JP2010204745A - Document processing apparatus and document processing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a document processor capable of easily correcting an error in writing of accumulated handwritten input information, in the document processor for accumulating handwritten input information input by handwriting. <P>SOLUTION: A document processor 100 is provided with an input/output I/F part 120 for accepting handwritten input information; a handwritten character recognition engine 124 for generating the character code information and THE evaluation value of the handwritten input information; a control part 122 for accumulating handwriting evaluation data 126a, in which the handwritten input information is associated with the character code information and the evaluation value; and a correction input detection part 121 for detecting the correction of the handwritten input information. The control part 122 extracts the handwritten input information, shown with the same character code information as the character code information of the handwritten input information, in which correction has been detected and evaluation value is a value which lies within a prescribed first range, from among the handwritten input information contained in the handwriting evaluation data 126a. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a document processing apparatus and a document processing method having a function of accumulating handwritten input information input by handwriting.

  When a person inputs document data into an electronic device such as a personal computer or a mobile phone, a method of inputting with a keyboard or a button has been generally used. However, in recent years, in a tablet input device, a tablet PC, a PDA (Personal Digital Assistants), a smartphone, a game terminal, etc., it has an input function by handwriting using a dedicated pen, etc. Devices equipped with a function of converting handwritten input information) into character code information have become widespread. Here, the character code information is an identifier for identifying handwritten input information.

  As another method of converting handwritten input information into character code information, handwritten data written on paper is captured as digital data into a personal computer by a scanner or the like, and the handwritten input information is converted into character code information by processing such as software. Electronic devices and software related to so-called OCR (Optical Character Reader) functions for conversion are also widely used.

  In a device or software that converts such handwritten input information into character code information, the character information indicated by the converted character code information is displayed on the screen. In addition, when such a device or software is used by the user, when the handwritten input information is converted into character code information, it is not converted into character code information intended by the user due to typographical errors or wrinkles during the user's handwriting input. (Character recognition is not possible).

  Regarding such erroneous conversion (misrecognition) of handwritten input information, in Patent Document 1, all converted character code information strings (document strings) having the same converted character code are converted into different character codes. This document describes a document processing apparatus that facilitates correction of character code information after erroneous conversion (erroneous recognition) by performing conversion. Further, in Patent Document 2, in a device that performs character code conversion for handwritten input each time, when there is no character code conversion candidate, a function that issues a warning that the input character is a typo, This document describes a document processing apparatus having a function of obtaining character code information intended by a user by outputting candidates.

Thus, handwritten input information is used as information for conversion to character code information, and the document processing apparatus holds the character code information after conversion and displays the character information indicated by the character code information. Several methods have been proposed for dealing with erroneous conversion (misrecognition) of handwritten input information based on the premise.
JP-A-6-119484 JP-A-8-44724

  However, since the conventional document processing apparatus is premised on displaying the character information indicated by the character code information, the conventional document processing apparatus assumes that the handwritten input information itself is displayed. There is a problem that an error in input information cannot be corrected.

  That is, the conventional document processing apparatus converts the handwritten input information into character code information, accumulates the character code information, and displays the character information indicated by the character code information. However, in recent years, a document processing apparatus having a function of accumulating and displaying handwritten input information itself is desired for application to an electronic notebook or the like.

  Then, in the case of a document processing apparatus that stores handwritten input information itself as digital data and displays the stored handwritten input information, the user makes a mistake in writing, such as with a handwriting on paper such as a notebook. The possibility of steadily accumulating data including the character is sufficiently higher than the case of temporarily performing handwritten input for conversion into character code information. For example, a case where “perfect” is written as “complete wall” remains, and data which is written as “complete wall” is accumulated for a long time.

  There is a case where such a mistake is noticed when the user makes a mistake and is corrected. However, whether or not the same error has been made in the past has required the user himself / herself to refer to all the past data and correct the data therein as necessary. Therefore, if the accumulated data is enormous, considerable labor is required.

  As described above, in the document processing apparatus that accumulates the handwritten input information in order to display the handwritten input information itself, there is a problem that it is not possible to easily correct errors in the accumulated handwritten input information.

  Therefore, in view of the above problems, the present invention provides a document processing apparatus that can easily correct errors in stored handwritten input information when storing the handwritten input information to display the handwritten input information itself. The purpose is to provide.

  In order to achieve the above object, a document processing apparatus according to the present invention is a document processing apparatus that accumulates handwritten input information input by handwriting, the input unit receiving the handwritten input information, and the received handwriting The handwritten information generation unit that generates an identifier for identifying input information, an evaluation value indicating a degree of error in the handwritten input information, and the generated handwritten input information are associated with the generated identifier and the evaluation value A storage unit for storing the handwritten evaluation data, a correction input detection unit for detecting correction of the handwritten input information included in the stored handwritten evaluation data, and a handwriting input information included in the stored handwritten evaluation data. The identifier is the same as the identifier for identifying the handwritten input information whose correction is detected, and the evaluation value is within a predetermined first range. And a extraction unit for extracting handwritten input information is.

  According to this, the handwritten input information including the same error as the corrected handwritten input information is extracted. Thereby, the user can easily extract handwritten input information including erroneous entries from the handwritten input information accumulated in the past. Therefore, when the handwritten input information is stored in order to display the handwritten input information itself, an error in the stored handwritten input information can be easily corrected.

  In addition, preferably, based on an instruction received by the correction receiving unit and a correction receiving unit that receives an instruction to correct the handwritten input information extracted by the extracting unit, the handwritten input information extracted by the extracting unit is corrected. And a correction unit.

  According to this, the handwritten input information is corrected based on the instruction to correct the handwritten input information. That is, the handwritten input information is corrected based on the user's instruction. For this reason, the user can select whether or not to correct the handwritten input information, and can easily correct errors in the stored handwritten input information.

  Preferably, the correction unit corrects the handwritten input information extracted by the extraction unit to the handwritten input information after correction of the handwritten input information detected by the correction input detection unit.

  According to this, the handwritten input information extracted by the extracting unit is corrected to the handwritten input information after correction of the handwritten input information detected by the correction input detecting unit. That is, the correction target handwritten input information is corrected so that the correction target handwritten input information is the same as the previously corrected handwritten input information. Thereby, since the user can perform correction again by utilizing the previously corrected handwritten input information, it is possible to easily correct erroneously written handwritten input information.

  Preferably, the handwritten information generation unit generates an identifier and an evaluation value corresponding to the corrected handwritten input information of the handwritten input information detected by the correction input detection unit, and the document processing device further includes: The update for updating the handwritten evaluation data on the handwritten input information whose correction is detected by the correction input detecting unit to the handwritten evaluation data in which the corrected handwritten input information is associated with the generated identifier and evaluation value A part.

  According to this, when handwritten input information is corrected, handwritten evaluation data is also updated. For this reason, since handwritten evaluation data corresponding to the latest handwritten input information is accumulated, it is possible to easily correct errors in the accumulated handwritten input information without making an erroneous determination based on the incorrect handwritten evaluation data. .

  Preferably, the extraction unit further extracts the handwritten input information when the evaluation value of the handwritten input information received by the input unit is a value within a predetermined second range, and the input unit Further, accepting the handwritten input information after the extracted handwritten input information is corrected, the handwritten information generating unit generates an identifier and an evaluation value corresponding to the corrected handwritten input information, The storage unit, when the evaluation value of the handwritten input information received by the input unit is not a value within a predetermined second range, the corrected handwritten input information, the generated identifier, and the evaluation Handwritten evaluation data associated with values is accumulated.

  According to this, when the input unit accepts the handwritten input information, the error in the handwritten input information is corrected and accumulated. For this reason, accumulation | storage of the handwritten input information containing an error can be reduced.

  Preferably, the handwritten information generation unit generates the identifier for identifying the handwritten input information in a predetermined unit, and the evaluation value indicating the degree of error in the handwritten input information in the predetermined unit. The storage unit stores handwritten evaluation data in which the predetermined unit of handwritten input information is associated with the generated identifier and the evaluation value, and the correction input detection unit stores the predetermined input Correction of a predetermined unit of handwriting input information included in the unit of handwriting evaluation data is detected, and the extraction unit includes the predetermined unit of handwriting input information included in the stored unit of handwriting evaluation data. The handwriting of a predetermined unit indicated by the same identifier as the identifier for identifying the handwritten input information of the predetermined unit in which the correction is detected and the evaluation value is a value within a predetermined first range To extract the force information.

  According to this, it is determined whether the handwritten input information for each predetermined unit is erroneous. As a result, the handwritten input information that does not include an error as a character but includes an error as a word can be corrected. For example, the word “complete wall” does not include an error in the word “wall” itself, but includes an error because the word “wall” must be “n”. Therefore, even in such a case, it is possible to easily correct errors in the stored handwritten input information.

  The present invention can be realized not only as such a document processing apparatus but also as a method in which the processing of each characteristic processing unit included in the document processing apparatus is used as a step or the processing of each processing part. Can be realized as a program for causing a computer to execute, or as an integrated circuit that performs processing of each processing unit. Needless to say, such a program can be distributed via a recording medium such as a CD-ROM or a transmission medium such as the Internet.

  ADVANTAGE OF THE INVENTION According to this invention, in the document processing apparatus which accumulate | stores the said handwritten input information in order to display the handwritten input information itself input by handwriting, the error of the stored handwritten input information can be corrected easily.

  Embodiments will be described below.

(Embodiment 1)
Hereinafter, Embodiment 1 of the present invention will be described in detail with reference to the drawings.

  FIG. 1 is an external view showing a configuration of a document processing apparatus 100 according to an embodiment of the present invention.

  As shown in the figure, the document processing apparatus 100 includes an input / output I / F unit 120. Then, when the user writes characters or the like in the input / output I / F unit 120 by handwriting with the input device 110, the document processing apparatus 100 accumulates the characters or the like input by handwriting as handwritten input information.

  Here, the input device 110 is an input device such as a dedicated pen using an electromagnetic induction function for inputting handwritten input information to the document processing apparatus 100.

  FIG. 2 is a block diagram showing a functional configuration of the document processing apparatus according to the embodiment of the present invention.

  As shown in the figure, the document processing apparatus 100 includes an input / output I / F unit 120, a correction input detection unit 121, a control unit 122, a character information search unit 123, a handwritten character recognition engine 124, an assist information generation unit 125, and an input. A data storage unit 126 and a recognition dictionary 127 are provided. The document processing apparatus 100 is an apparatus such as a mobile digital device such as a mobile phone, a PDA, or a game.

  The input / output I / F unit 120 accepts handwritten input information input by the input device 110 and converts it into digital information as handwritten stroke information, and a display function for displaying various digital information including accumulated handwritten input information And is controlled by the control unit 122.

  The input / output I / F unit 120 is a device in which various input devices such as a resistive film panel, a capacitance panel, and an electromagnetic induction panel are combined with various display devices such as a liquid crystal panel and an electronic paper panel. The input / output I / F unit 120 has a function of an “input unit” described in the claims.

  The handwritten character recognition engine 124 generates an identifier for identifying the handwritten input information received by the input / output I / F unit 120 and an evaluation value indicating the likelihood of the handwritten input information. Here, the identifier for identifying the handwritten input information is character code information indicating the character code of the handwritten input information. Further, the evaluation value is a value indicating the degree to which erroneous writing is included in the handwritten input information, and here, the evaluation value becomes higher as the degree of including erroneous writing is smaller.

  Specifically, the handwritten character recognition engine 124 converts handwritten input information input to the input / output I / F unit 120 into character code information in accordance with an instruction from the control unit 122, and which conversion result is included along with the conversion result. It has a function to generate an evaluation value indicating whether the degree is certain.

  The handwritten character recognition engine 124 uses data registered in the recognition dictionary 127 when converting handwritten input information into character code information. Here, a detailed description of a function for converting to character code information and a conversion algorithm when generating an evaluation value for the conversion result is omitted. The handwritten character recognition engine 124 has a function of a “handwritten information generation unit” described in the claims.

  The correction input detection unit 121 detects correction of handwritten input information included in the handwriting evaluation data 126a stored in the input data storage unit 126 described later. Specifically, the correction input detection unit 121 is a processing unit controlled by the control unit 122, and the handwritten input information input to the input / output I / F unit 120 is handwritten character information or corrects handwritten character information. It has a function of detecting whether the information is to be used.

  The control unit 122 has a function of performing various controls in the document processing apparatus 100, and according to a handwriting input operation from the user, the input / output I / F unit 120, the correction input detection unit 121, and the character information search unit 123. Various controls of the handwritten character recognition engine 124, the assist information generation unit 125, and the input data storage unit 126 are performed.

  In addition, the control unit 122 stores, in the input data storage unit 126, handwritten evaluation data 126 a in which handwritten input information is associated with an identifier and an evaluation value generated by the handwritten character recognition engine 124. In addition, the control unit 122 is indicated by the same identifier as the identifier for identifying the handwritten input information whose correction is detected from the handwritten input information included in the accumulated handwritten evaluation data 126a, and the evaluation value is a predetermined number. Handwritten input information that is a value within one range is extracted. Furthermore, the control unit 122 receives an instruction to correct the extracted handwritten input information. Then, the control unit 122 corrects the extracted handwritten input information based on the received instruction. Further, the control unit 122 updates the handwritten evaluation data 126a for the handwritten input information whose correction is detected to the handwritten evaluation data 126a in which the corrected handwritten input information is associated with the generated identifier and evaluation value. To do.

  The control unit 122 has functions of “accumulation unit”, “extraction unit”, “correction reception unit”, “correction unit”, and “update unit” described in the claims.

  The character information search unit 123 has a function of searching for desired data from the handwritten evaluation data 126 a stored in the input data storage unit 126 under the control of the control unit 122.

  The assist information generation unit 125 has a function of generating assistant information for correcting errors written in the input / output I / F unit 120.

  The input data accumulation unit 126 stores accumulated handwritten evaluation data 126a.

  Specifically, the input data storage unit 126 includes a nonvolatile memory function such as an HDD or a flash memory. Then, the input data storage unit 126 includes information (stroke information) indicating the handwritten input information itself input to the input / output I / F unit 120 and character code information obtained by converting the handwritten input information by the handwritten character recognition engine 124. And a function of accumulating evaluation values output from the handwritten character recognition engine 124 together with the character code information.

  3A to 3C are diagrams illustrating the configuration of stroke information.

  The stroke information is information including a combination of a parameter indicating the ON / OFF state of the pen, a parameter indicating the X coordinate, and a parameter indicating the Y coordinate as shown in the configuration shown in FIG. 3A.

  FIG. 3B shows a specific configuration example. Since P is “1”, the pen is ON (= the pen is in contact with the display), the X coordinate is 0x10, and the Y coordinate is 0x30. It shows that there is.

  The handwritten character “now” is 3 (P0, X0, Y0), (P1, X1, Y1), (P2, X2, Y2)... (Pn, Xn, Yn) as shown in FIG. It consists of a plurality of stroke information with one parameter as a set. The setting of what kind of stroke information is combined with one character or what sampling frequency the stroke information is acquired depends on the specifications of various character recognition engines. Will be omitted.

  4A and 4B are diagrams illustrating the configuration of the handwritten character recognition engine 124.

  FIG. 4A shows an input / output configuration of the handwritten character recognition engine 124. When stroke information is input to the handwritten character recognition engine 124, while referring to the recognition dictionary 127, for some character code information (C0, C1, etc. shown in the figure) and each character code information as output, An evaluation value (V0, V1, etc. shown in the figure) indicating how much it is likely is output. Here, the character code information is a code such as Shift JIS, for example, and the evaluation value is a value from 0 to 100, and an index indicating that the character recognition result is more probable as the value is closer to 100 is unique to the handwritten character recognition engine 124. This is the value to be calculated.

  FIG. 4B shows a specific configuration example of the output information. Character code information “0x11CD” has an evaluation value “97”, and character code information “0xAA22” has an evaluation value “70”. “0x11CD” having an evaluation value close to 100 indicates that the conversion result to character code information is most likely. In this way, the handwritten character recognition engine 124 converts the handwritten character information into character code information, and at the same time generates and acquires an evaluation value.

  FIG. 5 is a diagram illustrating an example of the handwritten evaluation data 126 a stored in the input data storage unit 126.

  As shown in the drawing, the handwritten evaluation data 126a is data in which handwritten input information (stroke information) is associated with an identifier (character code information) generated by the handwritten character recognition engine 124 and an evaluation value. The input data accumulation unit 126 accumulates such combinations of stroke information, character code information, and evaluation values with classifications such as Data # 1, Data # 2,.

  Here, for the sake of simplicity, an example is shown in which stroke information is input in units of one character, character code information and its evaluation value are output, and stored in the input data storage unit 126, but the specification is not limited. The handwritten character recognition engine 124 is provided with a function for converting and recognizing character code information in word units and phrase units, or a recognition dictionary, and an evaluation value that is an index indicating the certainty of conversion for each predetermined unit, The input data storage unit 126 may be configured to store the data.

  Next, a processing sequence when handwritten character information is input by the user in the document processing apparatus 100 will be described with reference to FIG.

  FIG. 6 is a flowchart showing an example of the operation of the document processing apparatus 100 when inputting handwritten input information according to Embodiment 1 of the present invention.

  (Step S501) The input / output I / F unit 120 accepts handwritten input information. Specifically, the user uses the input device 110 to input handwritten character information to the input / output I / F unit 120. The control unit 122 controls the input / output I / F unit 120 such as receiving input handwritten character information and displaying it on the input / output I / F unit 120. FIG. 1 shows an example of input and output of specific user handwritten character data. That is, the user inputs the character “today” by handwriting, and the character is displayed on the input / output I / F unit 120.

  (Step S502) Next, the handwritten character recognition engine 124 generates the character code information and the evaluation value of the handwritten input information received by the input / output I / F unit 120. Specifically, the control unit 122 inputs the input handwritten character information to the handwritten character recognition engine 124 and acquires the character code information and its evaluation value. The timing when the control unit inputs the handwritten character information to the handwritten character recognition engine 124 is, for example, after the input in units of one character is completed. In the example of FIG. 1, the control unit 122 determines that “now” of “today” has been input, and inputs the stroke information to the handwritten character recognition engine 124.

  (Step S503) Next, the control unit 122 accumulates handwritten evaluation data 126a in which the handwritten input information is associated with the character code information generated by the handwritten character recognition engine 124 and the evaluation value. Specifically, the control unit 122 stores the stroke information, which is handwritten input information, and the character code information and evaluation value with the highest evaluation value for each character obtained from the handwritten character recognition engine 124 as input data. Stored in the unit 126.

  As described above, the stroke information, the character code information, and the evaluation value related to the handwritten input information are accumulated in the handwritten evaluation data 126a.

  Next, a processing sequence for correcting handwritten character information once input by the user in the document processing apparatus 100 will be described with reference to FIG.

  FIG. 7 is a flowchart illustrating an example of the operation of the document processing apparatus 100 when correcting handwritten input information.

  (Step S701) The correction input detection unit 121 detects correction of handwritten input information. Specifically, when the control unit 122 controls the correction input detection unit 121, the correction input detection unit 121 receives data input to the input / output I / F unit 120 by the user using the input device 110. , To detect that the input is correction.

  8 to 10 are diagrams illustrating specific examples when handwritten input information is corrected in the document processing apparatus 100. FIG.

  FIG. 8 shows a specific operation example in which the user corrects handwritten input information. FIG. 8A shows an example in which the user has mistakenly entered the word “first”, which is originally “Koromo-Men”, with “Shime-Sen”. The user sets a change to the mode for correcting handwritten input information displayed on the input / output I / F unit 120 as shown in FIG. At this time, the correction input detection unit 121 detects that the handwritten input information is corrected.

  (Step S <b> 702) Returning to FIG. 7, the control unit 122 performs processing such as calculating area information first touched on the input device 110 after setting the correction mode in which the handwritten input information is corrected. Identify which handwritten input information is to be corrected, and let the character information search unit 123 search the data stored in the input data storage unit 126 for the specified handwritten input information, and character code information and evaluation before correction Get the value.

  (Step S703) Next, when the handwritten character is corrected by the user, the control unit 122 performs processing of the handwritten character recognition engine 124, and the stroke data stored in the input data storage unit 126 and the evaluation. The value is updated, and for example, the display as shown in FIG. Specifically, the handwritten character recognition engine 124 generates character code information and an evaluation value corresponding to the handwritten input information after the correction of the handwritten input information detected by the correction input detection unit 121. Then, the control unit 122 updates the handwritten evaluation data 126a for the handwritten input information whose correction is detected to the handwritten evaluation data 126a in which the corrected handwritten input information is associated with the generated identifier and evaluation value. To do.

  (Step S704) Next, the control unit 122 out of the handwritten input information included in the stored handwritten evaluation data 126a, the handwritten input information of the same character code information as the character code information of the handwritten input information in which the correction is detected. To extract. Specifically, the control unit 122 controls the character information search unit 123 so that the character information search unit 123 acquires the handwritten evaluation data 126a stored in the input data storage unit 126 in step S702. A search is performed to determine whether there is the same character code information as that before correction. If no character code information exists, the process ends.

  (Step S705) The control unit 122 further extracts handwritten input information whose evaluation value is a value within a predetermined first range. Specifically, by controlling the character information search unit 123, the control unit 122 also searches for the evaluation value if there is a character code information in step S704. If the evaluation value is larger than the predetermined value (P1), the handwritten input information is considered not to be erroneously written, and the process is terminated as it is.

  (Step S706) Next, when the evaluation value is smaller than the predetermined value (P), the control unit 122 generates assist information on the assumption that the handwritten input information is likely to be erroneous.

  FIG. 9 shows a configuration example of the handwritten evaluation data 126 a stored in the input data storage unit 126. Each data is managed for each predetermined data unit such as “Data ***”, and character code information and its evaluation value are stored therein.

  As shown in the figure, a plurality of pieces of code information indicating “0x8F89”, which is the “first” shift JIS code, are stored, but the evaluation values are different. Here, when an evaluation value smaller than 50 is highly likely to be erroneous, handwritten input information corresponding to the character code information indicated by “0x8F89” included in Data # n and Data # m is It is searched by the character information search unit 123, and becomes handwritten input information to be displayed as assist information in this step. The control unit 122 generates assist information by controlling the assist information generation unit 125 based on the handwritten input information searched as described above.

  FIG. 10A is a specific example in which the information generated by the assist information generation unit 125 is output to the input / output I / F unit 120. Of the handwritten evaluation data 126a stored in the input data storage unit 126, the fourth line in the data managed under the name “Diary 1” and the data managed under the name “Memo 20”. Since it is considered that there is a similar error in the third line, information on whether or not the same correction as the content corrected this time is output.

  As described above, the assist information generation unit 125 generates and displays button information or the like that accepts user input regarding how to correspond to other information on a place where there is a possibility of error. Here, there is a method in which a place where there is a possibility of error is once displayed as “confirmation of location”, and the user confirms it. However, for the sake of simplicity, the detailed explanation is omitted here.

  In addition, when correcting other places where there is a possibility of mistakes in the stroke information of handwritten input information, it is assumed that the size, color, etc. of the surrounding characters may not match, but depending on the size of the surrounding characters automatically Since the mechanism for changing the size is simple, a detailed description thereof is omitted here.

  (Step S707) Returning to FIG. 7, the control unit 122 accepts an instruction to correct the extracted handwritten input information. Specifically, the control unit 122 accepts whether or not the user corrects information on other places where there is a possibility of error. If the user does not make corrections (when the user selects the “No” button in FIG. 10A), the process ends.

  (Step S708) Next, based on the instruction received from the user, the control unit 122 corrects the extracted handwritten input information to the handwritten input information after the correction of the handwritten input information detected by the correction input detection unit 121. . Specifically, when the user receives an instruction to correct information on a place where there is a possibility of erroneous writing, the control unit 122 receives the input data storage unit 126 in the same manner as in step S703. The stroke information etc. of the handwritten evaluation data 126a stored in is updated, and a display as shown in FIG. 10B is performed.

  As described above, the document processing apparatus 100 holds the handwritten input information (stroke information) input by the input device 110, the character code information converted from the stroke information by the handwritten character recognition engine 124, and the evaluation value. When the character is corrected, the user can search for and correct the same past error information based on the character code information and the evaluation value that is held, and the user can input the erroneous handwritten input information recorded in the past. It is easy to make corrections.

  In other words, according to the document processing apparatus 100 according to the present invention, handwritten input information including an erroneous entry similar to the corrected handwritten input information is extracted. Thereby, the user can easily extract handwritten input information including erroneous entries from the handwritten input information accumulated in the past.

  Further, the handwritten input information is corrected based on an instruction to correct the handwritten input information. That is, the handwritten input information is corrected based on the user's instruction. For this reason, the user can select whether or not to correct the handwritten input information.

  Further, the handwritten input information extracted by the control unit 122 is corrected to the corrected handwritten input information of the handwritten input information detected by the correction input detection unit 121. That is, the correction target handwritten input information is corrected so that the correction target handwritten input information is the same as the previously corrected handwritten input information. Thereby, the user can perform the correction again by utilizing the previously corrected handwritten input information.

  In addition, when the handwritten input information is corrected, the handwritten evaluation data 126a is also updated. For this reason, since the handwritten evaluation data 126a corresponding to the latest handwritten input information is accumulated, an erroneous determination is not made by incorrect handwritten evaluation data.

  As described above, when the handwritten input information is stored in order to display the handwritten input information itself, an error in the stored handwritten input information can be easily corrected.

  Although the configuration of the document processing apparatus 100 has been described in the first embodiment of the present invention, the present invention is of course not limited to the above-described embodiment. Various modifications can be made without departing from the spirit of the present invention. For example, the present invention includes the following cases.

  (1) In the above embodiment, an example in which a button for shifting to the correction input mode is provided in the input / output I / F unit 120 for performing the correction input is not limited. There may be a method such as providing a button to shift to the correction input mode on the input device, shifting to the correction input mode by writing special characters, or shifting to the correction mode by overwriting existing characters. .

(Embodiment 2)
Hereinafter, Embodiment 2 of the present invention will be described in detail with reference to the drawings.

  A difference from Embodiment 1 of the present invention is a processing sequence when handwritten character information is input by the user in the document processing apparatus 100. Since other configurations are basically the same, detailed description is omitted. Hereinafter, different points will be described in detail.

  A processing sequence when handwritten input information is input by the user in the document processing apparatus 100 will be described with reference to FIGS. 11 and 12.

  FIG. 11 is a flowchart illustrating an example of the operation of the document processing apparatus 100 when inputting handwritten input information according to Embodiment 2 of the present invention.

  FIG. 12 is a diagram illustrating a specific example when handwriting input correction is performed in the document processing apparatus 100.

  (Step S1101) The input / output I / F unit 120 accepts handwritten input information. Specifically, the user uses the input device 110 to input handwritten character information to the input / output I / F unit 120. The control unit 122 controls the input / output I / F unit 120 such as receiving input handwritten character information and displaying it on the input / output I / F unit 120.

  (Step S1102) Next, the handwritten character recognition engine 124 generates the character code information and the evaluation value of the handwritten input information received by the input / output I / F unit 120. Specifically, the control part 122 inputs the information of the input handwritten input information into the handwritten character recognition engine 124, and acquires character code information and its evaluation value. The timing when the control unit 122 inputs the handwritten input information to the handwritten character recognition engine 124 is, for example, after the input in units of one character is completed.

  (Step S1103) Next, the control unit 122 further extracts the handwritten input information when the evaluation value of the handwritten input information received by the input / output I / F unit 120 is a value within a predetermined second range. To do. Specifically, the control unit 122 compares whether or not the evaluation value obtained from the handwritten character recognition engine 124 is lower than a predetermined value (P2). When the evaluation value is larger than the predetermined value (P2), it is assumed that there is a low possibility of error, and the process proceeds to step S1106 as it is. When the evaluation value is lower than the predetermined value (P2), the control unit 122 displays whether or not correction is to be performed, assuming that there is a possibility of error. FIG. 12 shows a specific display example of the document processing apparatus 100 that displays whether or not to correct errors.

  (Step S1104) Next, the control unit 122 accepts whether or not the user corrects the typographical error. If the user does not correct the error, the control unit 122 proceeds to the process of step S1106.

  (Step S1105) When the user corrects an error, the control unit 122 performs processing such as accepting correction of handwritten characters by the user, corrects handwritten input information, and performs character recognition processing in step S1102. Try again. Specifically, the input / output I / F unit 120 accepts handwritten input information after the extracted handwritten input information is corrected. Then, the handwritten character recognition engine 124 generates an identifier and an evaluation value corresponding to the corrected handwritten input information.

  (Step S1106) When the evaluation value of the handwritten input information received by the input / output I / F unit 120 is not a value within the predetermined second range, the control unit 122 generates and corrects the handwritten input information after generation. The handwritten evaluation data 126a in which the identifier and the evaluation value are associated is stored. That is, the control unit 122 stores the stroke information, which is handwritten input information, and the character code information and evaluation value with the highest evaluation value obtained from the handwritten character recognition engine 124 in the input data storage unit 126. Store.

  As described above, the document processing apparatus 100 estimates that there is an error in the handwritten input information input by the input device 110, and notifies the user that it is easy for the user to notice and correct the error. Become.

  As described above, the document processing apparatus 100 according to the present invention stores the handwritten input information itself as digital data, converts the handwritten input information into character code information, and stores it together with the evaluation value at the time of the conversion. When correction of erroneous handwritten input information is performed by the user, based on the accumulated character code information before conversion and the evaluation value, by searching whether there is another same error, It is possible to realize an erroneous correction correction function that displays whether or not the same error is likely to be extracted and whether or not to correct the error. Thereby, the user can easily correct erroneous handwritten input information recorded in the past.

  In addition, the user can also correct the handwritten input information input immediately before by giving a warning that there is an error based on the evaluation value at the time of character code conversion of the handwritten input information.

  Although the document processing apparatus has been described in the second embodiment of the present invention, it is needless to say that the present invention is not limited to the above embodiment. Various modifications can be made without departing from the spirit of the present invention. For example, the present invention includes the following cases.

  (1) In the above embodiment, it is determined whether or not there is a typo at the time when the input in units of one character is completed, and if there is a possibility that it is a typo, whether or not correction is performed is displayed. Although an example is shown, the timing is not limited. It may be configured to display that there is a possibility that a typographical error is included after completion of input of handwritten input information in a predetermined unit such as a word unit, a phrase unit, or a page unit.

  Further, for example, the following cases common to the first and second embodiments are also included in the present invention.

  (1) In the above embodiment, the handwritten character recognition engine 124 has shown an example in which the character code conversion and the evaluation value are output in units of one character. However, the unit is not limited. A configuration in which character code conversion results and evaluation values are obtained in word units, phrase units, and page units may be used. As a result, erroneous determination is made for the handwritten input information for each predetermined unit. For this reason, it is possible to correct the handwritten input information that does not include an error as a character but includes an error as a word. For example, the word “complete wall” does not include an error in the word “wall” itself, but includes an error because the word “wall” must be “n”. Therefore, even in such a case, it is possible to easily correct errors in the stored handwritten input information.

  (2) In the above embodiment, the handwritten character recognition engine 124 has shown an example in which the character code conversion and the evaluation value are output in units of one character. However, the evaluation value is not obtained from the handwritten character recognition engine 124, There may be a dedicated engine for obtaining a value, and the evaluation value may be obtained from the engine. As long as the character code information and the evaluation value information are stored, any number and calculation methods may be used.

  (3) In the above-described embodiment, an example in which stroke information, character code information, and evaluation values are accumulated has been shown, but the number of evaluation values is not limited. It may be configured to hold the evaluation value in units of one character, and further calculate the evaluation value in units of words and phrases and hold it together with the evaluation value as described above. In addition, in the case of a configuration that holds a plurality of evaluation values in this way, a configuration that searches from a combination of a character code and a combination of a plurality of evaluation values as a search for an error and determines an error may be made. Any method may be used as long as it is a method for searching for errors based on the evaluation value.

  In the document processing apparatus and the document processing method according to the present invention, the handwritten input information is held and the character code conversion information and the evaluation value at the time of conversion are stored. When performing the above, it is possible to search for other erroneously written parts based on the character code conversion information and the evaluation value. Therefore, it can be applied as a function that can easily correct typographical errors in a document processing apparatus that accumulates handwritten input information.

1 is an external view showing a configuration of a document processing apparatus in Embodiment 1 of the present invention. It is a block diagram which shows the function structure of the document processing apparatus in Embodiment 1 of this invention. It is a figure explaining the structure of the stroke information in Embodiment 1 of this invention. It is a figure explaining the structure of the stroke information in Embodiment 1 of this invention. It is a figure explaining the structure of the stroke information in Embodiment 1 of this invention. It is a figure explaining the structure of the handwritten character recognition engine in Embodiment 1 of this invention. It is a figure explaining the structure of the handwritten character recognition engine in Embodiment 1 of this invention. It is a figure which shows an example of the handwritten evaluation data accumulate | stored in the input data storage part in Embodiment 1 of this invention. It is a flowchart which shows an example of operation | movement of the document processing apparatus at the time of the input of the handwritten input information in Embodiment 1 of this invention. It is a flowchart which shows an example of operation | movement of the document processing apparatus at the time of correction of the handwritten input information in Embodiment 1 of this invention. It is a figure which shows the specific example at the time of handwriting input correction in the document processing apparatus in Embodiment 1 of this invention. It is a figure which shows the specific example at the time of handwriting input correction in the document processing apparatus in Embodiment 1 of this invention. It is a figure which shows the specific example at the time of handwriting input correction in the document processing apparatus in Embodiment 1 of this invention. It is a flowchart which shows an example of operation | movement of the document processing apparatus at the time of the input of the handwritten input information in Embodiment 2 of this invention. It is a figure which shows the specific example at the time of handwriting input correction in the document processing apparatus in Embodiment 2 of this invention.

DESCRIPTION OF SYMBOLS 100 Document processing apparatus 110 Input device 120 Input / output I / F part 121 Correction input detection part 122 Control part 123 Character information search part 124 Handwritten character recognition engine 125 Assist information generation part 126 Input data storage part 126a Handwritten evaluation data 127 Recognition dictionary

Claims (8)

A document processing apparatus for storing handwritten input information input by handwriting,
An input unit for receiving the handwritten input information;
A handwritten information generating unit that generates an identifier for identifying the accepted handwritten input information and an evaluation value indicating a degree of error in the handwritten input information;
A storage unit that stores handwritten evaluation data in which the handwritten input information is associated with the generated identifier and the evaluation value;
A correction input detection unit for detecting correction of handwritten input information included in the stored handwriting evaluation data;
The handwritten input information included in the stored handwritten evaluation data is indicated by the same identifier as the identifier for identifying the handwritten input information whose correction has been detected, and the evaluation value is within a predetermined first range. A document processing apparatus comprising: an extraction unit that extracts handwritten input information that is a value. further,
A correction receiving unit that receives an instruction to correct the handwritten input information extracted by the extraction unit;
The document processing apparatus according to claim 1, further comprising: a correction unit that corrects the handwritten input information extracted by the extraction unit based on an instruction received by the correction reception unit. The document processing apparatus according to claim 2, wherein the correction unit corrects the handwritten input information extracted by the extraction unit into corrected handwritten input information of the handwritten input information detected by the correction input detection unit. The handwritten information generating unit generates an identifier and an evaluation value corresponding to the handwritten input information after correction of the handwritten input information whose correction is detected by the correction input detecting unit,
The document processing apparatus further includes:
An update unit that updates the handwritten evaluation data for the handwritten input information whose correction is detected by the correction input detection unit to handwritten evaluation data in which the corrected handwritten input information is associated with the generated identifier and evaluation value The document processing apparatus according to claim 1. The extraction unit further extracts the handwritten input information when the evaluation value of the handwritten input information received by the input unit is a value within a predetermined second range,
The input unit further accepts handwritten input information after the extracted handwritten input information is corrected,
The handwritten information generation unit generates an identifier and an evaluation value corresponding to the handwritten input information after being corrected,
The storage unit, when the evaluation value of the handwritten input information received by the input unit is not a value within a predetermined second range, the corrected handwritten input information, the generated identifier, and the evaluation The document processing apparatus according to any one of claims 1 to 4, wherein handwritten evaluation data associated with a value is stored. The handwritten information generation unit generates the identifier for identifying the handwritten input information in a predetermined unit, and the evaluation value indicating the degree of error in the handwritten input information in the predetermined unit,
The storage unit stores handwritten evaluation data in which the predetermined unit of handwritten input information is associated with the generated identifier and the evaluation value,
The correction input detection unit detects correction of handwritten input information of a predetermined unit included in the stored handwritten evaluation data of the predetermined unit;
The extraction unit is the same identifier as the identifier for identifying the handwritten input information of the predetermined unit in which the correction is detected from the handwritten input information of the predetermined unit included in the stored handwritten evaluation data of the predetermined unit 6. The document processing apparatus according to claim 1, wherein handwritten input information in a predetermined unit is extracted and the evaluation value is a value within a predetermined first range. A document processing apparatus for storing handwritten input information input by handwriting is a method for correcting the handwritten input information,
An input step for the document processing device to accept the handwritten input information;
The document processing device generates an identifier for identifying the accepted handwritten input information, and a handwritten information generating step for generating an evaluation value indicating a degree of error in the handwritten input information;
An accumulating step in which the document processing apparatus accumulates handwritten evaluation data in which the handwritten input information is associated with the generated identifier and the evaluation value;
A correction input detecting step in which the document processing device detects correction of handwritten input information included in the stored handwritten evaluation data;
The document processing device is indicated by the same identifier as the identifier for identifying the handwritten input information in which correction is detected from among the handwritten input information included in the accumulated handwritten evaluation data, and the evaluation value is predetermined. An extraction step of extracting handwritten input information that is a value within the first range of A document processing apparatus for storing handwritten input information input by handwriting is a program for correcting the handwritten input information,
An input step for receiving the handwritten input information;
A handwritten information generation step of generating an identifier for identifying the accepted handwritten input information and an evaluation value indicating a degree of error in the handwritten input information;
An accumulation step of accumulating, in a memory, handwritten evaluation data in which the handwritten input information is associated with the generated identifier and the evaluation value;
A correction input detection step for detecting correction of handwriting input information included in the handwriting evaluation data stored in the memory;
The handwritten input information included in the handwritten evaluation data stored in the memory is indicated by the same identifier as the identifier for identifying the handwritten input information whose correction has been detected, and the evaluation value is a predetermined first value. A program that causes a computer to execute an extraction step of extracting handwritten input information that is a value within a range.

Images