JP6148193B2 - Form definition creation device, form definition creation method, and program - Google Patents

Description

  The present invention relates to a form definition creation apparatus, form definition creation method, and program for digitizing and managing forms.

  Conventionally, when a form such as a paper medium is read by an image reading device such as a scanner and OCR (engineering character recognition) is performed from an image, a reading definition body called a format definition that defines a reading position, the number of digits, a character type, and the like has been provided. It is used.

  When creating a new OCR format definition, (1) a form is input from a scanner and converted into image data, (2) the form image is read into the format definition program, and (3) a mouse is displayed on the displayed image. The reading position is set by drawing a frame in step (4), and (4) the number of digits, the character type, etc. are set in the created frame to make the reading item. In this way, when creating a format definition with a new form, it is necessary for the user to define the reading items one by one, and it takes a lot of time when the number of items is large. If there are many, it will take a great effort to create a format definition.

  Therefore, when creating a format definition for a similar form, create one type of form definition, and after copying the format definition for the rest, modify the position of the reading item and the number of digits to reduce man-hours. A method for creating a format definition has been developed (see Patent Documents 1 to 4).

  For example, <1> a new form is input from the scanner, converted into image data, read into the format definition program, <2> the format definition of the diverted form is read, and a frame for the read item is placed on the previously read form image. Overlay, <3> Move the frame of the read item to the item of the form image (change the number of digits and frame size if they are different), <4> Delete any unnecessary read items, A method has been developed to create a simple reading item if necessary.

JP 7-160803 A JP 2001-229341 A JP 2007-265389 A JP 2009-53827 A

  However, in the conventional method of reusing the format definition, it is assumed that the design of the original image (diversion source) and the input image (diversion destination) for creating a new format definition are the same. There was a problem that could not be handled.

  More specifically, when trying to divert the format definition, the position of the reading item may be misaligned even if it is the same type of document due to misalignment, skew, or distortion during printing or scanner input. Since it is a separate form, the read item on the form may have moved due to a change in the form design, the interval between mark items may have changed, and there may be no read item.

  For this reason, after reading the format definition of the diverted form, the user needs to adjust the reading position and size one by one, and search for and delete unnecessary reading items. For example, as a conventional method, there is a method in which a plurality of reading items are selected with a mouse or the like and moved at a time, but since the positional deviation of the reading items is not the same for all the reading items, all the reading positions do not match in the batch movement. Finally, there is a problem that the user needs to finely adjust the position of the reading item one by one.

  The present invention has been made in view of the above problems, and an object thereof is to provide a form definition creation device, a form definition creation method, and a program capable of appropriately diverting an already created format definition. To do.

  In order to achieve such an object, the form definition creating apparatus of the present invention is a form definition creating apparatus including at least a storage unit and a control unit for creating a format definition of a character recognition form, wherein the storage unit A format definition storage means for storing the format definition of the diversion source form; and an image storage means for storing an image of the diversion destination form read by the image reading device, wherein the control unit is a format of the diversion source form. Item position search means for searching individual read items based on the definition on the image of the diverted destination form, and item association means for moving the read items of the diversion source form based on the search result by the item position search means And comprising.

  A form definition method that is executed in a form definition creation device that includes at least a storage unit and a control unit to create a format definition for a character recognition form, wherein the storage unit stores a format definition of a diversion source form Definition storage means, and image storage means for storing an image of a diversion destination form read by the image reading apparatus, and each reading item based on the format definition of the diversion source form is executed in the control unit. And an item position searching step for searching on the image of the diversion destination form, and an item association step for moving the read item of the diversion source form based on the search result by the item position search means.

  A program for creating a form definition for a character recognition form and causing a form definition creation apparatus having at least a storage unit and a control unit to execute the format definition for storing a format definition of a diversion source form Storage means and image storage means for storing an image of a diversion destination form read by the image reading device, and in the control unit, each reading item based on the format definition of the diversion source form is assigned to the diversion destination. An item position searching step for searching on the form image and an item association step for moving the read item of the diversion source form based on the search result by the item position searching means are executed.

  According to the present invention, there is an effect that the format definition of the diversion source can be appropriately diverted even when the diversion source and the diversion destination are different in design.

FIG. 1 is a block diagram showing an example of a physical configuration of a form definition creation apparatus to which the present invention is applied. FIG. 2 is a block diagram showing an example of a functional configuration of a form definition creation apparatus to which the present invention is applied. FIG. 3 is a flowchart showing an example of processing in the present embodiment. FIG. 4 is a diagram schematically illustrating an example of the created format definition. FIG. 5 is a diagram illustrating an example of a newly scanned form image. FIG. 6 is a diagram in which the diversion source read items are directly associated with each other on the form image of FIG. FIG. 7 is a diagram schematically showing the result of overall positioning. FIG. 8 is a diagram illustrating an example of a search range set by the item position search unit 102c. FIG. 9 is a flowchart illustrating an example of the search process in step SA-6. FIG. 10 is a diagram schematically showing the movement of the reading item. FIG. 11 is a diagram in which an exclusive OR (XOR) for each pixel of the diversion-source read item image and the diversion-destination form image is obtained at a certain position. FIG. 12 is a diagram in which an exclusive OR (XOR) for each pixel of the diversion-source read item image and the diversion-destination form image is obtained at a certain position. FIG. 13 is a flowchart illustrating an example of a ruled line matching method. FIG. 14 is a diagram illustrating an example in which the user associates items via the input unit 112 under the control of the item association unit 102d. FIG. 15 is a diagram illustrating an example of contention due to overlapping search ranges. FIG. 16 is a diagram for explaining a criterion for determining the presence or absence of competition. FIG. 17 is a diagram for explaining evaluation criteria for resolving the conflict. FIG. 18 is a diagram illustrating a range excluded from the search range. FIG. 19 is a diagram for illustrating an example of a conflict resolution method. FIG. 20 is a diagram illustrating an example in which a conflict list is added. FIG. 21 is a diagram showing an example in which a deletion flag is added. FIG. 22 is a diagram illustrating an example in which items are further divided into entry frames and positioning is performed for each entry frame.

  Embodiments of a form definition creation apparatus, form definition creation method, and program according to the present invention will be described below in detail with reference to the drawings. Note that the present invention is not limited to the embodiments.

[1. Configuration of this embodiment]
Next, the configuration of the form definition creation device will be described with reference to FIG. FIG. 1 is a block diagram illustrating an example of a physical configuration of a form definition creation apparatus to which the present invention is applied. FIG. 2 is a block diagram illustrating an example of a functional configuration of a form definition creation apparatus to which the present invention is applied. Yes, only the part related to the present invention is conceptually shown.

  As shown in FIG. 1, the form definition creation apparatus according to the present embodiment includes a computer 10, an image scanner 12 that reads a form image, a keyboard 13, a mouse 11, and a display 14. The illustrated physical configuration is an example, and the present invention is not necessarily limited to this configuration.

  As shown in the functional configuration of FIG. 2, the form definition creation apparatus 100 according to the present embodiment schematically includes a control unit 102 (an image input control unit 102 a, an overall positioning unit 102 b, an item position search unit 102 c, item association). Unit 102d, conflict resolution unit 102f), and storage unit 106 (format definition file 106a, image file 106b). Here, the control unit 102 is a CPU or the like that comprehensively controls the entire form definition creation device 100. The storage unit 106 is a device that stores various databases and tables. Each unit of the form definition creation apparatus 100 is connected to be communicable via an arbitrary communication path. The control unit 102 and the storage unit 106 are realized by the above-described computer 10, the input unit 112 is realized by the above-described image scanner 12, the keyboard 13, the mouse 11, and the like, and the output unit 114 is the above-mentioned display 14 is realized.

  The storage unit 106 is a storage unit, and stores various databases and tables (format definition file 106a, image file 106b, etc.).

  Among the components of the storage unit 106, in FIG. 1, a format definition file 106a is a format definition storage unit that stores a format definition for character reading. For example, the format definition file 106a may store the format definition of the diversion source form. For example, when the user operates the mouse 11 or the keyboard 13 to set a reading item on a form image displayed on the screen and set the reading item attributes (handwritten / printed type, reading character type, etc.) A definition is created. Here, FIG. 4 is a diagram schematically showing an example of the created format definition.

  As shown in FIG. 4, in the format definition, reading items are set on the coordinates of the form image, and are subject to reading by OCR within the range specified in the reading items.

  The image file 106b stores a form image and a processed image read by the image reading unit of the input unit 112. For example, the image file 106b stores an image of a diversion destination form read by the image reading device. In addition to the image scanner 12, a digital camera, a WEB camera, or the like can be used as the input unit 112 as an image reading unit.

  The control unit 102 also has an internal memory for storing a control program such as an OS (Operating System), a program defining various processing procedures, and necessary data. And the control part 102 performs the information processing for performing various processes by these programs. In FIG. 1, the control unit 102 includes an image input control unit 102a, an overall positioning unit 102b, an item position search unit 102c, an item association unit 102d, and a conflict resolution unit 102f in terms of functional concept.

  Among these, the image input control unit 102a controls the input unit 112 to acquire a form image and store it in the image file 106b.

  The overall positioning unit 102b positions the entire images of the diversion source form and the diversion destination form.

  In addition, the item position search unit 102c searches each read item based on the format definition of the diversion source form on the diversion destination form image. Here, the item position search unit 102c may search for the reading item within a predetermined range from the position of the reading item associated with the image on the diverted advance form. For example, the item position search unit 102c, based on the positioning result of the entire diversion source form and diversion destination form image by the overall positioning unit 102b, determines the predetermined read item position on the diversion destination form image from a predetermined reading item position. The read item is searched within the range. The item position search unit 102c is not limited to this, performs character recognition in the diversion source form and diversion destination form images, and within a predetermined range from the position of the read item associated with the characters in the diversion destination form image. The read item may be searched for.

  Further, the item association unit 102d moves the read item of the diversion original form based on the search result by the item position search unit 102c. The item association unit 102d may define the result of moving the item in the format definition and reflect it in the format definition file 106a. Here, the item association unit 102d deletes the read item when the position of the corresponding read item cannot be identified from the search result by the item position search unit 102c.

  In addition, the conflict resolution unit 102f resolves the conflict when the item association unit 102d associates a plurality of diversion source reading items with each other by sharing an area. For example, the conflict resolution unit 102f resolves the conflict according to a predetermined evaluation criterion so as to meet the evaluation criterion.

  This is the end of the description of the configuration of the form definition creation apparatus 100.

[2. Processing of this embodiment]
Next, an example of processing in the present embodiment will be described in detail with reference to FIGS. 3 to 22 below. FIG. 3 is a flowchart showing an example of processing in the present embodiment.

  As shown in FIG. 3, the image input control unit 102a stores the image of the paper (form) read by the image reading unit of the input unit 112 in the image file 106b and displays it on the output unit 114 (step SA- 1).

  Then, the overall positioning unit 102b reads out the form image and the format definition as the diversion source from the format definition file 106a (steps SA-2 and SA-3).

  Then, the overall positioning unit 102b performs overall alignment so that the diversion source image overlaps the diversion destination image acquired in step SA-1 (step SA-4). Here, FIG. 5 is a diagram showing an example of a newly scanned form image, and FIG. 6 is a diagram in which the diversion-source read items are directly associated with each other on the form image of FIG. Here, in the figure, a broken line represents a reading item (the same applies to the following).

  As shown in FIG. 5 and FIG. 6, when the alignment process is performed so that the diversion source form image overlaps the diversion destination form image with reference to the diversion destination form image, the upper left corner of the image is used as a reference point. In a simple process of overlapping, as shown in FIG. 6, the reading position is shifted as a whole. Therefore, the overall positioning unit 102b corrects the overall reading position shift.

  More specifically, the overall positioning unit 102b may associate feature points extracted from both the diversion source and diversion form images, and symbols that serve as reference points such as crosshairs are printed in advance at the four corners of the form. If it is, the entire position may be aligned by detecting the symbol. For example, if the overall positioning unit 102b can associate three points, the affine transformation parameters can be obtained, so that rough positioning is possible. Further, as a simpler method, the overall positioning unit 102b may roughly correct the misalignment by the user dragging the entire image with the mouse. Here, FIG. 7 is a diagram schematically showing the result of overall positioning.

  As shown in FIG. 7, when a newly created form has been redesigned from the diverted form, there are misalignments and inconsistencies in the read items, and the form definition cannot be used as it is. That is, the position of the reading item is shifted due to the position shift of the entire paper, or the position of the reading field is shifted because the relative printing position of the frame is different. Therefore, the present embodiment performs the following processing.

  Returning to FIG. 3 again, the item position search unit 102c determines a search range for each read item based on the format definition of the diversion source form (step SA-5). The item position search unit 102c determines a search range for each read item on the image with a positional shift in the result of the alignment processing of the entire image by the overall positioning unit 102b. FIG. 8 is a diagram illustrating an example of a search range set by the item position search unit 102c.

  As shown in FIG. 8, in order to correct the position of the reading item of the format definition of the diversion source on the basis of the positioning parameter obtained by the overall positioning unit 102b, a search is performed with margins on the top, bottom, left and right. Set the range. Note that the margin value for determining the search range may be a fixed value determined in advance, or a margin for referring to the expansion / contraction rate. For example, when the overall positioning unit 102b obtains an affine transformation parameter, an approximate expansion / contraction rate can be obtained. The overall positioning unit 102 b temporarily stores the determined search range in the storage unit 106.

  Returning to FIG. 3 again, the item position search unit 102c executes a search within the search range (step SA-6). For example, the item position search unit 102c detects the position of the item by pattern matching within the search range for each item. Here, FIGS. 9 to 13 show examples of specific alignment methods. FIG. 9 is a flowchart illustrating an example of the search process in step SA-6.

  As shown in FIG. 9, first, the item position search unit 102c moves the read item within the search range of the diverted destination form image (step SA-61). FIG. 10 is a diagram schematically showing the movement of the reading item. For example, the item position search unit 102c may move the reading item to the initial position of the search range and shift the reading item by one pixel each time the process is repeated. That is, as shown in FIG. 10, the item position search unit 102c cuts out the image corresponding to the read item from the diversion source form image, superimposes it while shifting the search range, and searches for the most matching part. It may be moved by a technique.

  Returning to FIG. 9 again, the item position search unit 102c counts the number of pixels that do not match between the moved read item image and the diversion destination form image (step SA-62). Here, FIG. 11 and FIG. 12 are diagrams in which an exclusive OR (XOR) for each pixel of the diversion source read item image and the diversion destination form image is obtained at a certain position.

  The item position search unit 102c obtains an exclusive OR (XOR) for each pixel image and the diverted form image while shifting the image portion corresponding to the item. As a result, the value of the matching pixel is 0, and the value of the non-matching pixel is 1. Therefore, the degree of matching can be determined by summing the pixel values after XOR. As shown in FIG. 11, when matching is not possible at the moving position, the matching is shifted, so that the number of pixels obtained by exclusive OR is large. On the other hand, as shown in FIG. 12, when the matching is good, the images match, so the number of mismatch pixels obtained by exclusive OR becomes small.

  Returning to FIG. 9 again, the item position search unit 102c determines whether or not the number of non-matching pixels is smaller than a threshold value (step SA-63). By scanning the search range and finding a location where the number of black pixels is less than or equal to the threshold value, a matching location can be found. The threshold value may be, for example, the ratio of the number of black pixels to the area of the character frame range of the diversion destination image. Note that the two images of the diversion source and the diversion destination are different sheets, so they do not completely coincide with each other and a slight shift occurs. The part of the deviation remains as a difference, but by providing a threshold value at the upper limit of the difference, it can be considered that there is a match even if there is a slight deviation.

  When the number of mismatched pixels is larger than the threshold value (step SA-63, No), the item position search unit 102c returns the process to step SA-61, moves the item position, and performs the above-described steps SA-61 to SA-61. Repeat the process of SA-63.

  On the other hand, when the number of mismatch pixels is smaller than the threshold value (step SA-63, Yes), the item position search unit 102c determines that the matching is successful. When a plurality of character frames are included in the search range, a plurality of matching locations can be obtained by leaving a plurality of locations where the number of black pixels is equal to or less than the threshold value. At this time, it is also possible to select one having a certain distance between matching points so that the same frame is not detected twice.

  Note that the threshold value is not limited to setting in advance, and the search range may be exhaustively searched, and it may be determined that the matching is successful at the position where the number of mismatched pixels is minimized. Here, as another example of the matching method, FIG. 13 is a flowchart illustrating an example of the ruled line matching method.

  Although matching is difficult when the size of the character frame is expanded and contracted by the above-described method of FIG. 9, matching corresponding to expansion and contraction may be performed as described below with reference to FIG. Here, FIG. 13 is an example in which matching is performed using a ruled line pattern in the case of a character frame composed of ruled lines. A character frame can be found by associating each ruled line with a ruled line in a new form. In this method, by providing a range for the length of the ruled line that matches, it is easy to handle when the size is expanded or contracted.

  As shown in FIG. 13, first, the item position search unit 102c selects one of the longest vectors among the vectors constituting the frame of the diversion source form, and searches the diversion destination form for a vector close thereto (step SA). -611). The white line in the figure is the longest selected line.

  When the item position search unit 102c finds a corresponding vector, the item position search unit 102c determines a search range on the diversion destination form for the remaining vectors of the diversion source form frame with reference to this vector (step SA-612). ). Here, the broken line in FIG. 13 indicates the search range. The size of this search range is determined from the allowable range of expansion and contraction of the character frame, the upper limit of expansion and contraction of the image, and the like.

  Then, the item position search unit 102c searches the search range to determine whether a vector having a direction and length similar to the vector constituting the diversion source form frame exists (step SA-613). Finally, if the number found for the entire frame is equal to or greater than a threshold value (for example, 80%), it is considered that the two frames correspond to each other (see the technique of Japanese Patent Laid-Open No. 11-3431). The threshold value is, for example, the ratio of the number found to the number of character frame vectors in the diverted image.

  Note that which of the two methods described above is preferable is not uniquely determined, and may be selected by system setting. In addition, it is desirable to display a deletion location list to the user even when deletion is performed unconditionally. Note that the reading items that the user did not correspond to may be reset.

  Returning to FIG. 3 again, when the matching pattern of the diversion source reading item is found (step SA-7, Yes), the item association unit 102d moves the item to the matching position and changes the format definition. (Step SA-8).

  On the other hand, when the matching pattern of the diversion source read item is not found (step SA-7, No), the item association unit 102d deletes or moves the item (step SA-9). If the matching pattern is not found in the processing of the item position search unit 102c, the pattern does not match because the item has been deleted from the new form (no reading item) or the shape of the reading item has changed. (There is actually a reading item). In this case, the form definition creation apparatus 100 may delete the reading item unconditionally, or may newly define the reading item if the reading item actually exists. Here, FIG. 14 is a diagram illustrating an example in which the user associates items via the input unit 112 under the control of the item association unit 102d.

  As shown in FIG. 14, the item position search unit 102c issues a message to the user that the corresponding item is not found, and displays information about the item on the screen. The user determines the situation from the contents of the form, deletes the read item if it does not exist, and moves the read item to the correct location with a mouse or the like if it exists. That is, on this screen, items that have not been associated are listed at the edge of the screen, and the user can reset the reading items by dragging these items to the correct location on the image with the mouse. .

  Returning to FIG. 3 again, the conflict resolving unit 102f resolves the conflict when the item association unit 102d associates a plurality of diversion-source read items with a shared area (step SA-10). ). Here, a conflict resolution method will be described with reference to FIGS. FIG. 15 is a diagram illustrating an example of contention due to overlapping search ranges. The search range is indicated by brackets in the figure.

  As shown in FIG. 15, there is a case where a reading item is associated with the same position and a conflict occurs. When the item position search unit 102c determines the search range, if the distance between the items is short, the search ranges of a plurality of reading items may overlap as shown in FIG. In this case, when the shape of the reading item happens to be similar, there is a possibility that a plurality of reading items are associated with the same place. Since this is an obvious conflict, it is necessary to eliminate the duplication and make a one-to-one correspondence. Therefore, the conflict resolution unit 102f performs conflict resolution processing.

  First, it is investigated whether there is a conflicting reading item determined by the item position search unit 102c. FIG. 16 is a diagram for explaining a criterion for determining the presence or absence of competition.

  As an example, as shown in FIG. 16, the conflict resolution unit 102f determines that there is a conflict when the overlap between the read item indicated by the broken line and the read item indicated by the white outline exceeds 20% of the area of the read item. May be.

  Next, the conflict resolution unit 102f calculates an evaluation value for resolving the conflict for the conflicting reading items. Here, FIG. 17 is a diagram for explaining an evaluation criterion for resolving the conflict.

The following criteria may be set as an example of the form evaluation criteria as shown in FIG.
1) The size of the detected character frame (the closer to the diversion source, the higher the evaluation value)
2) Deviation from the position estimated by the overall positioning (the smaller the deviation, the higher the evaluation value)
3) Type near the item to be read (if the diversion source has the same type, the evaluation value is high)
4) Validity of field arrangement ("First name" item is on the left side of "First name" item, "Name" item is above "First name" field, the order of the frames in the mark item is not reversed, etc.)

  For example, the conflict resolution unit 102f may calculate an evaluation value for the reading frame that is competing based on these evaluation criteria. As a calculation method, for example, the conflict resolution unit 102f determines a threshold value for each of these four types of evaluation, and gives 1 if the threshold value is exceeded, and gives 0 if the threshold value is not exceeded or unknown. For example, the contention resolution unit 102f gives 1) if the difference between the character frame size and the diversion source is within a threshold value, and gives 1 if the positional deviation from the overall positioning is within the threshold value. May be. Note that the evaluation value is not limited to 0 and 1, but a plurality of evaluation values such as 0 to 5 can be given by having a plurality of threshold values. The conflict resolution unit 102f uses the sum of all the evaluations as the evaluation value of the read item.

  Then, the conflict resolving unit 102f determines that the reading frame having the highest evaluation value is correct after assigning all the evaluation values of the competing reading frames, and re-searches the remaining reading items corresponding to the position. . At this time, the conflict resolution unit 102f may correct the search range so as to exclude the place where the association has been determined. Here, FIG. 18 is a diagram illustrating a range excluded from the search range.

  As illustrated in FIG. 18, when the reading item is confirmed at the location “minute”, the conflict resolution unit 102 f excludes the determined reading item region from the search range of other reading items.

  The conflict resolution unit 102f performs conflict resolution processing when a new corresponding position is found by re-searching, and performs processing when the matching pattern is not found when it is not found. In this manner, the conflict resolution unit 102f can determine the reading item gradually by repeating the search and the resolution of the conflict, and obtain the final positioning result of the reading item.

  If the evaluation values are the same, the conflict resolution unit 102f cannot determine which is the correct reading item. In this case, the conflict resolution unit 102f may display a message to the user for the read item and control the user to resolve the conflict via the input unit 112. At this time, there is a method to resolve all the conflicts. Also, one reading item is determined by the user, and the remaining reading items that were in conflict are automatically re-searched on the system side, so that the correct one is obtained. A method of determining the position can also be taken.

  As another method, the conflict resolution unit 102f can select one reading item at random and make it correct. In this case, there is a possibility that the user is not positioned at the correct location. However, since the user finally confirms visually, the user can correct the correct position at that stage.

  The conflict resolving unit 102f does not search again for items with a low evaluation value, but obtains a plurality of locations corresponding to the search range in advance for each reading item and has them as candidates in descending order of the evaluation value. May be adopted. In this way, the search can be completed once, so the processing time can be shortened.

  FIG. 19 shows a data format when this method is adopted. FIG. 19 is a diagram for illustrating an example of a conflict resolution method.

  As shown in FIG. 19, each reading area on the diversion source format definition stores the search result in the search range on the diversion destination form as two pieces of information of area coordinates and evaluation values. At this time, as shown in FIG. 19, the candidates are arranged in descending order of evaluation values.

  The conflict resolution unit 102f checks whether or not there is a conflict after obtaining candidates for all the reading regions. This uses area coordinates in the data format of FIG. 19 to detect areas where the overlap of areas is equal to or greater than a threshold. If there is an overlap, add a conflict list to the candidate data. FIG. 20 is a diagram illustrating an example in which a conflict list is added. Here, the conflict is represented by the number of the reading item to be conflicted and its candidate number as shown in (5, 3). When there is no conflict, NULL is set to indicate that it is empty.

  Next, the conflict resolution unit 102f takes out one item having the highest evaluation value from the read items and refers to the conflict list. If there is no conflict, the read item is determined, and the process proceeds to the read item with the next highest evaluation value. If there is a conflict, the evaluation values are compared. If the evaluation value is higher, the reading item is determined, and a deletion flag is attached to the candidate for other reading items that compete. FIG. 21 is a diagram showing an example in which a deletion flag is added. If they are the same, the above-described processing at the same time is performed to determine the reading item.

  As shown in FIG. 21, these processes are performed for all the read items, thereby associating read items with no conflict. Here, FIG. 22 is a diagram illustrating an example in which items are further divided into entry frames and positioning is performed for each entry frame. In the case of a mark item, the interval between individual mark frames may change, but this method can be used to position the mark frame without shifting. Mark frames with the same shape are often lined up, and conflicts are likely to occur.However, because the left / right / up / down relationship of frames in the same mark item is often saved, check the left / right relationship and the up / down relationship. To resolve conflicts.

  Returning to FIG. 3 again, when the conflict is resolved as described above, the item association unit 102f determines the format definition of the diversion destination, stores it in the format definition file 106a, and causes the output unit 114 to display it. .

  Above, description of an example of the process in this Embodiment is finished.

[3. Summary of this embodiment and other embodiments]
As described above, according to the present embodiment, in the form definition creation device 100 that creates the format definition of the character recognition form, the format definition of the diversion source form is stored, and the image of the diversion destination form read by the image reading apparatus is stored. Then, individual reading items based on the format definition of the diversion source form are searched on the image of the diversion destination form, and the reading items of the diversion source form are moved based on the search result. Thereby, the already created form definition can be appropriately used. More specifically, since positioning is performed using only the pattern of the reading item, positioning can be performed following even if the reading item is moving. As a result, it is possible to reduce the time and effort for the user to move the reading item by hand.

  Further, according to the present embodiment, since the entire image of the diversion source form and the diversion destination form is positioned before the item position search process is performed, the positional deviation is eliminated and the overall positioning is performed. be able to.

  Further, according to the present embodiment, since the search for the read item is performed within a predetermined range from the position of the corresponding read item on the diversion destination form image, the format can be efficiently formed by limiting the search range. Definitions can be diverted.

  Further, according to the present embodiment, based on the positioning result of the entire diversion source form and diversion destination form image by the overall positioning, within the predetermined range from the position of the corresponding read item on the diversion destination form image The search for the reading item is performed. Thereby, this embodiment can search efficiently by making the rough position by whole positioning into a starting point.

  In addition, according to the present embodiment, when a plurality of diversion source reading items are associated with each other by sharing an area by the item association, the conflict is resolved. Therefore, the present embodiment has a conflict. Can be detected with high accuracy. That is, by moving or deleting one or more of the conflicting reading areas, the positions of the remaining conflicting reading items can be automatically rearranged based on the evaluation value.

  Further, according to the present embodiment, according to a predetermined evaluation standard, the conflict is resolved so as to meet the evaluation standard, so that a suitable form definition along the evaluation standard can be created.

  Further, according to the present embodiment, when the position of the corresponding reading item cannot be specified by the search result of the item position, the reading item is deleted, so that the unnecessary reading item is automatically determined. Can be deleted.

  Furthermore, although the embodiment of the present invention has been described so far, the present invention is not limited to the above-described embodiment, but can be implemented in various different embodiments within the scope of the technical idea described in the claims. It may be implemented.

  For example, the case where the form definition creation apparatus 100 performs processing in a stand-alone form has been described as an example. However, the form definition creation apparatus 100 performs processing in response to a request from a client terminal configured in a separate housing from the form definition creation apparatus 100, You may comprise so that the process result may be returned to the said client terminal.

  In addition, among the processes described in the embodiment, all or part of the processes described as being automatically performed can be performed manually, or the processes described as being performed manually can be performed. All or a part can be automatically performed by a known method.

  In addition, unless otherwise specified, the processing procedures, control procedures, specific names, information including registration data for each processing, parameters such as search conditions, screen examples, and database configurations shown in the above documents and drawings Can be changed arbitrarily.

  In addition, regarding the form definition creation device 100, each illustrated component is functionally conceptual and does not necessarily need to be physically configured as illustrated.

  For example, regarding the processing functions provided in each device of the form definition creation device 100, in particular, each processing function performed by the control unit 102, all or any part thereof is interpreted by a CPU (Central Processing Unit) and the CPU. It may be realized by a program to be executed, or may be realized as hardware by wired logic. The program is recorded on a recording medium to be described later, and is mechanically read by the form definition creation device 100 as necessary. That is, a storage unit 106 such as a ROM or a HDD (Hard Disk Drive) stores a computer program that cooperates as an OS (Operating System) to give instructions to the CPU and perform various processes. This computer program is executed by being loaded into the RAM, and constitutes the control unit 102 in cooperation with the CPU.

  The computer program may be stored in an application program server connected to the form definition creation device 100 via an arbitrary network, and may be downloaded in whole or in part as necessary. It is.

  The program according to the present invention can also be stored in a computer-readable recording medium. Here, the “recording medium” is an arbitrary “portable physical medium” such as a flexible disk, magneto-optical disk, ROM, EPROM, EEPROM, CD-ROM, MO, DVD, Blu-ray (registered trademark) Disc, etc. Or a “communication medium” that holds the program in a short period of time, such as a communication line or carrier wave in the case of transmitting the program via a network represented by LAN, WAN, or the Internet.

  The “program” is a data processing method described in an arbitrary language or description method, and may be in any format such as source code or binary code. The “program” is not necessarily limited to a single configuration, but is distributed in the form of a plurality of modules and libraries, or in cooperation with a separate program represented by an OS (Operating System). Including those that achieve the function. Note that a well-known configuration and procedure can be used for a specific configuration for reading a recording medium, a reading procedure, an installation procedure after reading, and the like in each device described in the embodiment.

  Various databases and the like stored in the storage unit 106 are storage means such as a memory device such as a RAM, SSD, and ROM, a fixed disk device such as a hard disk, a flexible disk, and an optical disk. Stores programs, tables, databases, web page files, etc.

  Further, the form definition creation device 100 is connected to an information processing device such as a known personal computer or workstation, and software (including a program, data, etc.) for realizing the method of the present invention is installed in the information processing device. May be realized.

  Furthermore, the specific form of distribution / integration of the devices is not limited to that shown in the figure, and all or a part of them may be functional or physical in arbitrary units according to various additions or according to functional loads. Can be distributed and integrated.

  As described above in detail, according to the present invention, it is possible to provide a form definition creation device, a form definition creation method, and a program that can appropriately divert already created form definitions. This is extremely useful when implemented in the field of image processing relating to an image read by a scanner.

DESCRIPTION OF SYMBOLS 10 Computer 11 Mouse | mouth 12 Image scanner 13 Keyboard 14 Display 100 Form definition creation apparatus 102 Control part 102a Image input control part 102b Whole positioning part 102c Item position search part 102d Item matching part 102f Conflict resolution part 106 Storage part 106a Format definition file 106b Image file 108 Input / output control interface unit 112 Input unit 114 Output unit 104 Communication control interface unit 200 External device

Claims (10)

A form definition creation device comprising at least a storage unit and a control unit for creating a format definition for a character recognition form,
The storage unit
Format definition storage means for storing the format definition of the diversion ledger,
Image storage means for storing an image of the diverted advance form read by the image reading device;
With
The control unit
Item position search means for searching for individual reading items based on the format definition of the diversion source form on the image of the diversion destination form,
Based on the search result by the item position search means, an item association means for moving the read item of the diversion original form,
Before performing the process according to the item position search unit, the overall positioning means for positioning the entire image of the diverted destination form a diversion original form,
Form definition creating apparatus comprising a. In the form definition creation device according to claim 1 ,
The item position search means includes:
A form definition creation device that searches for a read item within a predetermined range from the position of a corresponding read item on an image of a diversion destination form. In the form definition creation device according to claim 1 ,
The item position search means includes:
Based on the positioning result of the entire diversion source form and diversion destination form image by the overall positioning unit, the reading item is searched within a predetermined range from the position of the corresponding reading item on the diversion destination form image. Form definition creation device to perform. A form definition creation device comprising at least a storage unit and a control unit for creating a format definition for a character recognition form,
The storage unit
Format definition storage means for storing the format definition of the diversion ledger,
Image storage means for storing an image of the diverted advance form read by the image reading device;
With
The control unit
Item position search means for searching for individual reading items based on the format definition of the diversion source form on the image of the diversion destination form,
Based on the search result by the item position search means, an item association means for moving the read item of the diversion original form,
The above list correlating means, when a plurality of read items diverted source, associated share regions each other, a conflict resolution means for resolving the conflict,
Form definition creating apparatus comprising a. In the form definition creation device according to claim 4 ,
The above conflict resolution means
A form definition creation device that resolves a conflict so as to meet a predetermined evaluation standard. In the form definition creation device according to any one of claims 1 to 5 ,
The item association means
A form definition creation device that deletes a read item when the position of the corresponding read item cannot be identified from the search result by the item position search means. A form definition method that is executed in a form definition creation device that includes at least a storage unit and a control unit to create a format definition of a form for character recognition,
The storage unit
Format definition storage means for storing the format definition of the diversion ledger,
Image storage means for storing an image of the diverted advance form read by the image reading device;
With
Executed in the control unit,
An item position search step for searching for individual reading items based on the format definition of the diversion source form on the image of the diversion destination form;
Based on the search result in the item position search step, an item association step for moving the read item of the diversion source form,
Before performing the processing of the item position search step, an overall positioning step for positioning the entire image of the diversion source form and the diversion destination form,
A form definition creation method characterized by including: A form definition method that is executed in a form definition creation device that includes at least a storage unit and a control unit to create a format definition of a form for character recognition,
The storage unit
Format definition storage means for storing the format definition of the diversion ledger,
Image storage means for storing an image of the diverted advance form read by the image reading device;
With
Executed in the control unit,
An item position search step for searching for individual reading items based on the format definition of the diversion source form on the image of the diversion destination form;
Based on the search result in the item position search step, an item association step for moving the read item of the diversion source form,
In the item association step, when a plurality of diversion source reading items are associated with each other by sharing an area, a conflict resolving step for resolving the conflict,
A form definition creation method characterized by including: A program for creating a form definition for a character recognition form and causing a form definition creation apparatus having at least a storage unit and a control unit to execute the format definition,
The storage unit
Format definition storage means for storing the format definition of the diversion ledger,
Image storage means for storing an image of the diverted advance form read by the image reading device;
With
In the control unit,
An item position search step for searching for individual reading items based on the format definition of the diversion source form on the image of the diversion destination form;
Based on the search result in the item position search step, an item association step for moving the read item of the diversion source form,
Before performing the processing of the item position search step, an overall positioning step for positioning the entire image of the diversion source form and the diversion destination form,
A program for running A program for creating a form definition for a character recognition form and causing a form definition creation apparatus having at least a storage unit and a control unit to execute the format definition,
The storage unit
Format definition storage means for storing the format definition of the diversion ledger,
Image storage means for storing an image of the diverted advance form read by the image reading device;
With
In the control unit,
An item position search step for searching for individual reading items based on the format definition of the diversion source form on the image of the diversion destination form;
Based on the search result in the item position search step, an item association step for moving the read item of the diversion source form,
In the item association step, when a plurality of diversion source reading items are associated with each other by sharing an area, a conflict resolving step for resolving the conflict,
A program for running

Images