Methods for correcting ink data, information processing devices and computer program products

Abstract

The objective of this invention is to automatically correct ink data containing spelling errors in handwritten strings. The technical means is that the ink data correction method of this invention includes: steps S2 to S8, determining an ink data correction method by detecting spelling errors contained in the handwritten string represented by the ink data; and steps S10 to S12, correcting the ink data by operating the ink data according to the determined correction method.

Description

Technical Field

[0001] This invention relates to a method for correcting ink data, an information processing apparatus and a program, and more particularly to a method for correcting ink data representing strings, and an information processing apparatus and program for implementing such a correction method. Background Technology

[0002] With the increasing number of information processing devices (smartphones, tablets, laptops, etc.) that support input via styluses, digital handwriting is becoming increasingly popular. Digital handwriting enables features that were previously impossible with paper-based handwriting, such as reproducing the writing order and using writing speed as a benchmark.

[0003] When input is made using a stylus, the information processing device is configured to generate ink data containing one or more stroke data. Stroke data is data containing a series of coordinates representing the trajectory of the stylus's indicated position during the period from when the stylus tip contacts the contact surface until it leaves the contact surface. For example, when a user inputs a string using a stylus, the information processing device generates stroke data for each stroke and stores it as a series of ink data in a storage unit. The ink data generated in this way ultimately becomes data representing the string input by the user. An example of such ink data is disclosed in Patent Document 1.

[0004] Existing technical documents

[0005] Patent documents

[0006] Patent Document 1: Japanese Patent Application Publication No. 2019-036366 Summary of the Invention

[0007] The problem that the invention aims to solve

[0008] However, when users input strings by hand, spelling errors such as missing words, extra words, or misspellings may occur. These spelling errors are naturally reflected in the string represented by ink data. Therefore, it is desirable for the information processing device to automatically detect spelling errors and correct the ink data when handwritten strings represented by ink data contain spelling errors.

[0009] Therefore, one of the objectives of this invention is to provide a method, information processing apparatus, and program for automatically correcting ink data containing spelling errors.

[0010] Technical solutions for solving the problem

[0011] The ink data correction method of the present invention is a method for correcting ink data corresponding to a handwritten string, comprising the following steps: a determination step, which determines the ink data correction method by detecting spelling errors contained in the handwritten string represented by the ink data; and a correction step, which corrects the ink data by operating the ink data according to the correction method.

[0012] The information processing apparatus of the present invention includes: a correction method determination unit that determines a correction method for ink data by detecting spelling errors contained in a handwritten string represented by ink data; and a correction operation unit that corrects the ink data by operating the ink data according to the correction method.

[0013] The program of the present invention enables an information processing device to function as the following units: a correction method determination unit, which determines a correction method for the ink data by detecting spelling errors contained in a handwritten string represented by ink data; and a correction operation unit, which corrects the ink data by operating the ink data according to the correction method.

[0014] Invention Effects

[0015] According to the present invention, ink data corresponding to handwritten strings containing spelling errors can be automatically corrected. Attached Figure Description

[0016] Figure 1 This is a diagram illustrating an example of the structure and usage state of a tablet terminal 1 according to an embodiment of the present invention.

[0017] Figure 2 This is a diagram showing the internal structure of tablet terminal 1.

[0018] Figure 3 This is a diagram showing the function blocks of the main processor 10 and the data stored in the storage unit 11.

[0019] Figure 4 This is a diagram illustrating an example of the processing performed by the spelling error detection unit 20b.

[0020] Figure 5 This is a diagram illustrating the operation used to correct missing characters in ink data.

[0021] Figure 6 This diagram illustrates the operation used to correct the ink data for derivation.

[0022] Figure 7 This is a diagram illustrating the operation of ink data used to correct typos.

[0023] Figure 8 This is a flowchart representing the overall process of ink data correction processing executed by the main processor 10.

[0024] Figure 9 It means Figure 8 The flowchart shown details the character appending process.

[0025] Figure 10 It means Figure 8 The flowchart shown details the character deletion process.

[0026] Figure 11 It means Figure 8 The flowchart shown details the character replacement process.

[0027] Figure 12 This is a diagram showing the function blocks of the main processor 10 and the data stored in the storage unit 11 in a modified example of an embodiment of the present invention. Detailed Implementation

[0028] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

[0029] Figure 1 This diagram illustrates an example of the structure and usage of a tablet terminal 1 according to an embodiment of the present invention. The tablet terminal 1 is an information processing device for input via a stylus P, configured to have a panel surface 2 that serves as both a display surface and a contact surface. The input method using the stylus P is not limited; for example, active electrostatic discharge (AES) or electromagnetic induction (EMR) methods may be appropriately used. The following description will continue based on the premise of using an active electrostatic discharge method.

[0030] Figure 2 This is a diagram showing the internal structure of a tablet terminal 1. As shown in the diagram, the tablet terminal 1 is configured to have a main processor 10, a storage unit 11, a display 12, a contact sensor 13, and a sensor controller 14.

[0031] The main processor 10 is the central processing unit of the tablet terminal 1. By executing programs stored in the storage unit 11, it performs the functions of executing the operating system, drawing software, and various other applications of the tablet terminal 1. The drawing software includes the following functions: generating a file in the storage unit 11 for storing ink data; generating stroke data based on the coordinates and data successively supplied from the sensor controller 14 between the pen stroke and the pen lift (described later), drawing and displaying the stroke data sequentially on the display 12, and writing the generated stroke data as a series of interconnected ink data into the storage unit; and retrieving the generated ink data from the storage unit 11, drawing the ink data, and displaying it on the display 12.

[0032] The storage unit 11 is configured to include a main storage device, such as DRAM (Dynamic Random Access Memory), and an auxiliary storage device, such as flash memory or a hard disk. In addition to pre-stored programs executed by the main processor 10, the storage unit 11 also stores various data used or generated by the main processor 10. This data includes, in addition to the aforementioned ink data, identification information of the tablet terminal 1, login information for logging into the aforementioned operating system or various applications, etc.

[0033] The display 12 is a device that visually outputs the results of the processing of the main processor 10, such as a liquid crystal display or an organic EL display.

[0034] Contact sensor 13 is configured in Figure 1 The diagram shows a device with multiple sensor electrodes on the entire panel surface 2. The multiple sensor electrodes include multiple X electrodes extending along the Y direction and multiple Y electrodes extending along the X direction. Alternatively, a common electrode disposed within the display 12 can be used as multiple X electrodes, and the tablet terminal 1 configured in this way is referred to as "embedded".

[0035] The sensor controller 14 is an integrated circuit that uses the contact sensor 13 to detect the position of the stylus P and the finger within the panel surface 2 and to receive data transmitted by the stylus P. The data received from the stylus P includes a pen pressure value representing the pressure applied to the tip of the stylus P, a pen ID pre-assigned to each stylus P, etc. Regarding the pen pressure value, the sensor controller 14 is configured to detect the contact (pen placement) between the stylus P and the panel surface 2 and the removal (pen release) of the stylus P from the panel surface 2 based on the pen pressure value received from the stylus P. The sensor controller 14 sequentially supplies the coordinates of the detected position, the received data, and the pen placement or release information to the main processor 10.

[0036] The drawing software executing in the main processor 10 is configured to generate stroke data based on various data supplied from the sensor controller 14 and store it in the storage unit 11 in association with a pre-generated series of ink data. Specifically, regarding input from the stylus P, the drawing software is configured to generate stroke data by configuring a series of coordinates and pen pressure values ​​supplied from the sensor controller 14 in a time sequence from pen stroke to pen lift, and store it in association with a series of ink data in correspondence with user information indicating the user who input the stroke data. The specific content of the user information may be a pen ID supplied from the sensor controller 14, identification information of the tablet terminal 1, or login information of the operating system or drawing application.

[0037] The drawing software is also configured to retrieve generated ink data from the storage unit 11, draw it, and display it on the display 12. During this drawing process, interpolation curves, such as catmull-rom curves, are used to supplement the coordinates between a series of coordinates. Additionally, pen pressure values ​​are used to control line width or transparency.

[0038] return Figure 1 The string "patent aplication" shown in the figure is an example of a handwritten string represented by ink data. The tablet terminal 1 of this embodiment is characterized by automatically detecting spelling errors contained in such handwritten strings and automatically correcting the ink data based on that result. This aspect will be explained in detail below.

[0039] Figure 3 This is a diagram showing the function blocks of the main processor 10 and the data stored in the storage unit 11. This diagram only shows the parts related to the features described above.

[0040] like Figure 3 As shown, the main processor 10 is configured to functionally include a correction method determination unit 20 and a correction operation unit 21. The correction method determination unit 20 includes a character recognition processing unit 20a and a spelling error detection unit 20b. Furthermore, the correction method determination unit 20 and the correction operation unit 21 are preferably installed as functions in the aforementioned drawing software. Additionally, the storage unit 11 stores an ink database 30, a standard pattern database 31, and a word dictionary database 32.

[0041] The ink database 30 is a database that stores ink data generated by the main processor 10. The ink database 30 can store not only the ink data displayed by the main processor 10 of the tablet terminal 1, but also ink data generated by the main processors of other information processing devices.

[0042] The correction method determination unit 20 is a functional unit that determines the correction method for ink data by detecting spelling errors contained in a string represented by ink data stored in the ink database 30. The ink data to be detected is, for example, specified by the user. Specifically, the correction method determination unit 20 first performs character recognition processing using the character recognition processing unit 20a, thereby obtaining text data representing the string represented by the ink data to be processed. Next, the correction method determination unit 20 uses the spelling error detection unit 20b to detect spelling errors contained in the obtained text data. Finally, it determines the correction method for the ink data based on the content of the detected spelling errors. This will be explained in detail below.

[0043] The character recognition processing unit 20a is a functional unit that performs character recognition processing by using a standard pattern database 31, performing the same processing as so-called OCR (Optical Character Recognition). Specifically, when ink data is input, the character recognition processing unit 20a first analyzes its layout and cuts out lines and characters (including spaces). Next, the character recognition processing unit 20a extracts feature values ​​from each cut-out character.

[0044] The standard pattern database 31 stores a standard pattern for each character with the aforementioned feature values. The character recognition processing unit 20a, which extracts feature values ​​from each cut-out character, compares the extracted feature values ​​with the standard patterns in the standard pattern database 31 and selects one character for each cut-out character. The character recognition processing unit 20a outputs this selected series of characters as text data representing a string of ink data that is the object of processing.

[0045] Here, the text data output from the character recognition processing unit 20a is preferably stored in the ink database 30 in correspondence with the ink data to be processed. Therefore, when the correction operation unit 21 obtains the additional stroke data group (described later) from the ink database 30, it can determine the additional stroke data group to be obtained based on the text data stored in correspondence with each ink data.

[0046] Next, the spelling error detection unit 20b is a functional unit that uses a word dictionary database 32, which pre-stores data of words composed of combinations of one or more characters, to detect spelling errors in the text data output from the character recognition processing unit 20a. Specifically, the spelling error detection unit 20b first segments the text data output from the character recognition processing unit 20a into individual words. If it is a language that uses word segmentation, such as English, the segmentation can be performed based solely on spaces between characters. On the other hand, in the case of a language that does not use word segmentation, words are extracted by performing knowledge processing based on words stored in the word dictionary database 32, and segmentation is performed based on the results.

[0047] The spelling error detection unit 20b then determines whether any identical or similar words exist in the word dictionary database 32 for each of the more than one words obtained through segmentation. Here, two words are considered similar when, for example, the Levenstein distance or Yarro-Winkler distance between the words is below a predetermined value. If an identical word exists for a given word, the spelling error detection unit 20b concludes that the word does not contain a spelling error. Conversely, if neither an identical nor a similar word exists for a given word, the spelling error detection unit 20b concludes that the word does not contain a spelling error. This conclusion is based on the higher probability that the word is a proper noun or other word not found in the dictionary. On the other hand, if no identical word exists for a given word but a similar word exists, the spelling error detection unit 20b concludes that the word contains a spelling error.

[0048] Figure 4 This diagram illustrates an example of the aforementioned processing performed by the spelling error detection unit 20b. Figure 4 (a) represents ink data. Figure 4 (b) indicates based on Figure 4 (a) is the ink data output from the character recognition processing unit 20a as text data. Figure 4 In (b), the word delimiters 100 obtained as a result of the segmentation of the above-mentioned text data are also shown. Figure 4 (c) indicates that the spelling error detection unit 20b targets characters separated by delimiter 100. Figure 4 The text data of (b) consists of words found in the word dictionary database 32.

[0049] like Figure 4 As shown in (c), for the "patent" part of the text data, a completely identical word "patent" was found in the word dictionary database 32. Therefore, the spelling error detection unit 20b concludes that the word "patent" does not contain any spelling errors. On the other hand, for the "aplication" part of the text data, no completely identical word was found in the word dictionary database 32, but a similar word "application" was found. Therefore, the spelling error detection unit 20b concludes that the word "aplication" contains spelling errors.

[0050] return Figure 3The spelling errors detected by the spelling error detection unit 20b may include three types: missing characters, redundant characters, and misspelled characters. Missing characters are spelling errors where necessary characters are missing; redundant characters are spelling errors where unnecessary characters are mixed in; and misspelled characters are spelling errors where necessary characters are replaced with other characters. The correction method determination unit 20 determines the correction method for the ink data based on the detection results of these spelling errors. The determined correction method is to add characters in the case of missing characters, delete characters in the case of redundant characters, and replace characters in the case of misspelled characters.

[0051] Next, the correction operation unit 21 is a function that corrects ink data by operating the ink data based on the correction method determined by the correction method determination unit 20. Hereinafter, refer to... Figures 5-7 The processing performed by the correction operation unit 21 is explained in detail.

[0052] Figure 5 This is a diagram illustrating the operation used to correct missing characters in ink data. Figure 5 Example of ink data as the object of processing is shown in (a). The ink data in this example is... Figure 4 The word "patent aplication" shown in the text is missing a "p" compared to the correctly spelled "patent application".

[0053] By accepting Figure 5 The correction method determination unit 20 for the input ink data in (a) determines, for example, the correction method of appending "p" between "p" and "l" in "aplication". The correction operation unit 21 first starts from... Figure 3 The ink database 30 shown selects one or more ink data entries input by the same user who entered the ink data as the object of processing. Furthermore, the ink data selected here may include ink data that is currently being processed. Then, the correction operation unit 21 retrieves one or more stroke data entries from the selected ink data entries that represent the character "p" as the appended object. Hereinafter, the one or more stroke data entries retrieved in this way will be referred to as the "append stroke data set". Figure 5 The character 101 shown in (c) indicates that the additional stroke data group was obtained in this way.

[0054] Next, the correction operation unit 21 determines the width X of the character 101 to be added based on the size of each character in the ink data constituting the object of processing and the content of the character to be added. Then, among the multiple characters represented by the ink data constituting the object of processing, the character located on the right side of the same line as the insertion portion of the character 101 to be added is selected, and one or more selected characters are moved to the right according to the determined width X. Specifically, for each of the series of coordinates contained in the stroke data corresponding to the character to be moved, the values ​​of the moving distance are added together to move one or more selected characters to the right by a distance corresponding to the width X. Furthermore, the specific moving distance of each stroke data based on this operation is, for example, as follows: Figure 5 As shown in (b), it can be set to the width X plus the specified adjustment value α. However, the adjustment value α can be positive, negative, or zero.

[0055] Next, the correction operation unit 21 adjusts the values ​​of a series of coordinates constituting the additional stroke data group to make the drawing result of the ink data with the additional stroke data group appear natural. Specifically, the values ​​of the series of coordinates constituting the additional stroke data group are adjusted so that the character "p" displayed as a result of drawing the additional stroke data group is displayed at the position determined by the correction method determination unit 20 with a previously determined width X. In addition, the correction operation unit 21 can also adjust the insertion position of the additional stroke data group within the document constituting the ink data to make the reproduction order natural in the case of automatic ink data reproduction. After performing the adjustment, the correction operation unit 21 appends the adjusted additional stroke data group to the ink data and ends the ink data operation.

[0056] Here, in Figure 5 In the case of English, such as the example character, the series of strokes located to the right of the inserted stroke data group is moved to the right. However, depending on the language and its writing method, a different method may be used. For example, in languages ​​written from left to right, such as Arabic, the series of strokes located to the left of the inserted stroke data group is moved to the left. Additionally, in vertically written Japanese, the series of strokes located below the inserted stroke data group is moved downwards.

[0057] Figure 6 This diagram illustrates the operation used to correct ink data for defaced characters. Figure 6 (a) represents the ink data being processed in this example. The ink data is "patent application," with an unnecessary "l" appended compared to the correctly spelled "patent application." Figure 6In (a), the character 102 is equivalent to the "l".

[0058] By accepting Figure 6 The correction method determination unit 20 determines the correction method for the input ink data in (a) by deleting one "l" from the "appliedation". The correction operation unit 21 first obtains the width Y of the character "l" to be deleted from the ink data to be processed. Then, as... Figure 6 As shown in (b), one or more stroke data corresponding to the character "l" that is the object of deletion are deleted from multiple stroke data constituting ink data.

[0059] Then, the correction method determination unit 20 selects, from among the multiple characters represented by the ink data to be processed, a character located on one side of the same line as the deleted character "l" (in...). Figure 6 In the example (the characters on the right), the selected one or more characters are moved to the left based on the width Y. Specifically, the movement distance is subtracted from a series of coordinates contained in the stroke data corresponding to the character being moved, so that the selected one or more characters are moved to the left by a distance corresponding to the width Y. Furthermore, the specific movement distance of each character based on this operation is, for example, as shown below. Figure 6 As shown in (b), it can be set to the width Y plus the distance of the specified adjustment value β. However, the adjustment value β can be a positive value, a negative value, or zero. The correction operation unit 21 completes the operation of the ink data through the processing up to this point.

[0060] Here, in Figure 6 In the case of English, as in the example, the stroke data to the right of the deleted character is moved to the left. However, depending on the language and its writing system, a different method may be used. For example, in languages ​​written from left to right, such as Arabic, the stroke data to the left of the deleted character is moved to the right. Additionally, in vertically written Japanese, the stroke data below the deleted character is moved upwards.

[0061] Figure 7 This is a diagram illustrating the operation of ink data used to correct typos. Figure 7 (a) represents the ink data that is being processed in this example. This ink data is "patent application," with the "i" after "appl" replaced by an "e" compared to the correctly spelled "patentapplication." Figure 7 In (a), the character 103 is equivalent to the "e".

[0062] By accepting Figure 7The correction method determination unit 20 for the input ink data in (a) determines that the correction method replaces "e" in "applecation" with "i". First, the correction operation unit 21, which accepts this decision, first... Figure 7 (a) ~ Figure 7 As shown in (c), "e" is used as the deletion object, and the operation is performed with reference to... Figure 6 The operation described is the same as the operation of the ink data. Next, the correction operation unit 21 is as follows. Figure 7 (c) ~ Figure 7 As shown in (e), execution and reference Figure 5 The operation described is the same as the operation of the ink data, to append "i" between "l" and "c".

[0063] As a result of performing these operations, such as Figure 7 As shown in (e), the "e" in "applecation" is replaced with "i". Thus, the operation for correcting misspellings consists of a combination of operations for correcting extraneous characters and operations for correcting missing characters.

[0064] Next, the processing of the main processor 10 described above will be explained in detail from another perspective, referring to the processing flow of the main processor 10.

[0065] Figure 8 This is a flowchart illustrating the overall process of ink data correction processing executed by the main processor 10. As shown in the figure, firstly, the main processor 10 retrieves ink data from the ink database 30 (step S1) and then performs character recognition processing (step S2). The details of this character recognition processing are as described above.

[0066] Next, the main processor 10 segments the text data obtained through character recognition processing for each word (step S3). The details of this segmentation process are as described above. Afterward, the main processor 10 performs the processing steps S5 to S13 for each word in the text data obtained through the segmentation in step S3 (step S4).

[0067] Specifically, the main processor 10 first determines that... Figure 3The main processor 10 checks if a matching word exists in the dictionary database 32 (step S5). If it does, the main processor 10 concludes that there is no spelling error and moves on to the next word. If not, the main processor 10 further checks if a similar word exists in the dictionary database 32 (step S6). "Similar" here means as described above. If it is determined in step S6 that no word exists, the main processor 10 concludes that there is no spelling error and moves on to the next word. If it is determined that a similar word exists, the main processor 10 detects spelling errors in the word by comparing it with similar words in the dictionary database 32 (step S7). As described above, the detected spelling errors may include three types: missing words, redundant words, and misspelled words.

[0068] Next, the main processor 10 determines the ink data correction method based on the content of the spelling errors detected in step S7 (step S8). Here, the content of the spelling errors detected in step S7 is not limited to just one. In the case where a word contains multiple spelling errors, the main processor 10 selects one of them and determines the correction method used to correct the selected spelling error.

[0069] Next, the main processor 10 determines which of the correction methods decided in step S8 is character appending, character deletion, or character replacement (step S9). As a result, if it is character appending, character appending is performed (step S10); if it is character deletion, character deletion is performed (step S11); and if it is character replacement, character replacement is performed (step S12). Details of each step will be provided later. Figures 9-11 A detailed explanation will be provided.

[0070] After completing any one of steps S10 to S12, the main processor 10 determines whether it has completed the correction of all spelling errors contained in the word being processed (step S13). If it determines that the correction is not complete, it returns to step S8 to correct the remaining spelling errors. On the other hand, if the correction is complete, the main processor 10 moves on to the next word. When all words in the text data obtained by segmentation in step S3 have been processed, the main processor 10 ends the ink data correction process.

[0071] Figure 9 It means Figure 8 The flowchart shows the detailed content of the character appending process (step S10). The main processor 10, which performs this process, first selects the ink data (in the ink database 30) that is the object of the process from the ink database 30. Figure 8(Step S20) Obtain the additional stroke data group mentioned above from one or more ink data generated by the same user as the ink data obtained in step S1.

[0072] Next, the main processor 10 determines the width X of the character to be appended based on the size of each character constituting the ink data to be processed and the content of the character to be appended (step S21). Then, the main processor 10 selects, from the multiple characters represented by the ink data to be processed, the character located on the right side of the same line as the insertion portion of the character to be appended (step S22), and operates on the ink data to move one or more selected characters to the right by a distance corresponding to the width X (specifically, ...). Figure 5 The distance X + α shown (step S23). The details of this operation are as described above.

[0073] Next, the main processor 10 adjusts the values ​​of a series of coordinates that constitute the appended stroke data group to make the drawing result of the ink data with the appended stroke data group natural (step S24). The details of this adjustment process are as described above. Afterward, the main processor 10 appends the adjusted appended stroke data group to the ink data (step S25), ending the character appending process.

[0074] Figure 10 It means Figure 8 The flowchart shows the detailed process of character deletion (step S11). The main processor 10, which performs this process, first exports the width Y of the character to be deleted from the ink data (step S30). Then, the main processor 10 deletes one or more strokes corresponding to the character to be deleted from the ink data (step S31).

[0075] Next, the main processor 10 selects, from among the multiple characters represented by the ink data to be processed, the character located on the right side of the same line as the character to be deleted (step S32). Then, it operates on the ink data to be processed so that the selected one or more characters are moved to the left by a distance corresponding to the width Y (specifically, Figure 6 The distance Y + β shown is (step S33). The details of this operation are as described above. The main processor 10 terminates the character deletion process by completing step S33.

[0076] Figure 11 It means Figure 8 The flowchart shows the detailed content of the character replacement process (step S12). As shown in the figure, this process is a combination of character deletion and character appending. Specifically, the main processor 10 first selects the erroneous character as the deletion target and executes... Figure 10 The character deletion process shown is in step S40. Here, erroneous characters are deleted from the ink data being processed, and spaces created by the deletion are filled in. Next, the main processor 10 executes... Figure 9 The character appending process shown is used to append the correct character to the position where the erroneous character was deleted (step S41). Thus, the correct character is inserted into the ink data being processed at the position where the erroneous character was deleted. After completing step S41, the main processor 10 ends the character replacement process.

[0077] As described above, the ink data correction method, information processing apparatus, and program according to this embodiment can automatically detect spelling errors contained in a string represented by ink data and correct the ink data based on the result. Therefore, ink data containing spelling errors can be automatically corrected.

[0078] Furthermore, when correcting additional characters, since the additional stroke data group is obtained from the ink data previously entered by the same user who entered the ink data to be processed, it is possible to prevent characters with different handwriting from being mixed into the ink data.

[0079] Furthermore, after adjusting the values ​​of a series of coordinates constituting the additional stroke data group to make the drawing result of the ink data with the additional stroke data group more natural, the additional stroke data group is added to the ink data, so a natural drawing result can be obtained after correction.

[0080] The preferred embodiments of the present invention have been described above, but the present invention is not limited to such embodiments. The present invention can of course be implemented in various ways without departing from its spirit.

[0081] For example, in Figure 3 The text describes the case where more than one piece of ink data is selected from the ink database 30, which is input by the same user as the user who input the ink data as the object of processing. However, the ink data selected from the ink database 30 is not necessarily ink data input by the same user.

[0082] In addition, it was explained that in Figure 3In the correction method determination unit 20 shown, character recognition processing is performed by the character recognition processing unit 20a to obtain text data representing the string represented by the ink data as the processing object. However, it is not always necessary to obtain the text data. That is, in the correction method determination unit 20, character recognition processing can also be performed by the character recognition processing unit 20a to correct the size, tilt, etc. of the handwritten string represented by the ink data as the processing object, thereby standardizing the handwritten string. The standardized handwritten string is then pre-registered in the storage unit 11 instead of the text data. Thus, the correction method determination unit 20 uses the spelling error detection unit 20b to detect spelling errors contained in the obtained handwritten string, with the handwritten string itself as the target.

[0083] Figure 12 This diagram illustrates the function blocks of the main processor 10, configured as described above, and the data stored in the storage unit 11 in a modified embodiment of the above-described embodiment. As shown in the diagram, the main processor 10 in this modified embodiment has a character recognition processing unit 20c instead of a character recognition processing unit 20a, and a spelling error detection unit 20d instead of a spelling error detection unit 20b. Furthermore, the storage unit 11 has a handwritten string database 33 instead of a word dictionary database 32.

[0084] The Handwritten String Database 33 is a database that stores standardized handwritten strings. The Handwritten String Database 33 is structured to store instances of handwritten strings for as many words as possible.

[0085] The character recognition processing unit 20c standardizes strings by performing character recognition processing on ink data, which is the object of spelling error detection. Specifically, the character recognition processing unit 20c first cuts out lines and characters by performing the same processing as the character recognition processing unit 20a, and selects one character for each cut-out character. Then, the cut-out characters are shaped so that the shape of the selected character is close to the desired shape (i.e., the shape of the characters stored in the handwritten string database 33). As mentioned above, this shaping includes tilt correction, etc. In addition, the size of each shaped character is adjusted so that the size of each cut-out character is, for example, equal to each other.

[0086] The spelling error detection unit 20d detects spelling errors by comparing the shaped handwritten string output from the character recognition processing unit 20c with the handwritten strings registered in the handwritten string database 33. Specifically, the spelling error detection unit 20d first segments the handwritten string output from the character recognition processing unit 20c into words. This process can be the same as the text data segmentation process performed by the spelling error detection unit 20b. Then, for each word obtained through segmentation, it is determined whether a matching or similar handwritten string exists in the handwritten string database 33. If a matching handwritten string exists for a word, the spelling error detection unit 20d concludes that the word does not contain spelling errors. Similarly, if neither a matching nor a similar handwritten string exists for a word, the spelling error detection unit 20d also concludes that the word does not contain spelling errors. On the other hand, if no matching handwritten string exists for a word but a similar handwritten string exists, the spelling error detection unit 20d concludes that the word contains spelling errors. The correction method decision unit 20 and the correction operation unit 21, having accepted this conclusion, proceed as described in the above embodiment.

[0087] Thus, according to this variation, even without obtaining text data, it is possible to automatically correct ink data containing spelling errors in the same way as in the above-described embodiments.

[0088] Furthermore, in handwritten strings, such as English handwriting, there are cases where characters are connected. Therefore, the correction operation unit 21 can also determine the two stroke data that should be connected after the addition when appending stroke data groups to ink data, and change the shape (including position and / or thickness) of one or both of the determined two stroke data in a connected manner. Similarly, when deleting one or more stroke data and moving characters located on one side of the same line as the deleted character to the other side, it can determine the two stroke data that should be connected after the movement operation, and change the shape (including position and / or thickness) of one or both of the determined two stroke data in a connected manner. Thus, it is possible to achieve continuous strokes that connect characters.

[0089] in addition, Figure 3 The ink database 30, standard pattern database 31, and word dictionary database 32 shown are... Figure 12 At least one of the handwritten string databases 33 shown can also be a structure connected to the main processor 10 via a network.

[0090] Furthermore, cloud computing technology can also be used to execute over the network. Figure 3The main processor 10 shown performs some or all of the processing.

[0091] Label Explanation

[0092] 1. Tablet terminal

[0093] 2 Panel

[0094] 10 main processors

[0095] 11 Storage Department

[0096] 12 monitors

[0097] 13 Contact Sensors

[0098] 14 Sensor Controller

[0099] 20. Correction Method Decision Department

[0100] Character recognition processing unit 20a, 20c

[0101] 20b, 20d Spelling Error Detection Department

[0102] 21 Correction Operations Department

[0103] 30 Ink Database

[0104] 31 Standard Pattern Database

[0105] 32-word dictionary database

[0106] 33 Handwritten String Database

[0107] 101 Characters as append objects

[0108] 102 as the character to be deleted

[0109] 103 Replace characters in the object

[0110] P stylus

[0111] X and Y widths

[0112] α, β adjustment values

Claims

1. A correction method of ink data corresponding to a handwritten character string, comprising the steps of: a decision step of deciding a correction method of the ink data by automatically performing detection of a spelling error contained in the handwritten character string represented by the ink data; and a correction step of correcting the ink data by operating the ink data in accordance with the correction method.

2. The correction method of ink data corresponding to a handwritten character string according to claim 1, wherein the decision step acquires text data representing the handwritten character string by performing a character recognition process, and performs detection of a spelling error contained in the handwritten character string by performing detection of a spelling error contained in the text data.

3. The correction method of ink data corresponding to a handwritten character string according to claim 1, wherein the decision step performs normalization of the handwritten character string by performing a character recognition process, and performs detection of a spelling error contained in the handwritten character string by comparing with a handwritten character string registered in advance.

4. The correction method of ink data corresponding to a handwritten character string according to any one of claims 1 to 3, wherein the correction method is addition of a character, and the operation of the ink data performed by the correction step includes an operation of adding one or more stroke data representing a character to be added to a plurality of stroke data constituting the ink data.

5. The correction method of ink data corresponding to a handwritten character string according to claim 4, wherein the correction step acquires one or more stroke data representing the character to be added from one or more ink data input by a user who input the ink data.

6. The correction method of ink data corresponding to a handwritten character string according to claim 5, wherein the operation of the ink data performed by the correction step includes an operation of deciding a width of the character to be added, and moving a character located on one side of a portion where the character to be added is to be inserted among a plurality of characters constituting the handwritten character string to the one side in accordance with the decided width.

7. The correction method of ink data corresponding to a handwritten character string according to claim 4, wherein the operation of the ink data performed by the correction step includes an operation of deciding two stroke data to be connected after the addition of one or more stroke data, and changing a shape of one or both of the two stroke data to be connected so as to make the two stroke data to be connected.

8. The correction method of ink data corresponding to a handwritten character string according to any one of claims 1 to 3, wherein the correction method is deletion of a character, and the operation of the ink data performed by the correction step includes an operation of deleting one or more stroke data corresponding to a character to be deleted from a plurality of stroke data constituting the ink data.

9. The correction method of ink data corresponding to a handwritten character string according to claim 8, wherein ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ The operation of the ink data by the correction step includes an operation for moving a character located on one side of the deleted character to the other side among a plurality of characters constituting the handwritten character string according to the width of the character as the deletion target.

10. The correction method of ink data corresponding to a handwritten character string according to claim 9, wherein The operation of the ink data by the correction step includes an operation of deciding two stroke data that should be connected after the operation for the movement is performed, and changing the shape of one or both of the two stroke data in such a manner that the decided two stroke data become connected.

11. The correction method of ink data corresponding to a handwritten character string according to any one of claims 1 to 3, wherein The correction method is replacement of a character, The operation of the ink data by the correction step includes an operation of: deleting one or more stroke data corresponding to an erroneous character from a plurality of stroke data constituting the ink data; and adding one or more stroke data representing a correct character to a plurality of stroke data constituting the ink data.

12. The correction method of ink data corresponding to a handwritten character string according to claim 11, wherein The correction step acquires one or more stroke data representing the correct character from one or more ink data input by a user who inputs the ink data.

13. The correction method of ink data corresponding to a handwritten character string according to claim 12, wherein The operation of the ink data by the correction step includes an operation of: an operation for moving a character located on one side of the deleted character to the other side among a plurality of characters constituting the handwritten character string according to the width of the character as the deletion target; and an operation for deciding the width of the correct character, and moving a character located on one side of a portion where the correct character is inserted among a plurality of characters constituting the handwritten character string to the one side according to the decided width.

14. An information processing apparatus comprising: a correction method deciding section that decides a correction method of ink data by automatically performing detection of a spelling error contained in a handwritten character string represented by the ink data; and a correction operation section that corrects the ink data by operating the ink data according to the correction method.

15. The information processing apparatus according to claim 14, wherein the correction method deciding section acquires text data representing the handwritten character string by performing character recognition processing, and performs detection of a spelling error contained in the ink data by performing detection of a spelling error contained in the text data.

16. A computer program product comprising a computer program, wherein, The computer program is for causing an information processing apparatus to function as: a correction method deciding section that decides a correction method of ink data by automatically performing detection of a spelling error contained in a handwritten character string represented by the ink data; and a correction operation section that corrects the ink data by operating the ink data according to the correction method.

17. The computer program product according to claim 16, wherein The correction method determination section acquires text data representing the handwritten character string by performing character recognition processing, and performs detection of a spelling error included in the ink data by performing detection of a spelling error included in the text data.

Images