invention, it is preferable that the image processing apparatus further includes a determination unit that determines whether the overlapping column region extracted by the overlapping column region extraction unit is singular or plural. When it is determined that the number is singular, it is preferable to determine whether or not a plurality of characters included in the character string corresponding to the overlapping row area are characters designated by marking on the document.

  According to this preferable configuration, since it is determined whether the overlapping row region is singular or plural, it is possible to determine whether marking is performed over a plurality of character strings. Then, when there is a single overlapping row area, it is determined whether or not a plurality of characters included in the character string corresponding to the overlapping row area are characters designated by marking on the document. For this reason, when the character string currently marked is one line, the several character contained in the character string can be made into the character for determination irrespective of the said length of an overlap line area | region.

  According to the present invention, it is possible to specify a character that the user wants to specify even when the user has roughly marked.

1 is a block diagram illustrating a configuration of a network multi-function peripheral equipped with an image processing apparatus according to an embodiment. It is a block diagram which shows the structure of the control part which a network multifunction device has. It is a figure for demonstrating the column extraction process which the column extraction part which a control part has performs with respect to an example of the marked character string. It is a figure for demonstrating the character extraction process which the character extraction part which a control part has performs with respect to an example of the marked character. It is a figure for demonstrating the character extraction process which the character extraction part which a control part has performs with respect to another example of the marked character. 10 is a flowchart (first half) illustrating a processing procedure of a specified character extraction process performed by the network multifunction peripheral. 10 is a flowchart (second half) illustrating a processing procedure of a specified character extraction process performed by the network multifunction peripheral.

  DESCRIPTION OF EMBODIMENTS Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the drawings. First, the configuration of the network MFP 1 in which the image processing apparatus according to the embodiment is mounted will be described with reference to FIGS. 1 and 2. FIG. 1 is a block diagram showing the configuration of the network multifunction device 1. FIG. 2 is a block diagram illustrating a configuration of the control unit 30 included in the network multifunction peripheral 1.

  The network multifunction device 1 is configured to be able to send and receive information to and from devices such as a personal computer 5 connected to a LAN (Local Area Network) 90, and has a composite function including a print function, a scan function, and a FAX function. Equipment. The network multifunction device 1 has a function of reading a document on which characters are marked by a user, generating image data, and specifying a designated character designated by the marking. For the specified designated character, for example, OCR (character recognition) processing or editing processing such as deletion, color conversion, and enhancement is performed.

  In order to exhibit the above functions, the network multifunction peripheral 1 includes an operation panel 10, a display 11, a scanner 12, a printer 13, an NCU (Network Control Unit) 14, a modem 15, a LAN I / F (LAN interface) 19, and a control unit. 30. The NCU 14 controls the connection between the modem 15 and the public switched telephone network (PSTN) 91 and controls the FAX transmission / reception function.

  The scanner 12 reads a document and acquires image data. The scanner 12 functions as acquisition means described in the claims. On the manuscript, a plurality of character strings are arranged and described, and some characters are marked. Here, the marking is to color an area on the document with a color different from the background color of the document, that is, the background color. For example, the user can mark a character by drawing a line on the character with a color marker or the like. Moreover, a character can be marked by painting the area | region where the character was described. By marking the character in this manner, the user designates the character to be edited.

  The scanner 12 reads an original in color and generates image data including RGB data. For this reason, for example, the scanner 12 has a single-plate CCD to which color filters of red (R), green (G), and blue (B) are attached, and acquires data of RGB colors. Further, for example, a 3CCD system that splits light incident on the CCD into three primary colors of RGB by reflection of a dichroic film in the spectral prism may be used. Image data generated by the scanner 12 is output to the control unit 30.

  The control unit 30 includes a CPU 16, a ROM 17, a RAM 18, and the like, and specifies a designated character designated by marking from input image data. As shown in FIG. 2, the control unit 30 includes a marking extraction unit 31, a column extraction unit 32, and a character extraction unit 33 as functional components.

  The marking extraction unit 31 extracts a marking area from the image data input from the scanner 12. The marking area is a colored area on the document. For example, the marking extraction unit 31 converts the image data from the RGB color space to the HLS color space, and determines whether each pixel is a pixel in the marking area based on the position of each pixel included in the image data in the HLS color space. To identify. Here, the HLS color space is a color space including three components of hue (Hue), saturation (Saturation), and luminance (Lightness / Luminance).

  The marking extraction unit 31 uses a group of colored pixels as a marking region, and specifies the marking region using position coordinate data or the like. For example, if the marking area is rectangular, the marking area can be specified by position coordinate data at four corners. The marking extraction unit 31 outputs position coordinate data for specifying the marking area to the column extraction unit 32 and the character extraction unit 33. The marking extraction unit 31 functions as marking extraction means described in the claims.

  The column extraction unit 32 extracts a character string including a designated character from among a plurality of character strings described in the document. The column extraction unit 32 includes a column description region extraction unit 321, an overlapping column region extraction unit 322, a column calculation unit 323, a column determination unit 324, and a determination unit 325 in order to perform the above processing.

  The column description area extraction unit 321 extracts a column description area in which a character string is described for each character string from the image data. The column description area is a rectangular area surrounding the area where the character string is described. The column description region extraction unit 321 can extract the column description region by specifying the position coordinates of the four corners of the column description region.

  For example, generally, when performing OCR processing, position coordinate data that specifies four corners of a rectangular region in which a character string is described is generated. The column description area extraction unit 321 specifies the position coordinate data of the four corners of the rectangular area in which the character string is described, using the position coordinate data generated by the OCR process. The column description area extraction unit 321 functions as a column description area extraction unit described in the claims.

  The overlapping row area extraction unit 322 extracts the overlapping row area where the marking area and the row description area overlap. The overlapping row area extraction unit 322 functions as an overlapping row area extraction unit described in the claims. The column calculation unit 323 calculates the lengths of the overlapping column area and the column description area in the orthogonal direction orthogonal to the character string direction. The character string direction is a direction in which a plurality of characters included in the character string are arranged. The character string direction is specified, for example, in the longitudinal direction of the character string description area. The column calculation unit 323 functions as a column calculation unit described in the claims.

  The column determination unit 323 determines that the character string to be determined includes the designated character when the length of the overlapping column region in the orthogonal direction is equal to or greater than a predetermined ratio with respect to the length of the column description region in the orthogonal direction. To do. Then, when the length of the overlapping row area is smaller than a predetermined ratio with respect to the length of the row description area, it is determined that the character string does not include the designated character. The predetermined ratio can be set to 1/2, for example. The column determination unit 324 functions as a column determination unit described in the claims.

  The determination unit 325 determines whether the overlapping column region extracted by the overlapping column region extraction unit 322 is single or plural. This makes it possible to determine whether or not there are a plurality of lines of character strings that overlap the marking area, that is, whether or not marking is performed over a plurality of character strings. The determination unit 325 functions as a determination unit described in the claims. When the determination unit 325 determines that there are a plurality of overlapping column regions, the column calculation unit 323 and the column determination unit 324 determine whether the character string includes a designated character.

  With reference to FIG. 3, a column extraction method in which the column extraction unit 32 extracts a character string will be described using a specific example. As shown in FIG. 3, as an example, a plurality (three columns) of character strings 41 to 43 are arranged side by side in an orthogonal direction 45 that is orthogonal to each character string direction 44. In FIG. 3, a gray rectangular area is a marking area 46. In the example of FIG. 3, the marking region 46 extends over a plurality of character strings 41 and 42. In FIG. 3, the marking region 46 is shown in gray, but pink or yellow highlighter or any other color may be worn. The marking area 46 is extracted by the marking extraction unit 31 described above.

  The column description area extraction unit 321 extracts column description areas 411, 421, and 431 in which the upper character string 41, the interrupted character string 42, and the lower character string 43 are respectively described. The overlapping line area extraction unit 322 extracts the overlapping line area 412 where the marking area 46 and the line description area 411 of the upper character string 41 overlap. In addition, the overlapping line area extraction unit 322 extracts an overlapping line area 422 where the marking area 46 and the line description area 421 of the middle character string 42 overlap.

  In this case, the determination unit 325 determines that there are a plurality of overlapping row regions 412 and 422. Since it is determined that there are a plurality of overlapping string areas, the upper character string 41 corresponding to the overlapping string area 412 and the middle character string 42 corresponding to the overlapping string area 422 become the character strings to be determined.

  For the upper character string 41, the column calculation unit 323 calculates the length 413 of the overlapping column region 412 in the orthogonal direction 45 and the length 414 of the column description region 411 in the orthogonal direction 45. As the length, for example, the length in the orthogonal direction 45 can be calculated for a plurality of locations along the character string direction 44, and an average value of the calculated lengths can be used. Then, the column determination unit 324 determines whether or not the character string 41 includes a designated character based on the calculated length 413 of the overlapping column region 412 with respect to the length 414 of the column description region 411.

  In the example shown in FIG. 3, the length 413 of the overlapping column area 412 is not more than half the length 414 of the column description area 411, so that the character string 41 is determined not to include the designated character. Further, according to this method, the middle character string 42 is determined by the column determining unit 324 to include a designated character. Since the lower character string 43 has no overlapping row area and is not marked at all, it is determined that it does not include a designated character. As a result, even if the user has roughly marked the character string to be marked only in the middle character string 42 and the upper character string 41 has been marked, the user has marked it. Only the character string 42 can be extracted accurately.

  Note that FIG. 3 illustrates the case where there are a plurality of character strings overlapping the marking area 46. However, when there is a single character string overlapping the marking area 46, the determination unit 325 determines that the overlapping line area is singular. In this case, the column determination unit 325 determines that the character string corresponding to the overlapping column region includes a designated character. The column extraction unit 32 extracts the character string determined by the column determination unit 324 as including the designated character, and outputs the character string to the character extraction unit 33.

  The character extraction unit 33 illustrated in FIG. 2 extracts a designated character from a plurality of characters included in the character string extracted by the column extraction unit 32. The character extraction unit 33 includes a character description region extraction unit 331, an overlapping character region extraction unit 332, a character calculation unit 333, and a character determination unit 334 in order to extract a designated character.

  The character description area extraction unit 331 extracts, from the image data, a character description area in which a character is described for each character included in the character string to be determined. For example, when performing OCR processing, position coordinate data that specifies four corners of a rectangular area in which each character is described is generated. The character description area extraction unit 331 can specify and extract a character description area based on the position coordinate data of the four corners. The character description area extraction unit 331 functions as a character description area extraction unit described in the claims.

  The duplicate character region extraction unit 332 extracts a duplicate character region where the marking region and the character description region overlap. The overlapping character area extracting unit 332 functions as an overlapping character area extracting unit described in the claims. The character calculation unit 333 calculates the lengths of the overlapping character area and the character description area in the character string direction 44, respectively. The character calculation unit 333 functions as character calculation means described in the claims.

  The character determination unit 334 determines that the character to be determined is a designated character when the length of the overlapping character area in the character string direction 44 is equal to or greater than a predetermined ratio with respect to the length of the character description area in the character string direction 44. judge. Then, when the length of the overlapping character area is smaller than a predetermined ratio with respect to the length of the character description area, it is determined that the character to be determined is not a designated character. The predetermined ratio can be set to 1/2, for example. Note that the character extraction unit 33 extracts a character that is determined to be a designated character by the character determination unit 334.

  With reference to FIG. 4, the character extraction process in which the character extraction unit 33 extracts characters will be described using a specific example. First, the case where the first character “sentence” 51 from the right among the plurality of characters included in the character string “string character string sentence” shown in FIG.

  The character description area extraction unit 331 extracts the character description area 511 in which the character “sentence” 51 is described. The overlapping character area extraction unit 332 extracts an overlapping character area 512 between the marking area 46 extracted by the marking extraction unit 31 and the character description area 511 extracted by the character description area extraction unit 331. The overlapping character area 512 can be specified by position coordinate data.

  The character calculation unit 333 calculates the length 513 of the overlapping character area 512 in the character string direction 44 and the length 514 of the character description area 511 in the character string direction 44. As the length, for example, the length in the character string direction 44 is calculated for each of a plurality of positions along the orthogonal direction 45, and an average value of the calculated lengths can be used. Then, the character determination unit 334 determines whether or not the character “sentence” 51 is a designated character based on the calculated lengths of the overlapping character region 512 and the character description region 511.

  In the example shown in FIG. 4, since the length 513 of the overlapping character area 512 is not more than half the length 514 of the character description area 511, it is determined that the character 51 “sentence” is not a designated character. According to the above method, the second character “string” 52 from the right and the third character “character” 53 from the right match the overlapping character area and the character description area, and the character string direction 44 of the overlapping character area is the same. The length is at least half of the length in the character string direction 44 of the character description area. Therefore, the character “string” 52 and the character “character” 53 are determined to be designated characters.

  The fourth character “sentence” 54 from the right is only partially marked, but the length 543 of the overlapping character area 542 in the character string direction 44 is the length of the character description area 541 in the character string direction 44. More than half of 544. Therefore, the character “sentence” 54 is determined to be a designated character. Since the fifth character “sequence” 55 from the right is not marked at all, it is determined that there is no overlapping character area and it is not a designated character. In this case, the character extraction unit 33 extracts only the character “sentence” 54, the character “character” 53, and the character “string” 52 as designated characters.

  As shown in FIG. 4, the user performs rough marking, and the character “sentence” 54, the character “letter” 53, and the character “sentence” 51 on the right side of the character “string” 52 are marked. The character “sentence” 54 is not marked on the left side of the character. Even in this case, it is determined that the character “sentence” 54 on the left side is included in the designated character and the character “sentence” 51 on the right side is not included in the designated character, and the character “sentence” 54 and the character “character” that the user wants to mark are determined. 53, only the character “string” 52 can be accurately extracted.

  Next, with reference to FIG. 5, a character extraction process in which the character extraction unit 33 extracts characters will be described using another specific example. In the example shown in FIG. 5, the width of the marking region 46 is narrower than that of the example shown in FIG.

  In the first character “sentence” 61 from the right, the length 613 of the overlapping character region 612 in the character string direction 44 is smaller than half the length 614 of the character description region 611 in the character string direction 44. Therefore, it is determined that the character “sentence” 61 is not a designated character. In the second character “string” 62 from the right and the third character “character” 63 from the right, the length of the overlapping character area in the character string direction 44 is more than half of the length of the character description area in the character string direction 44. It is. Therefore, the character “string” 62 and the character “character” 63 are determined to be designated characters.

  The fourth character “sentence” 64 from the right is only partially marked, but the length 643 of the overlapping character area 642 in the character string direction 44 is the length of the character description area 641 in the character string direction 44. More than half of 644. Therefore, the character “sentence” 64 is determined to be the designated character. Since the fifth character “column” 65 from the right is not marked at all, it is determined that there is no overlapping character area and it is not a designated character. In this case, the character extraction unit 33 extracts only the character “sentence” 64, the character “character” 63, and the character “string” 62 as the designated characters.

  Also in this case, the user performs rough marking, and the character “sentence” 61 on the right side of the character string to be marked is marked, and the left part of the character “sentence” 64 is not marked. Even in this case, it is determined that the character “sentence” 64 on the left side is included in the designated character, and the character “sentence” 61 on the right side is not included in the designated character. "63" and only the character "string" 62 can be extracted correctly.

  In the example of FIG. 5, the line-shaped marking region 46 is located at the center in the orthogonal direction 45 of the character string, but the same applies even if it is located below or above the character string. Processing can be applied. Also in this case, only the character “sentence” 64, the character “character” 63, and the character “string” 62 that the user wants to mark can be extracted correctly.

  Subsequently, the operation of the network MFP 1 will be described with reference to FIGS. 6 and 7. 6 and 7 are the first half and the second half of the flowchart showing the processing procedure of the designated character extraction process performed by the network MFP 1.

  First, when the user performs an operation using the operation panel 10 so as to read the marked document, in step S101, the document set on the network multifunction peripheral 1 is read by the scanner 12, and the image data of the document is obtained. To be acquired. In step S102, the marking area 46 is extracted from the image data. In step S103, an OCR process is performed on the image data to extract a column description area in which a character string is described.

  Subsequently, in step S104, an overlapping column region in which the marking region 46 extracted in step S102 and the column description region extracted in step S103 overlap is extracted. Note that the marking area 46 extraction process performed in step S102, the column description area extraction process performed in step S103, and the overlapping column area extraction process performed in step S104 can be performed by the above-described method. The description is omitted here.

  In step S105, it is determined whether there is a character string overlapping the marking area 46. That is, it is determined in step S104 whether or not an overlapping row area has been extracted. If no overlapping row area is extracted, there is no marked character and the process ends. If an overlapping row area is extracted, the process proceeds to step S106.

  In step S106, it is determined whether or not the character string overlapping the marking area 46 is a single line. That is, it is determined whether the overlapping row area extracted in step S104 is singular or plural. If there is a single overlapping column area, the process proceeds to step S108. If there are a plurality of overlapping row areas, the process proceeds to step S107.

  In step S107, for the character string to be determined, the length in the orthogonal direction 45 between the column description area and the overlap string area is calculated, and whether or not the length of the overlap string area is half or more of the length of the column description area. Is judged. If it is determined that the length of the overlapping column area is not more than half the length of the column description area, the process proceeds to step S109. In step S109, it is determined that the character string to be determined is not a character string including the designated character, and the process proceeds to step S116.

  If it is determined in step S107 that the length of the overlapping column area is half or more of the length of the column description area, the process proceeds to step S108. In step S108, it is determined that the character string to be determined is a character string including a designated character. Subsequently, in step S110, the character description area of the character to be determined is extracted from the character string determined to include the designated character in step S108. In step S111, an overlapping character area in which the character description area extracted in step S110 and the marking area 46 overlap is extracted.

  In step S112, for the character to be determined, the length in the character string direction 44 between the character description area and the overlapping character area is calculated, and whether or not the length of the overlapping character area is half or more of the length of the character description area. Is judged. If it is determined that the length of the overlapping character area is not more than half the length of the character description area, the process proceeds to step S114. In step S114, it is determined that the character to be determined is not a designated character, and the process proceeds to step S115.

  If it is determined in step S112 that the length of the overlapping character area is half or more of the length of the character description area, the process proceeds to step S113. In step S113, it is determined that the character to be determined is a designated character. In step S115, it is determined whether or not there is a next character to be determined for the character string determined to include the designated character in step S108. If there is a next character to be determined, the process returns to step S110 to determine again whether or not the character to be determined is a designated character.

  If the determination is completed for all characters included in the character string, the process proceeds to step S116. In step S116, it is determined whether there is a character string to be determined next. If there is a next character string to be determined, the process returns to step S107, and determination processing is performed again to determine whether or not the character string to be determined includes the designated character. When the determination process is completed for all the character strings overlapping the marking region 46, a series of extraction processes is completed. Through the above processing, the designated character designated by the marking is specified. The specified designated character is displayed on the display 11 or the like. Further, the specified designated character may be subjected to OCR processing or various editing may be performed.

  According to the network multifunction peripheral 1 according to the present embodiment described above, the length in the character string direction 44 between the overlapping character area and the character description area is calculated. When the ratio of the length of the overlapping character area in the character string direction 44 to the length of the character description area 44 in the character string direction 44 is equal to or greater than a predetermined value, it is determined that the determination target character is the designated character. For this reason, even if the user has roughly marked the character and, for example, the character next to the character to be specified has been marked, or even if the character to be specified is not completely filled, the degree of marking Based on the above, it can be determined whether or not it is a designated character. Therefore, even if the user has roughly marked, it is possible to specify the character that the user wants to specify.

  Further, according to the network MFP 1, the length in the orthogonal direction 45 between the overlapping character string area and the column description area is calculated. When the ratio of the length in the orthogonal direction 45 of the overlapping row area to the length in the orthogonal direction 45 of the row description area is equal to or greater than a predetermined value, it is determined that the character string to be determined includes the designated character. For this reason, even if the user has roughly marked the character and, for example, the character string next to the character string to be specified has been marked, or even if the character string to be specified is not completely filled Whether or not the character string includes the designated character can be determined based on the degree to which the character has been set. Therefore, even if the user has roughly marked, it is possible to specify a character string including the character that the user wants to specify.

  Further, according to the network MFP 1, after the character string including the designated character is extracted, the designated character is extracted from a plurality of characters included in the extracted character string. That is, a character to be determined is extracted in character string units, and then it is determined whether or not the character included in the extracted character string is a character specified by marking. Therefore, it is possible to efficiently specify the character designated by the marking as compared with the case where all characters are determined one by one.

  Furthermore, according to the network multifunction device 1, since it is determined whether the overlapping row area is singular or plural, it is possible to determine whether marking is performed over a plurality of character strings. Then, when there is a single overlapping column area, it is determined whether or not a plurality of characters included in the character string corresponding to the overlapping column area are designated characters. For this reason, when the character string currently marked is one line, the several character contained in the character string can be made into the character for determination irrespective of the said length of an overlap line area | region. That is, as shown in FIG. 5, even when the marking is made in a line shape and the ratio of the length in the orthogonal direction 45 of the overlapping row region to the length in the orthogonal direction 45 of the row description region is smaller than a predetermined value, A plurality of characters included in the character string can be determined.

  Although the embodiment of the present invention has been described above, the present invention is not limited to the above embodiment, and various modifications can be made. For example, in the above embodiment, the determination is made based on whether or not the length of the overlapping character region or the overlapping column region is half or more of the character description region or the column description region, but is not limited thereto. The determination criterion can be arbitrarily set such that the length of the overlapping character region or the overlapping column region is one third or more of the character description region or the column description region.

  In the above embodiment, the column extraction unit 32 is provided. However, the column extraction unit 32 may not be provided. In this case, the character extraction unit 33 determines whether or not each character that overlaps at least part of the marking region 46 is a designated character. In the above embodiment, the scanner 12 is provided. However, the scanner 12 may not be provided. In this case, the image data of the document is acquired from the external device, and the marked characters are extracted by performing image processing on the acquired image data. Furthermore, in the above-described embodiment, the case where the characters are written horizontally is specifically described. However, the present invention can be applied even when the characters are written vertically. In this case, the character string direction is the vertical direction.

DESCRIPTION OF SYMBOLS 1 Network compound machine 12 Scanner 30 Control part 31 Marking extraction part 321 Column description area extraction part 322 Duplicate line area extraction part 323 Column calculation part 324 Column determination part 325 Judgment part 331 Character description area extraction part 332 Duplicate character area extraction part 333 Character Calculation unit 334 Character determination unit

Claims (4)

Acquisition means for acquiring image data of a document;
Marking extraction means for extracting a colored marking area on the document from the image data acquired by the acquisition means;
A character description region extracting means for extracting, from the image data, a character description region in which the character is described for each character included in a plurality of characters described side by side in the document;
An overlapping character area extracting means for extracting an overlapping character area where the character describing area extracted by the character description area extracting means and the marking area extracted by the marking extracting means overlap;
Character calculation means for calculating the length in the direction in which the plurality of characters are arranged for the overlapping character area extracted by the overlapping character area extraction means and the character description area corresponding to the overlapping character area;
Character determination for determining whether the character is a character designated by marking on the document based on the length of the overlapping character region with respect to the length of the character description region calculated by the character calculation unit Means,
An image processing apparatus comprising: Column description area extracting means for extracting, from the image data, a column description area in which the character string is described for each character string included in a plurality of character strings arranged and described in the document;
An overlapping column region extracting unit for extracting an overlapping column region where the column describing region extracted by the column describing region extracting unit and the marking region extracted by the marking extracting unit overlap;
About the overlapping row area extracted by the overlapping row area extraction means and the column description area corresponding to the overlapping row area, the length in the orthogonal direction orthogonal to the direction in which a plurality of characters included in the character string are arranged Column calculating means for calculating
Based on the length in the orthogonal direction of the overlapping row area with respect to the length in the orthogonal direction of the row description area calculated by the row calculation means, the character string is a character designated by marking on the document. Column determining means for determining whether or not to include,
The image processing apparatus according to claim 1, further comprising:   The character determination means determines whether or not the plurality of characters included in the character string determined to include a character marked on the original by the line determination means are characters specified by marking on the original. The image processing apparatus according to claim 2, wherein the determination is performed. A determination unit that determines whether the overlapping column region extracted by the overlapping column region extraction unit is singular or plural;
The character determination unit is designated by marking on the document for the plurality of characters included in the character string corresponding to the overlap column region when the determination unit determines that the overlap column region is singular. The image processing apparatus according to claim 3, wherein it is determined whether the character is a character.

Images