JP3540192B2 - Method and apparatus for checking and correcting optical character reading paper recognition result - Google Patents

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
SUMMARY OF THE INVENTION The present invention relates to an OCR paper recognition result confirmation and correction method that efficiently confirms and corrects a mark recognition result, a character recognition result, and a graphic area detection result performed by an optical character reading (hereinafter, abbreviated as OCR (Optical Character Reader)) paper recognition device. The invention relates to a method and an apparatus for implementing the method.
[0002]
[Prior art]
Conventionally, confirmation and correction of mark and character recognition results in an OCR paper recognition device are performed by displaying the mark and character recognition results on a display screen, referring to the input OCR paper for confirmation and correcting by keyboard input, and a display screen. In this method, the recognition result and the OCR paper image are displayed in a divided manner, checked while referring to the screen, and corrected by keyboard input.
[0003]
For this reason, in the method of referring to the OCR paper, the line of sight changes between the character position in the OCR paper and the position of the recognition result on the display screen, so that the efficiency is low and the handling of a large amount of OCR paper is complicated. The method of displaying the OCR paper recognition result and the OCR paper image separately on the display screen is simple because the OCR paper is not required. However, when the entire image of the OCR paper is displayed on the split screen of the display, a small OCR paper is displayed. Although OCR paper of A4 size or the like was good, not only was it difficult to read the character image, but also the correspondence between the character recognition result and the character image was difficult. For this reason, it is necessary to enlarge and display the OCR paper image, but this requires frequent scrolling of the screen.
[0004]
As described above, there is a problem that the work of confirming and correcting the recognition result of the OCR paper is not only complicated and time-consuming, but also burdens the operator who performs the confirmation and correction.
[0005]
[Problems to be solved by the invention]
The present invention is to simultaneously display the recognition result of the OCR paper and the enlarged image of the OCR paper image in conjunction with the display screen, and to clarify the image and field of the mark / character / graphic to be confirmed or corrected. Thus, the above problem is solved, and the task of confirming and correcting the recognition result of the OCR paper is made more efficient.
[0006]
[Means for Solving the Problems]
In order to solve such a problem, a method for confirming and correcting an OCR paper recognition result according to the present invention displays an OCR paper recognition result in a recognition result display area of a display screen, and also displays one of the OCR paper image recognition results. Displaying the corresponding image portion corresponding to the above in the image display area of the display screen, and instructing any of the recognition results in the recognition result display region, the corresponding image portion corresponding to the specified recognition result in the OCR paper image is displayed. A step of enlarging and displaying the image at the center or a predetermined position of the image display area; and a step of adding a confirmation mark to a corresponding image portion corresponding to the specified recognition result and displaying the image.. A step of displaying the recognition result in the recognition result display area and displaying a logical error of the recognition result in the logical error display area of the display screen; and a step of displaying a corresponding image portion corresponding to the designated logical error display And adding a confirmation mark to the display.
[0007]
Preferably, the recognition result is displayed in the recognition result display area., FigureThe shape area detection resultAlso deDisplay screenIllustrationThe process of displaying in the shape image display area and,fingerDisplaying the indicated graphic area detection result;ToHas a process to add and display a confirmation markYou.
[0008]
The OCR paper recognition result confirmation and correction device according to the present invention displays the OCR paper recognition result in a recognition result display area of a display screen and displays a corresponding image portion corresponding to one of the OCR paper images. Means for displaying the image in the image display area of the screen, and specifying one of the recognition results in the recognition result display area, the corresponding image portion corresponding to the specified recognition result in the OCR paper image is displayed at the center of the image display area or in a specified area. Means for enlarging and displaying the position and means for adding and displaying a confirmation mark on a corresponding image portion corresponding to the specified recognition result. Means for displaying the recognition result in the recognition result display area and displaying the logical error of the recognition result in the logical error display area of the display screen, and means for displaying a corresponding image portion corresponding to the designated logical error display And a means for adding and displaying a confirmation mark.
[0009]
Preferably, the recognition result is displayed in the recognition result display area., FigureThe shape area detection resultAlso deDisplay screenIllustrationMeans for displaying in the shape image display area and,fingerMeans for displaying the indicated graphic area detection result;ToHas a means to add and display a confirmation markYou.
[0010]
Claim 1,3According to the invention described in (1), the OCR paper recognition result and the OCR paper image are displayed on the same display screen, and when the character or mark recognition result on the display screen is designated, the corresponding image portion of the OCR paper image is centered on a predetermined display area. Alternatively, the mark / character image to be confirmed and corrected is clarified by enlarging and displaying the prescribed position and adding a confirmation mark. Also, OThe recognition result of the CR paper and the logical error of the mark / character recognition result for each field are displayed on the same display screen, and when a logical error display is instructed, the corresponding image portion of the specified OCR paper image is displayed in a predetermined manner. The field to be confirmed and corrected is clarified by enlarging and displaying it at the center of the area or at a predetermined position and adding a confirmation mark. ClaimsTwo, fourAccording to the invention described in (1), the recognition result of the OCR paper and the graphic image are displayed on the same display screen, and when a graphic area confirmation instruction is issued, the specified graphic image (graphic area confirmation result) of the OCR paper image is converted to a predetermined image. The graphic image and the field to be confirmed are clarified by being displayed at the center or a prescribed position of the display area and adding a confirmation mark. As described above, the present invention is significantly different from the prior art in that the image and field of the mark, character, and graphic to be confirmed can be clarified by the operation on the same screen, so that the confirmation and correction work can be made much more efficient.
[0011]
In the claims, the recognition result refers to the mark / character recognized by the OCR paper recognition device for each mark / character described on the input OCR paper, and the OCR paper image is described on the input OCR paper. Refers to the image information recognized by the OCR paper recognition device with respect to the input information, and the graphic region detection result refers to the image information recognized by the OCR paper recognition device with respect to the graphic described in each graphic region of the input OCR paper. Say. The corresponding image portion corresponding to the recognition result refers to image information on each mark / character on which the recognition result is based in the OCR paper image, and the corresponding image portion corresponding to the logical error display is the OCR paper image. It refers to image information on each mark / character on which a logical error has been detected in a paper image or a mark / character group belonging to a field where a logical error has been detected. Further, the confirmation mark is a mark for making a corresponding image portion to be confirmed / corrected stand out. Claim 1And 2In the step, each step may be performed separately or may be performed simultaneously. Claim3 and 4With respect to the means, each means may be independent or a part or the whole may be shared. Claims3 and 4The OCR paper recognition result confirmation and correction device described in (1) may be independent of the OCR paper recognition device, or may be partially or entirely shared.

[0012]
BEST MODE FOR CARRYING OUT THE INVENTION
FIG. 1 shows a procedure of an embodiment of a method for confirming and correcting an OCR paper recognition result according to the present invention. First, as an initial screen, a recognition result display window W on the display 6 screen is displayed.₁Display processing (step A1), a logical error display window W_TwoDisplay processing (step A2), mark / character image display window W_ThreeDisplay processing (step A3), graphic image display window W_FourIs performed (step A4). Recognition result display window W which is a recognition result display area₁In the display process, the recognition result of the OCR paper is displayed. Logical error display window W which is a logical error display area_TwoIn the display process, the logical error of the recognition result is displayed for each field. Mark / character image display window W which is the image display area_ThreeIn the display processing of, the corresponding image portion corresponding to the recognition result where the cursor is present is marked / character image display window W_ThreeIs enlarged and displayed at the center or the specified position. Graphic image display window W which is a graphic image display area_FourIn the display process, the designated graphic area detection result is displayed. First, window W_Three, W_FourDisplays a corresponding image corresponding to one recognition result at the top of the OCR paper and one graphic area detection result. If there is no logical error, a logical error display window W is displayed._TwoDoes not display anything.
[0013]
Next, the recognition result display window W₁Or logic error display window W_TwoIs confirmed / corrected by the logic error message (step A5). Recognition result display window W₁When one of the recognition results is specified by the user, the corresponding image portion corresponding to the specified recognition result is displayed in the mark / character image display window W._ThreeIs enlarged and displayed at the center or the specified position. In addition, the logic error display window W_TwoWhen one of the logical error displays is instructed, the corresponding image portion corresponding to the indicated logical error display is displayed in the mark / character image display window W._ThreeIs enlarged and displayed at the center of the image display area or at a predetermined position. Further, a confirmation mark is added to these corresponding image portions and displayed.
[0014]
If the recognition result has been corrected (step A6), the logical error display window W_Two(Step A7), and a logic error display window W_TwoConfirm that there is no longer a logical error message at the point of correction. This confirmation / correction work is performed for all recognition results (step A8). Next, the recognition result display window W₁The graphic image display window W is displayed while clicking the graphic button, which is the graphic region confirmation instruction area, with the mouse._FourTo confirm the entry figure (signature, seal) (step A9). When the figure area confirmation instruction is issued, the designated figure area detection result is displayed in the figure image display window W._FourIs displayed at the center or the specified position. A confirmation mark is added to the graphic area detection result and displayed. This is confirmed for all graphic images (step A10). With such a procedure, the OCR paper image can be checked and corrected. In the procedure of the present invention, after confirming and correcting the recognition result of the mark / character, the graphic image of the signature and the seal is confirmed, but the order is not limited.
[0015]
FIG. 2 is a block diagram showing a configuration of an embodiment of an apparatus for performing a method according to the present invention. 1 is an image storage unit for storing an OCR paper image, 2 is a recognition result storage unit for storing an OCR paper recognition result, and 3 is logic check information (a number of digits and a range of numbers for a field definition) stored in a definition information storage unit 4. , A rejection character, etc.), a logic error detection unit for performing a logic check of the OCR paper recognition result based on the OCR paper recognition result, and 4 describes the OCR paper recognition conditions (field definition and recognition position and character type related to the field, logic check information, etc.). A definition information storage unit for storing the obtained definition information; 5, an overall control unit for controlling the entire system; 6, a display for displaying OCR paper recognition results and OCR paper images; and 7, a OCR paper recognition result displayed on a display 6 screen. A keyboard 8 for inputting the correction data of the OCR paper 8 indicates the OCR paper recognition result displayed on the screen of the display 6. It is a mouse.
[0016]
FIG. 3 is a diagram showing an example of an OCR paper image. 10 is an OCR paper image, 11 and 12 are control marks indicating the left and right of a mark or a character line, 13 is a mark entry frame showing an area where a mark is to be entered, and 14 is a character. , A character entry frame indicating an area where a figure is to be entered, and 15 and 16 are figure entry frames indicating an area where a figure is to be entered.₁Is the first letter of the customer ID, I_TwoIs an entry mark indicating immediate delivery, I_ThreeIs the character to be corrected in the product name, I_FourIs the signature entered in the figure box,_FiveIs the seal stamped on the graphic entry box. However, the control marks 11 and 12, the mark entry frame 13, the character entry frame 14, and the graphic entry frames 15 and 16 are arranged with specified dimensions that can recognize the OCR sheet.₁, Entry mark I_Two, Character I to be corrected_Three, Signature I_Four, Seal I_FiveIs a color that can be sensed by a black or facsimile sensor, and the mark entry frame 13, character entry frame 14, graphic entry frames 15, 16 are described in a dropout color (a color that is not detected by the facsimile sensor). The fields in this embodiment are "customer ID", "product code", "immediate delivery", "product name", "signature", and "seal".
[0017]
FIG. 4 is a diagram showing an initial screen displayed on a display according to the present invention.₁Is a recognition result display window for displaying the recognition result of the OCR paper image, R₁Is the letter I₁Recognition result, R_TwoIs mark I_TwoRecognition result, R_ThreeIs the character I to be corrected_ThreeRecognition result, B₁Is a graphic display button (corresponding to a graphic area confirmation instruction area) for instructing display of information described in the next graphic entry frame;_TwoIs the recognition result R of the OCR paper image₁, A logical error display window for displaying a logical error for each field₁Is a scroll bar that scrolls through the messages in the logic error display window.₁Is the customer ID logic error message, E_TwoIs the product name logic error message, B_TwoIs a logic error detection button for instructing a logic check, W_ThreeIs the recognition result display window W₁Mark / character image display window for displaying the mark / character image in the OCR paper image 10 corresponding to the recognition result of the mark / character in the inside, S_TwoIs the mark / character image display window W_ThreeScroll bar for scrolling the image displayed on B_ThreeIs the mark / character image display window W_ThreeDisplay magnification specification box that allows the user to specify the display magnification of F₁Is the recognition result display window W₁The recognition result of the mark / character indicated by the mouse 8 (or the cursor) in the inside, or the logical error display window W_TwoA mark surrounding the character entry frame 14 with a mark / character confirmation / correction position mark which is a confirmation mark for highlighting a corresponding image corresponding to a recognition result of a mark / character having a logic error designated by the mouse 8 (or the cursor) therein. It is displayed as a frame. W_FourIs the figure entry frame I of the OCR paper image 10_FiveOr I₆Graphic image display window for displaying a graphic image (corresponding to a graphic area detection result) written in the box, S_ThreeIs a scroll bar for scrolling the image displayed in the graphic image display window, F_TwoIs a figure confirmation position mark, which is a confirmation mark that makes the figure area detection result corresponding to the figure entered in the figure entry frame to be confirmed stand out, and is displayed as a thick frame surrounding the figure entry frames 15 and 16.
[0018]
5 and 6 are diagrams showing a method of checking and correcting an OCR recognition result displayed on the display 6 according to the present invention. FIG. 5 is a diagram showing a display screen before correction, and FIG. 6 is a display after a partial correction. It is a figure showing a screen. I₁~ I_Five, W₁~ W_Four, R₁~ R_Three, B₁~ B_Three, S₁~ S_Three, E₁~ E_Two, F₁~ F_TwoIs similar to FIG. W_FiveIs the recognition result display window after character correction, W₆Is a logic error display window after logic check, W₇Is the character I to be corrected_ThreeMark / character image display window with W₈Is Seal I_FiveImage display window displaying the graphic image (graphic area detection result) of R_Three’Is the recognition result R_ThreeModified letter of P₁Is the character I to be corrected_ThreeClick position of mouse 8 to indicate_TwoIs the logical error message E_TwoClick position of mouse 8 to indicate_ThreeIs a click position of the mouse 8 for designating a graphic button.
[0019]
Next, the operation of FIG. 2 will be described with reference to FIGS. The procedure of the operation is shown in FIG.
[0020]
[Step A1] First, the overall control unit 5 reads the recognition result of the mark and the character from the recognition result storage unit 2. Then, a recognition result display window W prepared in advance₁Is used to display the mark / character recognition result on the upper left portion of the display 6.
[0021]
[Step A2] Next, the overall control unit 5 instructs the logic error detection unit 3 to perform a logic check. Upon receiving the instruction of the logic check, the logic error detection unit 3 reads the logic check information for each field stored in the definition information storage unit 4 and, based on the logic check information, stores the mark and the character stored in the recognition result storage unit 2. Perform a logic check of In the present embodiment, it is detected that the customer ID is not a five-digit number and that there is an unrecognizable character in the product name. Then, the logic error detection unit 3 stores the logic check result in the recognition result storage unit 2 and notifies the overall control unit 5 of the completion of the logic check. Upon receiving the logic check completion notification, the overall control unit 5 reads the logic check result from the recognition result storage unit 2 and prepares a logic error display window W prepared in advance._TwoTo display the logic error result in the lower left part of the display 6. In this embodiment, "Customer ID is a five-digit number" and "Product name is incorrect" are displayed.
[0022]
[Step A3] Next, the overall control unit 5 reads the coordinates and size of the character entry frame 13 corresponding to the head of the recognition result from the recognition result storage unit 2, and reads out the OCR paper image 10 stored in the image storage unit 1. First letter I₁Is read including the peripheral image of. Then, a mark / character image display window W prepared in advance_ThreeIs displayed on the upper right screen of the display 6 using the character I₁It is enlarged and displayed including the surrounding image of. At this time, the overall control unit 5₁Mark / character image display window W_ThreeAt the center of the character₁Mark and character confirmation correction position mark F₁Are displayed in layers.
[0023]
[Step A4] Next, the overall control unit 5 reads out the image coordinates and the size of the graphic entry frame 15 at the head of the recognition result from the recognition result storage unit 2, and reads out the OCR sheet image 10 stored in the image storage unit 1. Filling signature I_FiveIs read including the peripheral image of. Then, a previously prepared graphic image display window W_Four, Enter the signature I on the lower right screen of the display 6_FiveDisplays the message including the surrounding area. At this time, the overall control unit 5 writes the signature I_FourTo display a graphic image window W_FourAnd place the signature I_FourConfirmation position mark F on the image_TwoAre displayed in layers.
[0024]
[Step A5] By such a procedure, the initial screen of the display 6 shown in FIG. 4 can be displayed. However, at this time, the cursor position on the display 6 is the recognition result display window W₁In the recognition result at the beginning. From this state, the operator operates the cursor key of the keyboard 7 or the mouse 8 to display the recognition result display window W.₁When the recognition results are sequentially indicated, the corresponding mark or character image is displayed in the mark / character display window W._ThreeIt is displayed enlarged and can be checked. In addition, mark / character image display window W_ThreeAnd graphic image display window W_FourThe default magnification of the image displayed in is set to 100%._ThreeCan be changed by clicking with the mouse 8 and inputting a numerical value from the keyboard 7.
[0025]
[Step A6] The recognition result of the mark / character can be confirmed in this manner. Then, a recognition result display window W in which the cursor of the keyboard 7 exists.₁The recognition result of the mark or the character can be corrected from the keyboard 7 at that time. Specifically, the correction target character I_ThreeCan be corrected by using the cursor key of the keyboard 7 or the mouse 8 to display the recognition result display window W.₁Character I to be corrected in the recognition result of_ThreePosition (mouse click position P₁) Or a logic error display window W_TwoError message E₁Or E_TwoPosition (mouse click position P_Two) And click. Then, the overall control unit 5 sets the mouse click position P on the display 6.₁Or P_TwoOf the OCR paper image 10 corresponding to the recognition result specified from the display 8 coordinates of the OCR paper image 10 is read from the recognition result storage unit 2 and the correction target character I is read from the image storage unit based on this value._ThreeIs read out and the mark / character image display window W on the display 6 screen is read out.₇To enlarge. At this time, the overall control unit 5 sets the correction target character I_ThreeMark / character image display window W₇Character I to be corrected_ThreeMark and character confirmation correction position mark on the image of_F1Are displayed in layers. Then, the recognition result display window W₁Recognition result R_ThreeModified character R from keyboard 7_Three’(Letter“ ko ”). Then, the correction character R input from the keyboard 7_Three'(Character "") is read by the overall control unit 5 and the recognition result at the corresponding position in the recognition result storage unit 2 is rewritten. At the same time, the recognition result display window W on the display 6 screen_FiveModified letter R_Three’(Character“ 「”) is displayed again.
[0026]
[Step A7] Then, a logic error display window W_TwoLogic error detection button B_TwoIs clicked with the mouse 8 and clicked. Then, similarly to the method described in the initial screen display of the display 6, the general control unit 5 issues an instruction to the logical error detecting unit 3, stores the logical check result in the recognition result storage unit 2, and sends a logical check end notification to the general control. Notify section 5. Upon receiving the logical check end notification, the overall control unit 5 reads the logical check information from the recognition result storage unit 3 and displays the logical error display window W on the lower left screen of the display 6.₆Is displayed again. As a result, the logical error display screen W₆Error message E of product name_TwoIt can be seen that has disappeared.
[0027]
[Step A8] Recognition result display window W in the same procedure₁The customer ID of 5 digits, and the logical error detection button B_TwoIs clicked with the mouse 8, the customer ID logical error message E₁Can be turned off. Correction and logic check are sequentially performed for each mark, character, and logic error, thereby completing confirmation and correction.
[0028]
[Step A9] When confirming the figures to be entered in the figure entry frames 15 and 16, the recognition result display window W₁Graphic button B in₁(Figure area confirmation instruction area) position (mouse click position P_Three) Is designated with the mouse 8 and clicked on to enter the next figure (stamp I_FiveImage) to display the graphic image window W₈And can be checked. However, the overall control unit 5_FiveImage W of graphic image display window W₈In the center of the figure, and the figure position confirmation mark F_TwoAre displayed in layers. Thus, the correction target character I_ThreeConfirmation and correction of seal I_FiveFIG. 6 shows a display screen on which the graphic image has been confirmed.
[0029]
[Step A10] By sequentially performing correction and logic check on each graphic image, the confirmation correction is completed.
[0030]
In this embodiment, the mark / character image display window W_Three, W₇, Graphic image display window W_Four, W₈Was not scrolled, but the scroll bar S_Two, S_ThreeThe mouse₈By clicking on, it is possible to see adjacent entry marks, entry characters, figures, explanations, and the like. In addition, the logic error display window W_Two, W₆Scroll bar S₁Need to be sequentially displayed while clicking with the mouse 8. Furthermore, mark / character image display window W_Three, W₇, And graphic image display window W_Four, W₈If the display magnification of the image displayed on the OCR paper image is not appropriate (in this embodiment, the display image read from the OCR paper image 10 is not converted in size because the display magnification is 100%), the display magnification designation box B_ThreeIs clicked with the mouse 8 and a numerical value is input from the keyboard 7. Then, the overall control unit 5 interprets the numerical value from the keyboard 7, calculates the image size to be displayed, reads out and displays the image. The mark / character correction confirmation position mark F which is a confirmation mark according to the display magnification.₁Or figure confirmation position mark F_TwoIs calculated and displayed. In addition, as a logic check method, it is possible to include a survey of character types and a range of numbers, and it is also possible to collate a product name with a product name database. The image displayed in the mark / character display window and the confirmation mark to be superimposed as a result of the logical check can be displayed not only in the mark / character but also in the field.
[0031]
【The invention's effect】
According to the method and apparatus for confirming and correcting the OCR paper recognition result of the present invention, as described above, the mark and character recognition result and the enlarged OCR paper image are displayed on the screen of the same display device, and the characters or marks on the display screen are displayed. When the recognition result is indicated, the corresponding image portion of the OCR paper image is linked and enlarged to the center of the display area or a specified position and displayed, and a check mark is added, so that the mark / character image to be checked and corrected can be clarified. Therefore, there is an advantage that the operation of confirming and correcting the OCR paper recognition result can be efficiently performed without imposing a burden on the operator. In addition, the logical error between the mark and character recognition result and the mark / character recognition result for each field is displayed on the same screen, and when a logical error display is instructed, the corresponding image portion of the specified OCR paper image is linked and displayed. By enlarging the display at the center of the area or at the specified position and adding a confirmation mark, it is possible to clarify the field to be confirmed and corrected, not only to correct the mark and character recognition but also to confirm the meaning of the field There are advantages that can be reduced. In addition, the mark and character recognition result and the graphic image are displayed on the same screen, and when the graphic area confirmation instruction is given, the specified graphic area detection result is displayed at the center of the display area or at a prescribed position, and a confirmation mark is added. Since the graphic image and the field to be confirmed can be clarified, there is an advantage that the graphic area in which the signature or the like is described can be easily confirmed. As described above, according to the present invention, it is possible to clarify the image and field of the mark / character / graphic on the same screen, and there is an advantage that the operation of confirming and correcting the recognition result of the OCR paper can be efficiently performed.
[Brief description of the drawings]
FIG. 1 is a diagram showing a procedure of an embodiment of a method of confirming and correcting an OCR paper recognition result according to the present invention.
FIG. 2 is a block diagram showing a configuration of an embodiment of an apparatus for performing a method according to the present invention.
FIG. 3 is a diagram illustrating an example of an OCR paper image according to the present invention.
FIG. 4 is a diagram showing an initial screen displayed on a display according to the present invention.
FIG. 5 is a diagram showing a display screen before correction according to the present invention.
FIG. 6 is a diagram showing a display screen after a partial correction in the present invention.
[Explanation of symbols]
1. Image storage
2 ... Recognition result storage
3. Logic error detector
4: Definition information storage
5: Overall control unit
6. Display
7 ... Keyboard
8 ... Mouse
10 ... OCR paper image
11, 12 ... control mark
13 ... mark entry frame
14 ... Character entry frame
15, 16 ... figure entry frame
A1 to A10: steps of the method of checking and correcting the OCR paper recognition result
I₁… First entry character
I_Two… Entry mark
I_Three… Correction target character
I_Four… Signature
I_Five… Stamp
W₁, W_Five… Recognition result display window
W_Two, W₆… Logic error display window
W_Three, W₇… Mark / character image display window
W_Four, W₈… Figure image display window
R₁… First letter I₁Recognition result
R_Two… Entry mark I_TwoRecognition result
R_Three… Correction letter I_ThreeRecognition result
R_Three'... Recognition result R_ThreeCorrection character for
B₁… Figure display button
B_Two… Logic error detection button
B_Three… Display magnification specification box
E₁... Logical error message of customer ID
E_Two… Product name logic error message
F₁… Mark / character confirmation correction position mark
F_Two… Figure confirmation position mark
S₁, S_Two, S_Three…scroll bar
P₁, P_Two, P_Three… Mouse click position

Claims (4)

Displaying the recognition result of the optical character reading paper in the recognition result display area of the display screen, and displaying a corresponding image portion corresponding to one of the recognition results of the optical character reading paper image in the image display area of the display screen; When one of the recognition results in the recognition result display area is specified, the corresponding image portion corresponding to the specified recognition result in the optical character reading paper image is enlarged to the center of the image display area or a specified position. Displaying, displaying a confirmation mark on the corresponding image portion corresponding to the instructed recognition result, displaying the recognition result in the recognition result display area, and detecting a logical error in the recognition result. Detecting each field and displaying it in a logical error display area of the display screen; When the display is instructed, a step of enlarging and displaying a corresponding image portion of the optical character reading paper image corresponding to the indicated logical error display at a center or a predetermined position of the image display area; A method for confirming and correcting an optical character reading paper recognition result, comprising a step of adding a confirmation mark to a corresponding image portion corresponding to an error display and displaying the result. Displaying the recognition result and the graphic region confirmation instructing region in the recognition result display region, and displaying one graphic region detection result of the optical character reading paper image in the graphic image display region of the display screen; , when the graphic region confirmation instruction, and table Shimesuru step to the position of the center or the provisions of the indicated graphic region detection result figure type image display area of the optical character reading paper image, finger indicated graphic region detected 請 Motomeko 1 optical character reader sheet recognition result how to verify modifications described you and a step of displaying by adding a check mark in the results. And displays a recognition result of the optical character reading paper in the recognition result display area of the display screen and displays the corresponding image portion corresponding to a recognition result of the light Studies character reading paper image Image display area of the display screen Means for indicating one of the recognition results in the recognition result display area, enlarging the corresponding image portion of the optical character reading paper image corresponding to the specified recognition result to the center of the image display area or a predetermined position. Means for displaying and displaying a confirmation mark on a corresponding image portion corresponding to the instructed recognition result, and displaying the recognition result in the recognition result display area and logic of the recognition result. Means for detecting an error for each field and displaying the error in a logical error display area of the display screen; Means for enlarging and displaying a corresponding image portion of the optical character reading paper image corresponding to the specified logical error display at the center of the image display area or at a specified position; light Science character reading paper recognition result of the confirmation correction device you further comprising a means for displaying by adding a check mark in the appropriate image section corresponding to the error indication. And the recognition result and the figure region confirmation instruction area and displays on the recognition result display area, and means for displaying one of a graphic region detection result to the graphic image display area of the display screen of the optical character reading paper image , when the graphic region confirmation instruction, and table Shimesuru means the position of the center or the provisions of the indicated graphic region detection result figure type image display area of the optical character reading paper image, finger indicated graphic region detected 4. The apparatus according to claim 3, further comprising means for adding a confirmation mark to the result and displaying the result .

Images