JP2011181043A - Handwriting input system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To easily carry out edition over a plurality of pages while confirming a masking objective portion by the state of paper. <P>SOLUTION: This handwriting input system includes a data storage part for storing data of a recognition result executed by masking processing in a prescribed character or a prescribed pattern obtained by character-recognizing an image of a form; a printer for printing a dot pattern indicating positional information on a paper sheet face to the paper sheet face by ink containing carbon, and printing the data of the recognition result of the data storage part to the paper sheet face by ink containing no carbon to prepare a form for correcting the recognition result; a pen device allowing writing to the form, and reading the dot pattern existing in a writing position to generate a series of the positional information; and a correction part for correcting the character or the pattern in a calligraphy detection position according to a form of a calligraphy detected based on the calligraphy of the pen device, and updating the data of the recognition result in the data storage part. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention prints recognition result data (patterns of characters, figures, etc.) obtained by character recognition of a form image optically read by, for example, a scanner, and performs handwriting correction using a pen device. Regarding the input system.

  For example, when publishing a printed matter such as a document, it may be necessary to take a measure such as masking personal information described in the printed matter. In such a case, if it is on the computer screen, the document is usually opened with word processing software or the like, personal information in the document is masked, and printing is performed. Or, if the document exists only on paper, the personal information on the paper is painted black with a pen to make it undisclosed.

  By the way, the document to be disclosed is not limited to one page, but extends to several tens of pages. In such a case, the personal information is found visually from the document file or document image displayed on the display screen with word processing software or image processing software, and the personal information is blacked out by a key operation or mouse operation, or masked. Processing is performed, but it is extremely inefficient to carry out correction work of several tens of pages by visual inspection on the screen. In addition, when personal information is visually found from information displayed on a limited screen of the display, there is a possibility that an oversight may occur, and there is a risk that personal information will be leaked due to disclosure.

  As a conventional technique related to mask processing of a document file or an image, for example, an information publishing system that reads an image image and region data from a public data file and edits a public image masked by a region that restricts the disclosure on the screen has already been published. (For example, refer to Patent Document 1).

JP 2002-298122 A

  However, in the case of the above-described conventional technique, although the non-public area of the image image can be masked, the low visibility has not yet been solved due to the work on the display.

  The present invention has been made to solve such a problem, and provides a work-efficient handwriting input system capable of easily performing editing over a plurality of pages while confirming a mask target portion in a paper state. With the goal.

  In order to solve the above problems, the image processing apparatus according to the present invention stores recognition result data obtained by performing mask processing on a predetermined character or a predetermined figure obtained by character recognition of a form image. The data storage unit and the dot pattern indicating the position information on the paper surface are printed on the paper surface with ink containing carbon, and the recognition result data of the data storage unit is printed on the paper surface with the ink not containing carbon. A printing apparatus that creates a form for correcting recognition results, a pen apparatus that can write on the form, reads the dot pattern existing at the writing position, and generates the position information series, and the pen apparatus The handwriting of the pen device is detected from the position information sequence generated by the step, and the character or figure at the handwriting detection position is corrected according to the detected form of the handwriting, Characterized by comprising a correction section for updating the data of the recognition result of the data storage unit.

  According to the present invention, it is possible to easily perform editing over a plurality of pages while confirming a mask target portion in a paper state.

It is a figure which shows schematic structure of the handwriting input system of one Embodiment of this invention. It is a block diagram which shows the structure of the computer of the handwriting input system of FIG. It is a figure for demonstrating the representation method of 2-bit information by the dot printed on a form. It is a figure which shows an example of the dot pattern comprised by the combination of the dot which shows the 2-bit information of FIG. It is a figure which shows the arrangement | sequence of the data corresponding to the dot pattern of FIG. It is a figure which shows the structure of a digital pen. It is a flowchart which shows a series of processes from a form scan to a correction process. It is a flowchart which shows the detail of a recognition process. It is a flowchart which shows the detail of a correction process. (A) is a figure which shows the example of description (1st Embodiment) of a digital pen document, (b) is a figure which shows the example of a display of a mask completion PDF file. (A) is a figure which shows the example of description (2nd Embodiment) of a digital pen document, (b) is a figure which shows the example of a display of a mask completion PDF file. (A) is a figure which shows the example of description (3rd Embodiment) of a digital pen document, (b) is a figure which shows the example of a display of a mask completion PDF file. (A) is a figure which shows the example of description (4th Embodiment) of a digital pen document, (b) is a figure which shows the example of a screen display of a mask completion PDF file.

  Hereinafter, a handwriting input system according to an embodiment of the present invention will be described in detail with reference to the drawings. FIG. 1 is a diagram showing a configuration of a handwriting input system.

  As shown in FIG. 1, the handwriting input system of this embodiment includes a computer 3, a scanner 1 as a form reading device connected to the computer 3 by a USB cable, a digital pen 4 as a writing and dot pattern reading device, and A printer 5 as a printing device and a monitor 2 connected to the computer 3 via a monitor cable are provided. The computer 3 includes a memory, a hard disk device, an external interface device, a CPU, and the like, and is connected to a keyboard, a mouse, and the like as an instruction device or an input device.

  The scanner 1 reads a form P1 on which patterns such as characters, figures, and images to be recognized are written. The monitor 2 is, for example, a liquid crystal display device and displays information input from the computer 3 on a screen.

  The printer 5 prints the character recognition result data on the blank form P2 with ink that does not contain carbon. In addition, the printer 5 prints a minute dot pattern with a special arrangement representing position information on the paper surface with a carbon-containing ink on the blank form P2.

  That is, the printer 5 prints the second recognition result read from the data storage unit 15 on the paper surface of the form P2 with the first ink that does not contain carbon, and generates a dot pattern for specifying the position on the paper surface. Is printed on the paper surface of the form P2.

  A form P3 for correction is obtained by printing character recognition result data and a dot pattern by the printer 5.

  The digital pen 4 is a pen device that inputs handwritten correction contents on a correction form P3 and reads a dot pattern printed on the form P3. The internal configuration of the digital pen 4 will be described later with reference to FIG.

  That is, in this handwriting input system, the computer 3 prints the data of the character recognition result in which the portion corresponding to the personal information is masked from the image of the form P1 obtained from the scanner 1 on the blank form P2 together with the dot pattern. Then, a form P3, which is a correction entry form, is created, and the character recognition result data is corrected by the correction function in accordance with the handwriting entered by the operator with the digital pen 4 in the correction part of the form P3.

  As shown in FIG. 2, the computer 3 includes an entry information reception unit 6, an image reception unit 7, a mask processing unit 8, a character recognition unit 9, a dictionary 10, a first and last name dictionary 11 as a personal information dictionary, a correction processing unit 12, and a memory. 14, a data storage unit 15, a print control unit 16, a personal information table 17, and a control unit 20 that comprehensively controls these units.

  Each component including the entry information reception unit 6, the image reception unit 7, the mask processing unit 8, the character recognition unit 9, the correction processing unit 12, the print control unit 16, and the like is an external storage to which a hard disk device (HDD) is applied, for example. Hardware such as a device, a CPU and a main memory, an operating system installed in the external storage device, control software, and the like are realized in cooperation.

  Specifically, the dictionary 10 is stored in an external storage device. The data storage unit 15 is, for example, a database constructed in an external storage device. Furthermore, the external storage device or the main memory stores programs necessary for handwriting input, various tables, and the like. The program is read from the main memory and executed by the CPU.

  The image receiving unit 7 receives the image of the form P1 read by the scanner 1 and stores it in the memory 14.

  The entry information receiving unit 6 collectively receives a series of dot patterns (position information) corresponding to the handwriting of the digital pen 4 from the digital pen 4 side and stores it in the memory 14. In the mask processing unit 8, the memory 14 stores a series of dot patterns (position information) received from the digital pen 4.

  The dictionary 10 stores a large number of character images (character images) and character codes (text data) associated with the character images.

  The character recognition unit 9 reads the image of the form P1 from the memory 14, performs character recognition on the recognition target range with reference to the dictionary 10, and text data (this is referred to as “first recognition result”) as the character recognition result. The data is stored in the data storage unit 15.

  The character recognition unit 9 recognizes a first pattern including either a character or a graphic from an image of the form P1 in which either a character or a graphic is described, and the first recognition result of the character recognition result is stored in the data storage unit 15. It functions as a recognition unit that stores information.

  Specifically, the character recognition unit 9 matches the character image for each character cut out from the recognition target area with reference to the dictionary 10 and the character image in the dictionary 10 (performs character recognition processing), The first recognition result of the recognition result is output to the data storage unit 15. The data storage unit 15 stores the character image obtained when the character recognition process is performed and the text data matched therewith in association with each other.

  A list of first and last names (first name and last name) and first name (first name) is stored in the first and last name dictionary 11. In addition to the first and last name dictionary 11, as a dictionary for detecting personal information, for example, address, age, gender, telephone number A dictionary storing seal stamps and the like may be used in combination.

  The mask processing unit 8 performs mask processing on the first recognition result with reference to the first name / name dictionary 11 and stores the result in the data storage unit 15. More specifically, the character or character string of the first and last name (first name and last name or last name or first name) of the first name surname dictionary 11 existing in the first recognition result is replaced with a mask (for example, shaded characters) and stored. Store in unit 15.

  That is, the mask processing unit 8 compares the first recognition result recognized by the character recognition unit 9 with the name (personal information) in the first name / name dictionary 11 and masks the name (personal information) included in the first recognition result. Processing is performed, and the second recognition result with the name (personal information) masked is stored in the data storage unit 15.

  The correction processing unit 12 functions as a handwriting detection unit that detects handwriting from a series of dot patterns read by the digital pen 4. The correction processing unit 12 has a correction function corresponding to the detected handwriting. For example, when the form of the handwriting is a range selection line such as “square” or “ellipse”, a mask process (shading process) is performed on the text data included in the range. When the form of the handwriting is a strike-through line, for example, “double line”, etc., deletion of the mask of the text data on which the double line overlaps (cancellation of shading) is executed.

  The correction processing unit 12 corrects the portion entered by the digital pen 4 in the second recognition result with a correction function corresponding to the detected handwriting.

  The correction processing unit 12 detects the handwriting of the digital pen 4 from the series of dot patterns (position information) generated by the digital pen 4 (information that is the basis for handwriting detection), and according to the form of the detected handwriting, The character or figure at the handwriting detection position is corrected, and the recognition result data in the data storage unit 15 is updated.

  When the handwriting form is, for example, a range designation line that designates a range, the correction processing unit 12 performs mask processing on the character recognition result data existing in the range of the range designation latitude line. When the handwriting is a double line, the correction processing unit 12 replaces the masked character with the original character when the character overlapping the double line is masked. That is, the masked character is restored to the original text data.

  In the data storage unit 15, for example, character recognition result data (first recognition result) obtained by character recognition of the image of the form P1 and a predetermined character or a predetermined figure are masked. The recognition result data (this is called “masked data” or “second recognition result”) is stored.

  Further, in the data storage unit 15, the position information (coordinate data) of a predetermined corner portion of the paper serving as the reference position of the form P3, the position information (coordinate data) of each part on the form P3 represented by the dot pattern, The form data of the form P3 indicating the relative positional relationship is stored. This form data also includes mask location information.

The print controller 16 controls the printer 5 to print the second recognition result and the dot pattern (see FIG. 4) on the blank form P2 when a print instruction is issued from the screen of the computer 3. At this time, the second recognition result is printed with ink not containing carbon (this is referred to as “first ink”), and the dot pattern is printed with ink containing carbon (this is referred to as “second ink”). The printer 5 is controlled to do so.
As shown in FIG. 8, the personal information table 17 includes detected mask target data (first name and last name), a detection position (how many lines of which page of the form P3), and characters or figures converted by mask processing. Correspondingly stored. That is, the personal information table 17 functions as a personal information storage unit in which personal information to be masked is stored.

  The control unit 20 controls each of the above units. The control unit 20 secures a work area in the memory 14 and executes various processes. More specifically, the control unit 20 controls the character recognition unit 9, the mask processing unit 8, and the correction processing unit 12 during, for example, character recognition processing and correction processing. For example, the control unit 20 controls the character recognition unit 9 to detect a character recognition target range in the image of the form P1, and performs image processing such as image cut-out processing for each character in the range, or a cut-out image. The character recognition process is executed.

  That is, the computer 3 performs character recognition using the dictionary 10 from the image of the form P1 read by the scanner 1, obtains text data corresponding to each pattern image as a character recognition result, and then the first name surname dictionary 11. Is used to specify a mask portion (name entry portion) of the text data, mask processing is performed on the character at the portion, and character recognition result data after mask processing is held in a hard disk device or the like.

  Here, with reference to FIGS. 3 to 5, a dot pattern printed on the blank form P <b> 2 by the handwriting input system will be described. FIG. 3 is a diagram for explaining a method of expressing 2-bit information using dots on the paper, and FIG. 4 is a diagram illustrating an example of a dot pattern composed of combinations of dots indicating 2-bit information in FIG. 5 is a diagram showing an arrangement of data corresponding to the dot pattern of FIG.

  As shown in FIG. 3, a gray dot pattern as a plurality of marks representing position information on the paper surface is printed on the entire paper surface of the blank paper document P2 to form a paper document P3. In this dot pattern, 6 × 6 vertical and horizontal dots d1, d2,... D36 arranged in a grid pattern with an average interval of 0.3 mm are misaligned with respect to the reference positions (intersections of vertical and horizontal grids) k1, k2,. The values of the respective dots d1, d2,... D36 are shown. A combination of the values of these dots d1, d2,..., D36 becomes one piece of position information indicating a predetermined location on the paper surface of the form P3.

  Specifically, as shown in FIG. 4, each dot is arranged at a position slightly shifted (shifted) from the reference position in four directions, up, down, left, and right, and the positional relationship with respect to the reference position. Gives 2 bits of encoded information 0-3. Therefore, the arrangement of data shown in FIG. 5 is obtained from the dot pattern shown in FIG. 4, and this arrangement of data indicates positional information (two-dimensional coordinate position) on the paper surface of the form P3 after printing the dot pattern. . With such a configuration, a pattern of 4 to the 36th power can be generated by combining the values of 6 × 6 dots in the vertical and horizontal directions, and theoretically for a vast two-dimensional space (coordinate plane) equivalent to 60 million square kilometers. Location information can be given. Therefore, if the form P3 to be corrected is a sheet of about A4, it is possible to determine that all forms P3 are different. That is, a plurality of types of dot patterns having this position information are printed on the form P3 over the entire sheet.

  Here, the print range of the dot pattern on the paper surface of the form P3 composed of 6 × 6 dots in the vertical and horizontal directions shown in FIG. 4 is displayed by the camera 33 (described later) built in the digital pen 4 at a time. It is a range that can be captured. Further, as shown in FIG. 1, the form P3 is printed with an entry frame for the user to enter, for example, a sentence or the like. This entry frame is previously defined (set) in the memory 14 as a character recognition range (character reading range) on the sheet P3.

Next, the digital pen 4 will be described with reference to FIG.
The digital pen 4 is a pen-type input device that can optically read the dot pattern printed on the form P3 with the second ink containing carbon. The digital pen 4 is also called an electronic pen.

  As shown in FIG. 6, the digital pen 4 includes a casing (housing) 31 formed in a shape that can be easily held by the user. A pen portion 32 is provided at the tip portion (the pen tip portion) of the casing 31. Further, an ink cartridge 34 connected to the pen portion 32 is mounted inside the casing 31. That is, writing can be performed with the ink that oozes out from the ink cartridge 34 by moving the tip portion of the pen portion 32 while pressing it against the paper surface of the form P3. The ink cartridge 34 is filled with ink not containing carbon (for example, dye-based ink).

  In addition, a camera 33 is provided in the casing 31 for capturing an image of a position where the pen unit 32 is in contact with the paper surface of the form P3 and a region in the vicinity thereof. The camera 33 includes an illumination unit such as a light emitting diode that emits infrared rays, an image sensor such as a CCD sensor or a CMOS sensor, and an optical system such as a lens. As described above, the camera 33 has a field of view of 6 × 6 dots in the vertical and horizontal directions on the paper surface of the form P3, and takes, for example, 50 or more snapshots per second.

  As the snapshot, since the light emitted from the illumination unit is infrared light, the print portion of the ink containing carbon is reflected, and only the dot pattern printed on the paper surface is photographed. In addition, a writing pressure (pressurization) sensor 35 that detects the writing pressure of the pen unit 32 is provided inside the casing 31 in order to detect the timing when the pen unit 32 fills in the form P3. Yes. That is, the camera 33 functions as a dot pattern reading unit that can optically read the dot pattern printed with the second ink.

  Further, the casing 31 includes a CPU 36, a memory 37, a battery 38, a communication unit 39, and the like. The CPU 36 performs overall control of each part in the digital pen 4, processing on video data captured by the camera 33, for example, extraction of a dot pattern on the form P3 from the video data, and conversion of the extracted dot pattern into position information Perform processing to do.

  The memory 37 is used as an area for temporarily storing video data captured by the camera 33, a work area for processing video data by the CPU 36, and the like. Further, the memory 37 stores a program necessary for processing video data by the CPU 36, a conversion table referred to for converting a dot pattern into position information, and the like.

  The battery 38 supplies power to the hardware including the communication unit 39 in the casing 31. The communication unit 39 exchanges information with the computer 3 by wireless communication such as Bluetooth (registered trademark) communication or IrDA communication. Here, instead of such wireless communication, communication with the computer 3 may be performed by wire using a USB cable or the like. When the USB cable is applied, power can be supplied from the computer 3 side to the digital pen 4 side, so that the battery 38 can be deleted.

  In the digital pen 4 configured as described above, the camera 33 reads the dot pattern printed on the paper surface when the pen tip moves on the paper by a writing operation.

  More specifically, the CPU 36 monitors the change in the output of the writing pressure sensor 35 while the camera 33 is imaging the contact position of the tip of the pen unit 32 and the area in the vicinity thereof on the paper surface of the form P3. When it is determined that the pen unit 32 is pressed onto the form P3 and the output of the writing pressure sensor 35 has reached a predetermined threshold value, the video data captured by the camera 33 is taken into the memory 37.

  Next, the CPU 36 extracts a dot pattern from the video data stored in the memory 37, refers to the conversion table stored in the memory 37, and decodes (calculates) position information corresponding to the dot pattern. To do. The position information is generated at predetermined time intervals. The position information includes x and y coordinate values (coordinate data) with respect to predetermined reference coordinates in the form P3 and time data (time stamp) indicating the time when the pen unit 32 traces the dot pattern. That is, the CPU 36 stores the x and y coordinate values in the memory 37 in time series in association with the time data.

  Therefore, the digital pen 4 can write on the form P3 through the pen unit 32, and also reads a dot pattern existing at the position where the writing was made on the form P3, and moves it to the movement locus (handwriting) of the pen unit 32. Corresponding position information series, that is, handwriting information including stroke information such as character writing order (writing direction) and writing speed, writing pressure, writing time, and the like is generated. The series of position information (handwriting information) obtained in this way is transferred to the computer 3 by the communication unit 39.

Next, the operation of the handwriting input system will be described with reference to FIGS.
In this handwriting input system, the scanner 1 scans the set form P1 (document scanning step S101), and sends an image of the form P1 to the computer 3. In the computer 3, the image receiving unit 7 receives the image of the form P 1 and stores it in the data storage unit 15.

  Subsequently, the character recognizing unit 9 reads the image of the data storage unit 15 and performs a process of recognizing patterns such as characters and graphics included in the image with reference to the dictionary 10 (character recognition step S102). As a recognition result, the image of each pattern, the corresponding text data, and the pattern position information (x coordinate, y coordinate) in the image of the form P1 are stored in the data storage unit 15.

  Next, the mask processing unit 8 reads the text data of the recognition result from the data storage unit 15 and refers to the first and last name dictionary 11 to detect (extract) personal information of first and last names (names and the like) included in the text data ( The personal information detection step S103) performs mask processing on the detected first and last names (names, etc.) and stores them as second recognition results (masked data) in the data storage unit 15.

  Thereafter, when there is a print instruction, the print control unit 16 controls the printer 5 to print the second recognition result (masked data) on the blank form P2 (automatic masked document printing step S104), thereby creating the form P3. To do.

  Here, a dot pattern readable by the Anoto digital pen 4 is printed by the printer 5 together with the second recognition result (masked data) subjected to the masking process on the form P3. This form P3 is called a digital pen document, and FIG.

  In the digital pen document of the printing example shown in FIG. 10A, “Nippon Taro” after “however” is a name, that is, a mask portion, and this portion is printed with shading. The characters “Nihontaro” are legible.

  The operator (user) confirms whether or not the personal information is appropriately masked by visually confirming the printed digital pen document.

  In the character recognition and the extraction of personal information from the image read from the form P3, since the form P1 as the reading source is handwritten, a 100% reliable result cannot be obtained. For this reason, there is a possibility that some error has occurred in the document that has been automatically masked.

  Therefore, after performing the correction process using the form P3 and the digital pen 4 (correction process step S200), the “recognition” operation indicating the completion of the correction is performed, whereby the correction processing unit 12 receives the second recognition result. Is corrected (mask completed), converted into a PDF file and stored in the data storage unit 15. The correction processing unit 12 performs a process of painting a shaded portion black when converting to a PDF file. As a result, as shown in FIG. 10B, the mask completion PDF in which the personal information 63 is painted black in the corrected second recognition result is obtained.

  Here, the character recognition processing step S102 will be described in detail. In the character recognition processing step S102, as shown in FIG. 8, an image of the form P1 is obtained by scanning the form P1, which is the original document (step S201). And the text data which is a 1st recognition result is obtained by character-recognizing the obtained image (step S202).

  The first recognition result is text-searched and compared with the data stored in the first and last name dictionary 11 to extract the matched personal information (step S204), and the extracted personal information is stored in the personal information table 17 ( In step S207, the coordinates of the area of the personal information and the first recognition result are stored in the data storage unit 15 (step S208).

  Finally, the mask processing unit 8 converts the personal information into shaded characters, and prints the automatically masked document as the second recognition result as a form P3 (digital pen document) (step S209).

  Subsequently, the correction processing step S200 will be described. In the correction processing step S200, as shown in FIG. 9, the digital pen 4 is moved onto the paper surface of the form P3 (digital pen document) (step S301), and the digital pen 4 is operated (step S302). The series of read dot patterns (position information) is sequentially input to the entry information receiving unit 6 (step S303) and is taken into the memory 14.

  The correction processing unit 12 reads a series of dot patterns (position information) from the memory 14 to detect a handwriting, and matches or resembles a handwriting form (line type) for each correction function set in advance. The presence or absence of a thing is discriminated (step S304).

  As a result of this determination, if the line type is a double line (“double line” in step S304), the correction processing unit 12 deletes the shading of the character at the position overlapping the double line (deletes the mask). (Step S305), and the personal information table 17 is updated (Step S306).

  If the line type is a square (“square” in step S304), the correction processing unit 12 determines whether or not there is an input from the digital pen 4 in the area surrounded by the square (step S304). S307).

  When there is an input (entry) from the digital pen 4 (“Yes” in step S307), the character (for example, “A”) entered in the area is recognized based on the handwriting (step S308) and recognized. The mark such as the character or figure that has been added is added to the corresponding name column in the personal information table 17 (step S309). For example, a mark such as “A” is added to the registered “Taro Nihon” column in the personal information table 17.

  On the other hand, when there is no input (entry) from the digital pen 4 for a certain time (“No” in step S307), a mask is added to the text data in the area, that is, shaded characters are added (step S310).

  If there is a correction end operation (Yes in step S311), the corrected document is stored in the data storage unit 15 (step S312).

Here, a specific example of addition and deletion of mask portions in the correction processing step S200 will be described.
(Add mask part)
When the operator confirms the form P3, as shown in FIG. 10A, the character string “Nippon Taro” on the first line is the shaded character 61 and should be masked even though it is masked. When the character string “Nippon Hanako” (name) on the third line, which is personal information, is not masked, a range designation line 62 indicated by a bold line is placed on the character string “Nippon Hanako” with the digital pen 4. Handwritten to enclose the character string “Nippon Hanako”. Here, the range designation line 62 is described as a rectangular frame as much as possible.

  At this time, the digital pen 4 transmits the dot pattern read from the paper surface by the pen tip camera 33 to the computer 3 in real time.

  In the computer 3, the correction processing unit 12 determines which coordinate of which page the pen point moves from the dot pattern information received by the entry information receiving unit 6 and the preset dot pattern position information. (This is called handwriting detection), and a rectangle connecting the outer shapes of the character strings “Nippon Hanako” within the range surrounded by the detected handwriting “range designation line 62” is recognized, and the rectangular area is recognized. Is determined as a mask location, and data of the mask location (the coordinates of the two vertices of the diagonal of the rectangle) is added to the memory 14.

(Delete mask part)
When a portion that should not be masked is masked (shaded characters) or masked by mistake, such as the character string “Abandoned” on the second line of the digital pen document in FIG. In other words, the correction processing unit 12 detects the handwriting of the double line 63 and cancels (deletes) the mask of the part by drawing the double line 63 at the place with the digital pen 4.

When the operator corrects all the locations and then performs a determination operation for the corrected locations, the correction processing unit 12 converts the second recognition result reflecting the corrected locations into a masked PDF file, and stores it in the data storage unit 15. Output such as storing or displaying on the screen of the monitor 2 is performed.
As a result, as shown in FIG. 10B, the masked PDF file is displayed on the screen of the monitor 2 with the mask portion 64 which is a portion of the personal information masked black. The correction location confirmation operation may be performed by pressing the ENTER button on the keyboard, or by using the digital pen 4 to point a predetermined position (such as a correction end mark) on the form P3.

  Conventionally, when an existing image processing tool of image processing application software (for example, Photoshop or GIMP) is activated and an image is corrected on the screen, it is difficult to overlook the whole work as compared with paper. There is a high risk that masks will be overlooked. A high-definition display having the same resolution as that of paper has been developed, but a high-definition display is extremely expensive at about 1 million yen per unit.

  In that respect, the digital pen 4 has a remarkably low price of about several tens of thousands of yen, and since it is paper, the visibility is extremely high and low-cost and reliable correction can be performed.

  Also, other handwritten pen devices such as ultrasonic pens cannot identify the page, so there is a drawback in that it is necessary to give an operation on the computer which page is being edited. However, the Anoto digital pen 4 obtains in real time which page the coordinate is on, so there is no need to specify the page on the computer, and editing across multiple pages is extremely easy. Can be done.

  As described above, according to the handwriting input system of the first embodiment, the work efficiency can be stored electronically after correcting the portion by handwriting while confirming the mask target portion in a paper state. Can provide a good handwriting input system.

Hereinafter, modified examples (second embodiment to fourth embodiment) from the example (first embodiment) shown in FIG. 10 will be described.
(Second Embodiment) FIG. 11 shows a second embodiment. In the second embodiment, in addition to the mask method described in the first embodiment, as shown in FIG. 11B, a function for creating a mask completion PDF file in which the name is replaced with a pseudonym such as “A” is provided. It is added. In this case, the operator designates a range with the digital pen 4 so as to surround the personal information of the mask omission, for example, “Nippon Hanako”, among the text data printed on the form P3, as shown in FIG. A line 62 is drawn, and then, for example, a character 65 “B” is added inside the range designation line 62.

  At this time, the correction processing unit 12 recognizes the character “B” 65 inputted after the range is designated, and recognizes the recognized character “B” in the column of the name “Nippon Hanako” newly registered in the personal information table 17 this time. Register additional.

  In addition, when the same name as the name already registered in the personal information table 17 is detected at the time of character recognition and mask processing at the time of additional registration, the name detected thereafter is used as the name of the personal information table 17. Replace with the corresponding kana.

  Also in this case, since the character recognition process cannot obtain 100% performance, an error occurs in the mask portion. Therefore, in this example, a name overlooked by the automatic mask process is enclosed in a rectangle, and characters such as kana “A” and “B” written therein are inserted. The correction processing unit 12 recognizes the handwritten range selection line and the character to determine the character at the mask target position and the replacement kana.

  In this example, when the corrected recognition result is output as a PDF file, as shown in FIG. 11B, inverted characters (white characters) Kana “A” and “B” overlap with a black background mask. The masks 66 and 67 are replaced.

  (Third Embodiment) FIG. 12 shows a third embodiment. In this example, as shown in FIG. 12A, when the operator encloses a portion to be masked with the digital pen 4 with a range designation line 62 slightly larger, the correction processing unit 12 is in the enclosed area. A mask is generated in accordance with the rectangle of the nearest character string (concatenated outline of each character). As a result, as shown in FIG. 12B, the mask 63 is displayed with an appropriate size for the characters, and can be masked in the appearance of the recognition result data.

  Based on the same principle, even when an area is surrounded slightly smaller, a mask is generated in accordance with the character rectangle (character outline) closest to the area. As a result, even when the operator performs a somewhat rough operation with the digital pen 4, the mask can be put on the correct position.

  (Fourth Embodiment) FIG. 13 shows a fourth embodiment. In this example, as shown in FIG. 13A, when the operator encloses personal information to be masked (character string “Nippon Hanako”) with a range designation line 62 with the digital pen 4, the correction processing unit 12 Then, the same character string as the character string “Nippon Hanako” is detected from the entire document, and a warning flag is set in the corresponding character string of the second recognition result. After correcting the second recognition result on the paper in this way, a mask completion PDF file is created and stored in the data storage unit 15.

  When the mask completion PDF file, which is the final recognition result, is displayed on the screen of the monitor 2 using a PDF file browsing tool, as shown in FIG. “Hanako” is shaded 68, or a mark 69 or a message “warning” is displayed near the location to notify the operator of mask leakage. According to this embodiment, forgetting to mask the same name can be prevented.

  In addition, this invention is not limited only to the said embodiment, You may deform | transform a component in the range which does not deviate from the summary in an implementation stage. For example, each constituent element described in the above embodiment may be realized by a program installed in a storage such as a hard disk device of a computer, or the program is stored in a computer-readable electronic medium: program The computer may realize the functions of the present invention by causing the computer to read from the electronic medium. Examples of the electronic medium include a recording medium such as a CD-ROM, flash memory, and removable media. Further, the configuration may be realized by distributing and storing components in different computers connected via a network, and communicating between computers in which the components are functioning.

  DESCRIPTION OF SYMBOLS 1 ... Scanner, 2 ... Monitor, 3 ... Computer, 4 ... Digital pen, 5 ... Printer, 6 ... Entry information reception part, 7 ... Image reception part, 8 ... Mask processing part, 9 ... Character recognition part, 10 ... Dictionary, DESCRIPTION OF SYMBOLS 11 ... Last name dictionary, 12 ... Correction processing part, 14 ... Memory, 15 ... Data storage part, 16 ... Print control part, 17 ... Personal information table, 20 ... Control part, 31 ... Casing, 32 ... Pen part, 33 ... Camera , 34 ... ink cartridge, 35 ... sensor, 35 ... writing pressure sensor, 37 ... memory, 38 ... battery, 39 ... communication unit.

Claims (4)

A data storage unit in which data of a recognition result obtained by performing mask processing on a predetermined character or a predetermined figure obtained by character recognition of an image of a form;
A dot pattern indicating positional information on the paper surface is printed on the paper surface with carbon-containing ink, and the recognition result data is printed on the paper surface with the recognition result data of the data storage unit using the carbon-free ink. A printing device to create,
A pen device that can write on the form and reads the dot pattern present at the writing position to generate the position information series;
The handwriting of the pen device is detected from the position information sequence generated by the pen device, and the character or the figure at the handwriting detection position is corrected according to the detected form of the handwriting, and the data storage unit is recognized. A handwriting input system comprising: a correction unit that updates result data. The correction unit is
2. The handwriting input system according to claim 1, wherein when the form of the handwriting is a range designation line surrounding a character or a figure as a recognition result, a mask process is performed on the character or the figure in the corresponding range. The correction unit is
2. The handwriting input system according to claim 1, wherein when the form of the handwriting is a strikethrough drawn so as to overlap a masked character or figure, the mask of the corresponding character or figure is deleted.   When the form of the handwriting is a combination of a range designation line surrounding a character or graphic as a recognition result and a replacement character or graphic, the character or graphic is replaced with a replacement character or graphic. The handwriting input system according to claim 1.

Images