JP2006323502A - Information processor, and its control method and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an information processor, and its control method and program capable of efficiently and accurately executing handwritten character recognition. <P>SOLUTION: The information processor recognizes a character corresponding to manually inputted trajectory data, and post-processes the candidate character string of the character recognition result. Based on processing type due to a command for instructing execution of various processings including input processing and processing to the candidate of the character recognition result, the information processor determines whether a newly inputted character is connected to the character string to be post-processed. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an information processing apparatus that inputs trajectory data of handwritten characters, recognizes characters, and performs post-processing on the character recognition results, a control method thereof, and a program.

  Conventionally, there is a character recognition device that recognizes a locus (stroke) of an input handwritten character. In this type of character recognition device (for example, Patent Document 1), in order to designate a word break, the user inputs a stroke in a specific area near the character input frame and designates it, thereby clarifying the position of the break. Need to be directed.

Moreover, in the on-line handwritten character recognition apparatus that performs post-processing in units of character strings, the handwritten character string input until the recognition start button is pressed is set as a target for post-processing.
JP-A-6-208655

  However, in the conventional character recognition device, the device cannot recognize a word break unless the user inputs a word break each time. Therefore, the user is forced to perform input work other than the originally intended character input at the time of input, and natural usability is impaired.

  When post-processing is performed in units of character strings, handwritten character strings input until the recognition start button is pressed are set as post-processing targets. Therefore, if the recognition start button is pressed during the input of the character string, the characters input after that are not linked to the recognized characters, and the recognition rate is not lowered or the intended recognition result cannot be obtained. There is a case.

  SUMMARY An advantage of some aspects of the invention is that it provides an information processing apparatus that can efficiently and accurately execute handwritten character recognition, a control method therefor, and a program.

In order to achieve the above object, an information processing apparatus according to the present invention comprises the following arrangement. That is,
An information processing apparatus that inputs handwritten character trajectory data, recognizes characters, and performs post-processing on the character recognition results,
Input means for inputting the locus data;
Character recognition means for recognizing characters corresponding to the trajectory data input by the input means;
Post-processing means for post-processing candidate character strings of character recognition results by the character recognition means;
Instruction means for instructing execution of various processes including input processing by the input means and processing for the character recognition result candidates;
Based on the processing type instructed to be executed by the instruction means, a new character input after the instruction by the instruction means is continued to the post-processing character string that has already been post-processed by the post-processing means. Determination means for determining whether or not.

  Preferably, the apparatus further includes insertion means for inserting, as the newly input character, a character whose character code corresponding to the newly input character is fixed.

  Preferably, the determination unit is newly input as a character recognition result by the character recognition unit with respect to trajectory data input by the input unit when the processing type is input processing by the input unit. The candidate character is continued to the post-processing character string that has already been post-processed by the post-processing means.

  Preferably, the input unit, the instruction unit, an operation screen including a candidate display area for displaying candidate characters as a result of character recognition by the character recognition unit, and an application that operates on the information processing apparatus are generated. Display means for displaying an application screen for switching and a switching operation screen for switching between a plurality of types of applications operating on the information processing apparatus.

Preferably, the input means includes an input area for inputting the trajectory data,
The instruction means includes
A changing unit for changing candidate characters on the candidate display area;
A conversion unit for converting a candidate character on the candidate display area to a kana-kanji character;
A deletion unit for deleting the candidate character before the cursor displayed on the candidate display area is located;
A blank character insertion portion for inserting a blank character on the candidate display area;
A candidate character string determining unit for determining a candidate character string on the candidate display area and inserting the candidate character string on the application screen.

  Preferably, the determination unit is configured such that the processing type is input processing for the input area, timer event processing that occurs when a non-input time for the input area has passed a predetermined time, deletion processing by the deletion unit, If the switching operation screen is one of the switching processes, a candidate character newly input as a character recognition result by the character recognition unit with respect to the trajectory data input to the input area is already updated by the post-processing unit. Continue to the post-processing character string to be processed.

Preferably, the information processing apparatus further includes an insertion unit that inserts characters stored in a storage area secured in the internal memory by the operating system that controls the information processing apparatus,
In the case where the type of processing is insertion processing by the insertion means, the determination means determines a character newly input by the insertion processing as a post-processing character that has already been post-processed by the post-processing means. Continue to the column.

  Preferably, the post-processing means calculates a chain probability between candidate character strings of the character recognition result.

In order to achieve the above object, a method for controlling an information processing apparatus according to the present invention comprises the following arrangement. That is,
A method for controlling an information processing apparatus that inputs handwritten character trajectory data, recognizes characters, and performs post-processing on the character recognition results.
An input step for inputting the trajectory data;
A character recognition step for recognizing characters corresponding to the trajectory data input in the input step;
A post-processing step of post-processing the candidate character string of the character recognition result by the character recognition step;
An instruction process for instructing execution of various processes including an input process by the input process and a process for the candidate character recognition result;
Based on the processing type instructed to be executed by the instruction process, new characters input after the instruction by the instruction process are continued to the post-processing character string that has already been post-processed in the post-processing process. A determination step of determining whether or not.

In order to achieve the above object, a program according to the present invention comprises the following arrangement. That is,
A program that realizes control of an information processing apparatus that inputs trajectory data of handwritten characters, recognizes characters, and performs post-processing on the character recognition results,
A program code of an input process for inputting the locus data;
A program code of a character recognition step for recognizing characters corresponding to the trajectory data input in the input step;
A program code of a post-processing step of post-processing a candidate character string of a character recognition result by the character recognition step;
A program code of an instruction process for instructing execution of various processes including an input process by the input process and a process for the candidate character recognition result;
Based on the processing type instructed to be executed by the instruction process, new characters input after the instruction by the instruction process are continued to the post-processing character string that has already been post-processed in the post-processing process. And a program code of a determination process for determining whether or not.

  ADVANTAGE OF THE INVENTION According to this invention, the information processing apparatus which can perform handwritten character recognition efficiently and accurately, its control method, and a program can be provided.

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

  In the present invention, a post-processing continuation determining unit corresponding to the type of user's instruction input is provided in an information processing apparatus that performs post-processing of character strings by handwriting input in units of character strings. The post-processing continuation determination unit will explain a configuration for realizing post-processing in a natural and appropriate processing unit even if the user does not particularly instruct instruction input such as ending or continuation of a sentence. . Thereby, in this invention, it can implement | achieve raising the probability that the character used as the recognition result with respect to the handwritten input which a user inputs will turn into the recognition result which a user intends.

  In addition, an insertion unit is provided for inserting a general text character string (character code string) composed of character codes as a part of the character string to be post-processed. A configuration will be described in which post-processing is normally realized even when a text character string is inserted into the post-processing target character recognition result editing area by the insertion unit. Thereby, even when the recognition result of a handwritten character and a general text character string coexist, the input efficiency of the text which a user intends can be improved.

<Embodiment 1>
FIG. 1 is a diagram illustrating an example of a usage pattern of the character recognition device according to the first embodiment of the present invention. FIG. 2 is a diagram illustrating a hardware configuration of the portable information device that functions as the character recognition device according to the first embodiment of the present invention.

  Reference numeral 1 denotes a portable information device such as a PDA (personal digital assistant), a mobile phone, a digital camera, or a notebook PC (personal computer).

  This portable information device 1 is for inserting a display unit 2, an indicator 3, a camera unit 4, a coordinate input unit 5, a CPU 6, a RAM 7, a ROM 8, and a storage medium (for example, smart media, compact flash (registered trademark), etc.) Media I / F (interface) 9 having a slot, keyboard (a soft keyboard displayed on the display unit 2 or a normal keyboard) 10, an external I / F (USB, Bluetooth, etc.) for connecting to other information devices 11) A network I / F 12 is provided.

  The portable information device 1 displays an image captured by the camera unit 4 on the liquid crystal display unit 2, and uses trajectory data (user's handwriting input operation) to input additional information such as memo information added to the image by the coordinate input unit 5. Handwritten trajectory data).

  The CPU 6 is connected to various components of the portable information device 1 such as the RAM 7, the ROM 8, the coordinate input unit 5, and the display unit 2 via the system bus 13. Further, the CPU 6 executes various controls according to programs stored in the ROM 8.

  The RAM 7 stores various data such as image data, input trajectory data, and position information thereof, and functions as a data work area and a temporary save area.

  The ROM 8 stores a control program for portable information equipment including various programs for realizing the present invention. More specifically, the ROM 8 recognizes handwritten character recognition in addition to various programs such as a post-processing control processing program for inputting handwritten characters, a camera unit control program, a display unit control program, and a handwriting locus data recognition program, which will be described later. Data such as recognition dictionary data, post-processing dictionary data (chain probability calculation dictionary data), various tables, setting data, and the like are stored.

  The input locus data can be recognized by a corresponding character code by a conventional handwritten character recognition function and input to the portable information device 1. In addition, the input trajectory data is added to the photographed image as an additional information as an image (image data) as it is without being recognized, or the text data obtained by the character recognition function is added to the photographed image as additional information. be able to.

  The image data of the captured image to which the additional information is added can be transmitted to the external terminal 50 such as a personal computer via the external I / F 11 by a user instruction operation. Alternatively, after storing the image data on a storage medium, the image data can be stored or printed by mounting the image data on an external terminal 50 or a printer having a media slot.

  Further, the image data is attached to an e-mail created by the e-mail function realized by the portable information device 1, and then on the network through a network such as a local network or a public line connected via the network I / F 12. It is also possible to transmit to other external terminals.

  Reference numeral 2 denotes a display unit, which includes, for example, a liquid crystal or an LED. When the display unit 2 is composed of a liquid crystal display unit, it includes a liquid crystal display element, a liquid crystal control circuit, and a display memory. The display unit 2 is connected to the CPU 6 via the system bus 13. Further, in accordance with an instruction from the CPU 6, image data and locus data are displayed, and various operation screens are displayed.

  By arranging the coordinate input unit 5 so as to overlap the upper surface of the display unit 2, a general transparent resistance film digitizer (so-called touch panel) that performs coordinate input can be realized. The user can input trajectory data by pressing the coordinate input surface (display screen of the display unit 2) of the coordinate input unit 5 with the pointing tool 3.

  More specifically, the coordinate input unit 5 is connected to the CPU 6 via the system bus 13 via an internal control circuit (not shown). Thereby, the position coordinate data of the pressed position is notified by the CPU 6 by pressing on the coordinate input surface of the coordinate input unit 5 using the pointing tool 3.

  Based on the position coordinate data, the portable information device 1 displays an operation screen such as a file list display screen by a GUI (graphic user interface) realized on the display unit 2 according to the processing procedure in the ROM 8. Also, the contents of the data file corresponding to the data file icon (display representing the data file) on the operation screen at the position indicated by the input position coordinate data are displayed, and the image corresponding to the input position is edited. Various processes such as performing can be executed.

  Further, by using the pointing tool 3, it is possible to realize execution instructions and screen operations of various functions realized by the portable information device 1. For example, the above-described electronic mail function is also used by using the pointing tool 3. can do.

  An indicator 3 is used for inputting handwritten locus data, soft buttons on the display screen, soft keyboard, and the like.

  A camera unit 4 includes components such as a lens, a CCD 4a, and a CCD control circuit (not shown) that are components of a general digital camera. The camera unit 4 is controlled by the CPU 6 in the portable information device 1, and the captured image data is temporarily stored in the RAM 7 under the control of the CPU 6. As described above, an image captured by the camera unit 4 can be attached to an e-mail and transmitted.

  More specifically, the CCD control circuit of the camera unit 4 is connected to the CPU 6 via the system bus 11. For this reason, the operation of the CCD 6 is controlled by the CPU 6 via the CCD control circuit, and a photographed image by the CCD 6 is stored in the RAM 7.

  The media I / F 9 is an interface unit of a storage medium generally used in a digital camera or the like, for example, a compact flash (registered trademark) card. In particular, the media I / F 9 can read / write data in the connected media and store data such as image data.

  The keyboard 10 is a general keyboard, and a key code such as alphanumeric characters is input by pressing a key. The input key code is read by the CPU 6 via the system bus 13.

  The external I / F 11 is, for example, a USB interface unit. This is controlled by the CPU 6 and can be connected to an external terminal such as a personal computer to exchange various data such as trajectory data and image data.

  The network 1 / F12 is composed of a communication card, for example. This has a built-in controller and antenna for connecting to a public line such as a PHS line. By passing a prescribed command to the controller in the communication card, it is possible to connect to the public line and send / receive e-mail.

  Further, if this communication card is changed to a network card such as a local area network card, it is possible to support local area network connection. The protocol for sending and receiving e-mail here is the same as the conventional method.

  The system bus 13 is a communication path that allows various components of the portable information device 1 including the CPU 6, the RAM 7, and the ROM 8 to be connected to each other to transmit and receive data.

  Reference numeral 50 denotes an external terminal such as a personal computer. The external terminal 50 has standard components (for example, CPU, RAM, ROM, hard disk, external storage device, network interface, external interface, display, keyboard, mouse, etc.) mounted on a general-purpose computer. Yes.

  In the first embodiment, the external terminal 50 can be connected to the portable information device 1 via a general external interface such as a USB or a network, and can transmit and receive the image data 6.

  Next, a functional configuration of the portable information device according to the first embodiment will be described with reference to FIG.

  FIG. 3 is a diagram showing a functional configuration of the portable information device according to the first embodiment of the present invention.

  In addition to the various functional configurations of the portable information device (enclosed by the broken line portion), FIG. 3 shows the coordinate input unit 5 and the display unit 2 that are hardware for the sake of explanation. This functional configuration may be realized, for example, by a control program stored in the ROM 8 under the control of the CPU 6, or a part / all of it may be realized by a dedicated hardware circuit.

  In FIG. 3, reference numeral 31 denotes a processing type determination unit that determines the processing type according to the range of position coordinate data received from the coordinate input unit 2. Then, processing for distributing the input coordinate data to each processing unit is performed according to the determined processing type.

  Reference numeral 32 denotes a text editing processing unit which executes editing processing such as deletion, kana-kanji conversion, and insertion into a character input area of an application on the text character string obtained by the handwritten character recognition unit 35.

  Reference numeral 33 denotes a post-processing continuation determination unit that determines continuation and termination (stop) of the post-processing character string (character string to be post-processed) according to each processing type received from the processing type determination unit 31. The result is transmitted to the post-processing unit 34.

  Reference numeral 34 denotes a post-processing unit. Here, in particular, post-processing in units of character strings is executed, and as the algorithm, for example, an algorithm similar to the algorithm described in JP-A-10-198759 can be used. Specifically, the handwritten character recognition unit 35 recognizes the candidate handwritten character string by considering the chain probability between the characters and the degree of coincidence between the standard character dictionary data in the recognition dictionary and the like. Determine recognition candidates for character strings.

  Note that post-processing is not limited to this, and any post-processing may be used as long as it is post-processed on a character string obtained by recognition processing such as morphological analysis or statistical processing.

  Reference numeral 35 denotes a handwritten character recognizing unit that receives a position coordinate data string of handwritten characters from the processing type discriminating unit 31 and outputs a candidate character code by matching processing with the standard character dictionary data in the recognition dictionary. .

  A recognition candidate display selection unit 36 performs a selection instruction process for handwritten character recognition candidates. This process itself is a conventional process. For example, when the display of the first candidate of a handwritten character is tapped with the pointing tool 3, a candidate group from the first candidate to the fifth candidate of the character is displayed as a candidate. Display in the area. Then, when the arbitrary candidate is tapped from the candidate group, a process of replacing the tapped candidate character with the first candidate displayed is performed.

  In addition, although this functional configuration has been described by taking as an example a case where it is realized on the portable information device 1, it is not limited to this. That is, it goes without saying that the functional configuration shown in FIG. 3 of the present invention can be realized even in a general-purpose terminal (information processing apparatus) such as a personal computer having a function equivalent to or higher than that of the portable information device.

  In the present invention, in various processes realized in the handwritten character recognition environment, the post-processing continuation determining unit 33 determines whether or not to continue the post-processing in the post-processing in character string units, in particular, depending on the processing type. In order to realize this determination by the post-processing continuation determination unit 33, for example, a post-processing continuation determination table is stored on the RAM 7.

  Here, the post-processing continuation determination table will be described with reference to FIG.

  FIG. 4 is a diagram showing a post-processing continuability determination table according to the first embodiment of the present invention.

  This post-processing continuation determination table determines whether to continue the processing type and whether or not to continue the post-processing (the entire character string including new characters that are continued in the candidate character string of the already obtained character recognition result Information indicating whether or not to be a character string to be processed is managed in association with each other.

  For example, when the process type determined by the process type determination unit 31 is the candidate change process for the handwritten character recognition candidate, the post-processing continuation determination unit 33 refers to the post-processing continuation determination table to determine at that time. Then, it is determined that the post-processing for the character input before the candidate selection instruction is temporarily stopped (stopped) at that time.

  As another example, one character by character cutting time (processing (timer event) that occurs when a pen-up time (time when no operation by the pointing tool 3 is not performed: non-input time) elapses) has passed. When the character data segmenting operation is performed, the post-processing continuation determining unit 33 refers to the post-processing continuation determination table to determine the character input at the time when the operation is performed and subsequent characters. It is determined that the post-processing for the input character is continued.

  Thus, in the present invention, every time an input operation or event occurs with the pointing tool 3 or the like, the continuation / cancellation of post-processing is determined based on the input operation type and event type, thereby making it easier for the user to use the operating environment. Can be provided.

  This post-processing continuation determination table is merely an example, and various configurations can be realized depending on the application and purpose. That is, it goes without saying that a post-processing continuation determination table for various setting contents can be configured according to the processing type or event for which post-processing is desired to be applied to the character string obtained by the recognition processing according to the user's intention and purpose. .

  Next, a specific example of character recognition processing according to the first embodiment will be described with reference to FIG.

  FIG. 5 is a diagram for explaining a specific example of character recognition processing according to the first embodiment of the present invention.

  FIG. 5 shows an example of a display screen when the handwritten character input utility program that realizes the character recognition processing of the present invention operates on an e-mail application that operates on the portable information device 1.

  In the first embodiment, an example in which a handwritten character input utility program is operated on an application screen generated by an e-mail application will be described as an example. However, the present invention is not limited to this. That is, it goes without saying that the present invention is also effective for other application programs that require character input, for example, other programs such as a file browser, a photo viewer, a scheduler, and a word processor program.

  In FIG. 5, an area 51 is an area composed of various input fields for inputting an e-mail destination and subject. For example, by pressing an arbitrary position in the area 51 with the pointing tool 3, a cursor for text input is displayed in the corresponding input field. In this state, the user can execute the character input in the input field using the handwritten character input utility program.

  The area 52 is an area for inputting the text of the e-mail. For example, the cursor for text input is displayed in the area 52 by pressing an arbitrary position in the area 52 with the pointing tool 3. In this state, the user can execute character input in the area using the handwritten character input utility program.

  The screen 53 is a handwritten character input screen realized by a handwritten character input utility program. The details will be described with reference to FIG. 6. When character input is necessary on an arbitrary application program, the user executes character input by handwriting input by starting this handwritten character input utility program. be able to.

  Of course, it is also possible to assign a handwritten character input utility program in advance to an application program that requires the handwritten character input utility program and start the handwritten character input utility program simultaneously with the activation of the application program.

  When the user inputs handwritten characters (trajectory data input) on the handwritten character input screen 53 by the handwritten character input utility program, the locus data is recognized by the handwritten character recognition unit 35 of FIG. The character font corresponding to the character code obtained as the recognition result is once displayed on the handwritten character input screen 53.

  If the character on the handwritten character input screen 53 is a desired character, the user confirms the character so that the currently displayed character is the cursor position displayed on the e-mail application program. Inserted into.

  The area 54 is an area in which a control group for switching various application programs and a control (button or icon) for instructing activation of a handwritten character input utility program are arranged.

  In the area 54, a button 55 is a start button for a handwritten character input utility program. When the button 55 is pressed with the pointing tool 3, the handwritten character input utility program is activated and the handwritten character input screen 53 is displayed. On the other hand, when the button 55 is pressed while the handwritten character input utility program is activated, the handwritten character input utility program is terminated and the handwritten character input screen 53 is erased.

  The button 56 is a file browser activation button. When the button 56 is pressed with the pointing tool 3, a file browser application for browsing various data managed by the portable information device 1 is activated.

  The button 57 is an activation button for an electronic mail application. When this button 56 is pressed with the pointing tool 3, an e-mail application is activated.

  The button 58 is an activation button for the photo viewer. When this button 56 is pressed with the pointing tool 3, a photo viewer (photo display application) for browsing image data managed by the portable information device 1 is activated.

  The button 59 is a scheduler activation button. When this button 59 is pressed with the pointing tool 3, the scheduler application is activated.

  Next, details of the handwritten character input screen 53 generated by the handwritten character input utility program will be described with reference to FIG.

  FIG. 6 is a diagram showing an example of a handwritten character input screen generated by the handwritten character input utility program according to the first embodiment of the present invention.

  In FIG. 6, an area 61 is an input character buffer area, and displays a first candidate for a handwritten character recognition result. Here, when the input of one handwritten character is confirmed and the recognition process is completed, the first candidates of the recognition result are sequentially displayed. For example, if "Today" is entered by handwriting input, if the trajectory data is the same character as the shape registered in the recognition dictionary (standard character dictionary data), the recognition result is "Today" It is displayed in area 61.

  Here, when characters that are not intended candidates are included in the character group displayed in the input character buffer area 61, the user selects the characters that are not intended candidates by pressing the pointing tool 3. By doing so, the candidate can be changed.

  Specifically, when the user presses and selects an arbitrary character in the input character buffer area 61, the first to fourth candidate groups for the character are displayed in the candidate display area 66 in the order of the candidates. The If there is an intended candidate in the candidate group, the user presses the candidate with the pointing tool 3 to select the candidate, thereby selecting the character selected in the input character buffer area 61 in the candidate display area 66. The selected character can be replaced as the first candidate character.

  Note that the first candidate character group displayed in the input character buffer area 61 recognizes the recognition result of the handwritten character recognition unit 35 when the function of the post-processing unit 34 is explicitly disabled by user designation. The candidate ranking is reflected as it is. On the other hand, when the function of the post-processing unit 34 is explicitly enabled by user designation, based on the recognition result of the recognition result of the handwritten character recognition unit 35, the chain probability of the character string and the recognition result Based on the degree of coincidence, the order of the recognition candidates is changed, and the display of characters in the input character buffer area 61 is controlled.

  The button 62 is a conversion button. When this button 62 is pressed with the pointing tool 3, kana-kanji conversion of hiragana characters in the input character buffer area 61 is performed. For example, when this button 62 is pressed while “Aiga” is input in the input character buffer area 61, for example, “Love is” is converted. If this conversion candidate is not intended, the next candidate is displayed by pressing the button 62 again.

  The button 63 is a BackSpace key button (delete button). When the button 63 is pressed with the pointing tool 3, if a character is displayed in the input character buffer area 61, for example, 1 before the cursor displayed in the input character buffer area 61 is positioned. The character is deleted. On the other hand, if no character is entered in the input character buffer area 61, the character on the left side of the cursor displayed on the application (for example, the area 52 on the e-mail application) is deleted.

  The button 64 is a Space key button (blank character insertion button). When the button 64 is pressed with the pointing tool 3, when a character string is displayed in the input character buffer area 61 and there is an insertion position after the character string, a blank character ( (Blank code) is inserted.

  The button 65 is an Enter key button (candidate character string confirmation / line feed input button). When the button 65 is pressed with the pointing tool 3, if a character string is displayed in the input character buffer area 61, the character string is confirmed and the right side of the cursor displayed on the application is displayed. The character string is inserted. On the other hand, when there is no character in the input character buffer area 61, a line feed code is inserted on the right side of the cursor displayed on the application, and the cursor is broken.

  An area 66 is a candidate display area for handwritten character recognition. Here, as described above, the first to fourth recognition candidates corresponding to the selected character in the character string in the input character buffer area 61 are displayed.

  As described above, when an arbitrary character is pressed and selected from the characters displayed in the candidate display area 66 with the pointing tool 3, replacement with the character selected in the input character buffer area 61 is performed.

  For example, when the character string of the recognition result “good today” is displayed in the input character buffer area 61, if “RI” is pressed and selected with the pointing tool 3, it is displayed in the candidate display area 66. “Li”, “I”, “I”, “” ”, etc. are displayed in that order as the first to fourth candidates for“ RI ”. The user can select any character from them. For example, if “I” is selected by pressing with the pointing tool 3, “Ri” in the input character buffer area 61 is replaced with “I”, and the entire character string is changed from “Good for today” to “Today for today”. It will be changed to “good”.

  An area 67 is a handwritten character input area. Here, character input can be performed by alternately writing handwritten characters in the two regions 67a and 67b.

  For example, if “I” is entered by handwriting in the area 67a, and then the first stroke (stroke) constituting “heaven” is written by handwriting in the area 67b, the input of “I” in the area 67a is confirmed. The recognition process for “I” is started. Here, when the recognition of “I” ends, the first candidate of the recognition result of “I” is displayed in the input character buffer area 61. At that time, the stroke of the handwritten character input to the area 67a is Erased.

  Next, processing for the area 67b is executed. This is because when there is a character to be written next when the handwritten strokes constituting “heaven” have been input, when the next character is written in the area 67a, the recognition process of “heaven” is started. .

  On the other hand, when there is no character to be input next, when the operation (for example, pressing) time from the pointing tool 3 has elapsed for a certain time (pen-up time), it is determined that the input to “heaven” is completed, Perform recognition processing. Alternatively, when a portion other than the handwritten character input area 67 is touched with the pointing tool 3, it may be determined that input to “heaven” has been completed at that time.

  Then, when the recognition process of “heaven” is completed, the first candidate of the recognition result of “heaven” is displayed on the right side of the character of the recognition result of “yes” in the input character buffer area 61, and “ The handwritten strokes that make up “Ten” are erased.

  It should be noted that these handwritten character input processes themselves can be realized in the same manner as the conventional handwritten character input process.

  Further, various operations (instructions) controls (buttons, menus, etc.) for instructing execution of various processes shown in FIGS. 5 and 6 are examples implemented on the GUI. It may be realized by a physical button or key on the device.

  Next, processing executed in the first embodiment will be described with reference to FIGS.

  FIG. 7 is a flowchart showing the character string post-processing continuation determination process according to the first embodiment of the present invention.

  This process is mainly executed by the post-processing continuation determination unit 33. In executing this process, the work area on the RAM 7 used by this process is initialized.

  First, the characters input by the user using the pointing tool 3 and editing operations are input as position coordinate data by the coordinate input unit 22. Then, depending on the position indicated by the position coordinate data, various operation contents such as input of handwritten characters, editing operation, candidate selection operation, and other operations are determined. This determination is realized by the processing type determination unit 31 in FIG. 3 as described above.

  In the first embodiment, as an example of determining the input type by the processing type determination unit 31, a case where the input type is realized by the handwritten character input utility program shown in FIG. 6 will be described as an example. The handwritten character input utility program is started by pressing the button 55 in the handwritten character input screen of FIG. 5 as described above.

  The flowchart will be specifically described below.

  In step S <b> 702, it is determined whether or not it is a candidate change process for changing a character string candidate in the input character buffer area 61. In this determination, for example, it is determined whether or not the pointing tool 3 is pressed on the character string in the input character buffer area 61 of FIG. In particular, when position coordinates in the area are input, it is determined that the candidate change process is being performed. If it is candidate change processing (YES in step S702), the process proceeds to step S703. On the other hand, if it is not candidate change processing (NO in step S702), the process proceeds to step S704.

  In step S703, a character string post-processing completion flag is set. This character string post-processing completion flag is a flag indicating whether or not character string post-processing is completed.

  For example, when the user performs the candidate change process, the character string of the first candidate displayed in the input character buffer area 61 is different from the intended character string at that time. When the candidate is changed, the candidate display character string after the change becomes a character string desired by the user. Therefore, at this point, the character string post-processing completion flag is set in, for example, the RAM 7 in order to end (stop) the post-processing of the character string input so far. When this character string post-processing completion flag is set, the post-processing unit 34 in FIG. 3 is instructed to complete post-processing.

  In step S704, the presence / absence of handwritten character input is determined. This determination is based on the presence or absence of input of position coordinates in the handwritten character input area 67 of FIG. If there is a handwritten character input (YES in step S704), the process proceeds to step S705. On the other hand, if there is no handwritten character input (NO in step S704), the process proceeds to step S706.

  In step S705, processing for adding recognition candidates to the post-process character string is performed. After one character of the input handwritten character trajectory is determined in the handwritten character input area 67 of FIG. 6, the position coordinate data of the handwritten character trajectory input to the handwritten character recognition unit 35 of FIG. Next, character feature data is extracted from the position coordinate data using a conventional online handwritten character recognition algorithm.

  Next, handwritten characters are input in the order of standard character dictionary data having a high degree of matching by the matching process between the feature data and the standard character dictionary data (standard character feature data) in the recognition dictionary. Candidates up to the fourth candidate are determined as first candidates other than characters. Then, the character codes corresponding to the determined first to fourth candidates and the evaluation value of the matching degree are passed to the post-processing unit 34.

  The post-processing unit 34 determines a combination having a high chain probability between the character strings from the character candidate string including the character code and the evaluation value of the recognition candidate, and uses the high combination as a first candidate as an input character buffer area. The display in 61 is changed.

  For example, the case where the display in the input character buffer area 61 is initially “12-4” by the handwritten character recognition unit 35 will be described.

  Here, paying attention to “RO”, for “RO”, the first candidate is “RO”, the second candidate is “3”, and the third candidate is “END”. Assume that the fourth candidate is “ru”. In this case, combinations of character strings include “12 4”, “1234”, “12 end 4”, and “12 ru 4”.

  In this case, in the chain probability, the combination of “1234” becomes a high evaluation value, and the combination of “1234” is selected if there is little difference between the evaluation values of the first candidate “RO” and the second candidate “3”. It is. As a result, the display in the input character buffer area 61 is switched from “12-4” to “1234”.

  Next, in step S706, it is determined whether or not the character cut time is up.

  In the handwritten character input area 67 of FIG. 6, when a predetermined time (predetermined time) has passed after the last trajectory of the trajectory constituting one character is input, Generate a cut-up time up event. This configuration is also a general process employed in conventional online handwritten character recognition.

  Therefore, when the character cut time elapses, this character cut time time-up event occurs, that is, when the character cut time time-up occurs (YES in step S706), the process proceeds to step S707.

  On the other hand, if it is another event or other position coordinate input (NO in step S706), the process proceeds to step S708.

  In step S707, the post-processing unit 34, which has added recognition candidates to the post-processing character string, performs a process of giving a continuation instruction. Note that in the conventional character recognition processing that performs post-processing, the post-processing is terminated when it is determined that the continuous input has been completed (for example, when the character cutting time is up).

  On the other hand, in the first embodiment, when the input of the last locus of the locus group of the handwritten characters intended by the user is confirmed by the character cut time time-up event, the locus data in the handwritten character input area 67 is temporarily changed. It passes to the handwritten character recognition part 35. Then, the candidate character output from the handwritten character recognition unit 35 and its evaluation value are passed to the post-processing unit 34, and a post-processing continuation instruction is issued to the post-processing unit 34 to wait for input.

  As a result, in the present invention, even if the character cut time is increased, post-processing can be continued for subsequent input.

  In step S708, it is determined whether or not the recognition start area has been pressed. In the first embodiment, the recognition start area defines an area outside the handwritten character input area 67 in FIG. 6. When a position coordinate for this area is input, a recognition start event is first generated.

  Specifically, when unprocessed trajectory data exists in the handwritten character input area 67 when coordinates outside the handwritten character input area 67 in FIG. 6 are input, a recognition start event is first generated. If this recognition start event has occurred (YES in step S708), the process proceeds to step S709, where the position coordinate input data is held as it is, and the process at that position is executed as the next event.

  For example, when the button 65 is pressed with the pointing tool 3 in the state where the locus of the handwritten character of “Ki” is input in the handwritten character input area 67, first, an area outside the handwritten character input area 67 (recognition). A recognition start event is generated by pressing the (start area), and handwritten character recognition is performed. Next, after the recognition result is displayed in the input character buffer area 61, processing for pressing the button 65 is executed. On the other hand, in the case of other events (NO in step S708), the process proceeds to step S710.

  In step S710, it is determined whether a one-character erasing process has been instructed. When the position coordinates on the button 63 are input (YES in step S710), it is determined that one character is erased, and the process proceeds to step S711. On the other hand, if it is other position coordinate input (NO in step S710), the process proceeds to step S712.

  In step S711, a one-character deletion process is executed. When there is a character in the input character buffer area 61, the rightmost character in the input character buffer area 61 is deleted. Next, processing for deleting a recognition candidate in the post-processing character string in the post-processing unit 34 corresponding to the erased character is executed. On the other hand, if there is no character in the input character buffer area 61, the character on the left side of the cursor position on the application is erased, and there is no processing target data in the post-processing unit 34. finish.

  In step S712, it is determined whether or not there is application switching processing. When the position coordinates on the buttons 56, 59, etc. in FIG. 5 serving as an application switching instruction are input (YES in step S712), it is determined that the application switching operation is performed, and the process proceeds to step S713. On the other hand, if it is other position coordinate input (NO in step S712), the process proceeds to step S714.

  In step S713, the post-processing character string is continued. In the first embodiment, the input of the handwritten character input utility program is continued even when the application is switched. This assumes that a request for changing an application may occur during input. Therefore, a post-processing continuation instruction is issued to the post-processing unit 34 to wait for input.

  In step S714, it is determined whether or not there is an instruction for kana-kanji conversion processing.

  If the position coordinates of the button 62 in FIG. 6 are input (YES in step S714), it is determined that the process is a kana-kanji conversion process, and the process proceeds to step S715. If it is other position coordinate input (NO in step S714), the process proceeds to step S716.

  In step S715, kana-kanji conversion processing is executed. Since the kana-kanji conversion process itself may be various kana-kanji conversion processes conventionally performed, detailed description thereof will be omitted. Briefly, when a character is input in the input character buffer area 61, for example, when a hiragana character of “Tenki” is input, kana-kanji conversion of the character string is performed, “weather”, etc. Convert to When this kana-kanji conversion process is executed, a process for converting an input character string into a character string having a different character code is performed. For this reason, here, since it does not make sense to continue the post-processing, a character string post-processing completion flag for the post-processing unit 34 is set.

  In step S716, it is determined whether a character string insertion is instructed. If a character string insertion command or the like is generated by a user instruction (YES in step S7164), the process advances to step S717. In step S717, processing similar to that in step S715 is executed. On the other hand, if it is any other position coordinate input or other event (NO in step S716), the process ends.

  With the processing described so far, it is possible to control continuation or cancellation of post-processing that is processing in units of character strings of handwritten characters in accordance with input processing and editing processing of each handwritten character input.

  7 is not limited to the order shown in FIG. 7 (step S702, step S704, step S706, step S708, step S710, step S712, step S714, step S716). The order can be changed according to the use and purpose.

  Next, the handwritten character recognition process will be described with reference to FIG.

  FIG. 8 is a flowchart showing handwritten character recognition processing according to the first embodiment of the present invention.

  When the handwritten character recognition process is started, the work area on the RAM 7 used by this process is first initialized. This process is mainly executed by the handwritten character recognition unit 35. The position coordinate data of the trajectory for one character to be recognized is input as trajectory data for one character by the areas 67a and 67b on the handwritten character input area 67 in FIG. The

  In step S902, trajectory data for one character is read into the work area. For example, the trajectory data for one character is composed of trajectory data for one stroke if it is a character of one stroke. Moreover, if it is a character of 2 strokes or more, it is composed of a plurality of stroke data of one stroke. The trajectory data of one stroke starts with the XY position coordinate data of the start point, for example, and the XY position coordinate data along the shape of the trajectory continues to the XY position coordinate of the end point. Then, the position coordinate data is normalized by a conventional algorithm, and feature points (feature data) are extracted.

  In step S903, recognition processing is executed on the obtained trajectory data, and candidate characters (character codes) as the recognition results are stored in the RAM 7. Here, matching processing is performed between the feature data of the trajectory data extracted in step S902 and the standard character dictionary data of the recognition dictionary. Then, the character code corresponding to the standard character dictionary data having the highest degree of coincidence and the score (evaluation value) of the degree of coincidence are stored as recognition results, and the character codes corresponding to the standard character dictionary data are recognized in order from the highest degree of coincidence. Store as a result candidate.

  In step S904, recognition result candidates are registered in the post-processing unit 34. That is, the character code of the recognition result stored in step S903 and its evaluation value are registered in the post-processing unit 34. The post-processing unit 34 stores character codes that are recognition results that are sequentially recognized and their evaluation values.

  Here, for example, a case where a handwritten character “factory” is input will be described. First, by the handwritten character recognition unit 35, the first candidate character of the recognition result for “Ku” is “Ku” and its evaluation value is 100, the second candidate character is “D” and its evaluation value is 100. It is assumed that the third candidate character is “e” and its evaluation value is 90.

  Next, it is assumed that the first candidate character of the recognition result for “place” is “lift” and its evaluation value is 98, the second candidate character is “place”, and the coincidence evaluation value is 96.

  In this case, the post-processing unit 34 calculates the chain probability of the candidate combination character string using the post-processing dictionary. The combination is “Kanage”, “Factory”, “E”, “E”, “Y”, “Y” etc. Then, the post-processing unit 34 determines a candidate character string having the highest linkage probability among these combinations.

  In step S905, the candidate character string is read from the post-processing unit 34.

  As described above, for example, even when the recognition result in the handwritten character recognition processing unit 35 when “factory” is input as the handwritten character is “Kyogo”, the processing by the post-processing unit 34 in step S904 is performed. When the candidate character string “factory” having a high number chain probability is determined, “factory” is read as a candidate character string of handwritten characters. In this case, the recognition result in the handwritten character recognition processing unit 35 can be replaced with “factory” as “Koujo”.

  In step S906, the recognition result is displayed. For example, in the input character buffer area 61 of FIG. 6, “factory” obtained as an initial recognition result is replaced with “factory” in the process of step S905 and displayed.

  This series of processes will be described in time series. When “Kaku” is input first and the first stroke of the next character “K” is input, at that time, the handwritten character corresponding to the input “K” is input. The “result” of the recognition result by the recognition unit 35 is displayed in the input character buffer area 61.

  Next, when the input of the character of “place” is confirmed by, for example, the character cutting time, recognition by the handwritten character recognition unit 35 for “place” is started, and if the first place of the recognition result is “lift”, In the input character buffer area 61, “Draft” is displayed as the recognition result.

  Next, the post-processing unit 34 determines a candidate character string having the highest chain probability among the combinations in the recognition candidates by the handwritten character recognition unit 35. If the candidate character string is “factory”, the input character buffer area The display of 61 is replaced with “factory” from “working”.

  Furthermore, if the user instruction to be input next is a handwritten character, for example, “GA”, the input character string is “factory is” post-processing target and post-processing is executed.

  On the other hand, if the post-processing result “factory” is not the intended recognition result but an instruction such as candidate change processing to another candidate or single character deletion processing is given, the character string up to that point is determined And the continuation of the post-processing is stopped (that is, the character string post-processing completion flag is set once at this time). Thereafter, the input handwritten character is handled as the first character to be post-processed (in this case, the character string post-processing completion flag is reset).

  Thereafter, the handwritten character recognition process is terminated. Here, the stored trajectory data for one character is deleted, and the work area or the like is released.

  As described above, according to the first embodiment, the post-processing continuation determination unit that determines whether or not to continue the post-processing for the input character string is provided according to the processing type defined by the user's instruction input. This makes it possible for the user to input and edit characters naturally without having to worry about continuation or confirmation of the input character string (ie, without performing an operation to specify a post-processing unit delimiter). An environment for recognizing handwritten characters can be realized. Thereby, the handwritten character recognition environment where the input of the character which a user intends is easy is realizable.

<Embodiment 2>
In the first embodiment, the character string to be post-processed is described as an example of handling the recognition result of handwritten characters. However, in a normal portable information device, text characters copied from other applications in addition to handwritten character input There are many cases where a column is inserted and used. In such a case, the character of the recognition result of the handwritten character and the inserted text character string are mixed in one sentence. In this case, the conventional handwritten character recognition apparatus cannot realize post-processing in such a situation.

  Therefore, in the second embodiment, in addition to the configuration of the first embodiment, an insertion unit 101 (FIG. 9) for inserting a text insertion character string into the post-processing unit 34 is added. As a result, it is possible to efficiently realize post-processing in a state where a character as a result of handwritten character recognition and an inserted text character string are mixed in one sentence.

  The inserted text character string includes, for example, a character string copied on an arbitrary application, a character string input in the past, and a character string obtained by converting a handwritten recognized hiragana to kana / kanji.

  First, the functional configuration of the portable information device according to the second embodiment will be described with reference to FIG.

  FIG. 9 is a diagram showing a functional configuration of the portable information device according to the second embodiment of the present invention.

  In addition, about the structure same as FIG. 3 of Embodiment 1, the same reference number is added and the detailed description is abbreviate | omitted.

  In the second embodiment, as described above, the insertion unit 101 for inserting the text insertion character string into the post-processing unit 34 is provided. The insertion unit 101 is a processing unit that is called when another application program or a text character string saved in the past is inserted into the input character buffer area 61 in the handwritten character input utility program of FIG.

  The insertion unit 101 executes a process for handling a candidate character string that is a recognition result of a handwritten character input to the input character buffer area 61 and the inserted text character string as a target character string for post-processing. Then, the processed candidate character string is transmitted to the post-processing unit 34.

  Next, a specific example of character recognition processing according to the second embodiment will be described with reference to FIGS.

  10 to 12 are diagrams for explaining a specific example of character recognition processing according to the second embodiment of the present invention.

  10 to 12 show examples of display screens when a handwritten character input utility program operates on a word processor application.

  In FIG. 10, reference numeral 111 denotes a text input area of the word processor application. Reference numeral 112 denotes a display that conceptually represents a text character string to be inserted into the post-processing unit 34. The state where the text “city” is stored in a text cut buffer area (clipboard area: storage area secured by the OS in the internal memory), which is a function of the OS (operating system) on which the word processor application operates. ing.

  FIG. 11 shows a state where the text character string “city” is inserted into the input character string buffer area 61 from the state of FIG. 10. FIG. 12 shows a state in which a handwritten “thought” is to be input after the character string “store lost city” in the input character string buffer area 61 of FIG.

  Next, an insertion process for inserting (adding) the inserted character string into the post-processing unit 34 by the insertion unit 101 will be described with reference to FIG.

  FIG. 13 is a flowchart showing insertion processing according to the second embodiment of the present invention.

  In addition, the state transition of FIGS. 10-12 is also demonstrated along this flowchart.

  In another application or the like, a character string is selected by a conventional operation using a menu, an operation key, or the like, and copied to a specific storage area (system character string storage area).

  Next, the character string stored in the specific storage area (system character string storage area) is inserted into the place where the text cursor is displayed by an instruction such as a command or a menu.

  Since the process so far is a general process in a general computer or portable information device, a detailed description thereof will not be given.

  The cursor is moved after “lost” in the character string “lost memory” in the input character string buffer area 61 in FIG. 10, and the “city” stored in the system character string storage area there. Is inserted.

  When the insertion is instructed, the character string “store lost city” in the input character string buffer area 61 of FIG. 11 is displayed on the screen. The text insertion process itself is the same process as that of a conventional portable information device.

  This process is mainly executed by the insertion unit 101. In executing this process, the work area on the RAM 7 used by this process is initialized.

  Next, in step S142, the stored text character string to be inserted (inserted text character string) is read into the system character string storage area. For example, the character code of the insertion text character string “city” shown in FIG. 10 is stored in the RAM 7. Further, the position information of the position of the cursor when the insertion is instructed is stored.

  In step S143, in order to insert the inserted text character string into the post-processing unit 34 as a confirmed character string, the processing for processing the character string data into the same structure (format) as the character data processed by the post-processing unit 34 Execute. The character string data (post-processing character string data) processed by the post-processing unit 34 is composed of a character code and its evaluation value. Therefore, since the inserted text character string is originally defined by the character code, in the processing process, the character string data for post-processing is created by setting the evaluation value to 100 which is the highest score for the character code. To do.

  For example, in the case of “city”, post-processing character string data is created (character code of “city” and its evaluation value 100, character code of “city” and its evaluation value 100), and the position of the cursor Are output to the post-processing unit 34.

  Thereafter, a process for releasing the work area used by the process is executed, and the process is terminated.

  The post-processing unit 34 determines the insertion position of the post-processing character string data corresponding to the inserted text character string, based on the position information indicating the insertion cursor position. Then, based on the determination result, the post-processing character string data is inserted into the data position corresponding to the cursor position in the post-processing character string storage area of the post-processing unit 34.

  For example, in the example of FIG. 10 and FIG. 11, the cursor exists at a position to the right of “ta” in the character string “memory of lost”. Therefore, the cursor position is to the right of the fourth character, and the data at the insertion cursor position is data of 4 indicating the fourth character.

  If the character string “memory of lost” in FIG. 10 is a recognition result in handwritten character input, the data of “lost” of the first character of post-processing is the first candidate “lost” and its evaluation value 98. The second candidate “arrow” and its evaluation value 90 are included. Thereafter, in the same manner, for example, “ta” of the fourth character is the first candidate “ta” and its evaluation value 97, the second candidate “E” and its evaluation value 60, and the third candidate “thick” and its evaluation. The value is 59. The fifth character “NO” is the first candidate “NO” and its evaluation value 97, the second candidate “0” and its evaluation value 80, and the third candidate “O” and its evaluation value 78.

  In this state, the character code “Me” and its evaluation value 100, and the character code “City” and its evaluation value 100 are inserted into the post-processing character string data. Here, since the insertion position is after the fourth character data, it is inserted after the third candidate “thick” which is the last data of “ta” of the fourth character, and the fifth character is “Miyako”. The data is the first candidate “Miyako” and its evaluation value 100.

  Subsequently, the sixth character is “city”, and the data is the first candidate “city” and its evaluation value 100. The seventh character is “NO”, and the data is the first candidate “NO” and its evaluation value 97, the second candidate “0” and its evaluation value 80, and the third candidate “O” and its evaluation value 78. The following data will be sequentially shifted backward.

  As described above, the content of the post-processing character string data processed by the post-processing unit 34 is updated. That is, here, post-processing character string data is generated in a state where a character as a result of handwritten character recognition and an inserted text character string are mixed in one sentence.

  Next, an example in which handwritten characters are input after the processed character string data is updated will be described with reference to FIG.

  In FIG. 12, when a handwritten character “thought” is input in the regions 67 a and 67 b of the handwritten character input 67, first, a handwritten character recognition process is executed as in the process described in the first embodiment. In this case, for example, the first candidate “think” of the recognition result for “think” and its evaluation value 98, the second candidate “megumi” and its evaluation value 67, and the first candidate “i” for the recognition result “yes”. And the evaluation value 95, the second candidate “i”, and the evaluation value 90 are obtained.

  When this handwritten “thought” is input, the post-processing character string data in the post-processing unit 34 is “memory of lost city”, and the cursor position where the recognition result of the handwritten character is inserted is “ It exists behind "".

  In this case, the first candidate “O” (evaluation value 96) and the second candidate “A” (evaluation value 77) of the tenth character of the post-processing character string data are recognized for “think”. The first candidate “thought” (evaluation value 98), the second candidate “Megumi” (evaluation value 67), the first candidate “i” (evaluation value 95) of the recognition result of “i”, and the second candidate The data of “i” (evaluation value 90) is inserted.

  Based on the post-processing character string data processed in this way ("Lost memory of a lost city"), the post-processing described in the first embodiment is performed, the recognition candidate display candidates are changed, and input Displayed in the character buffer area 61.

  The post-processing with the post-processing character string data will be briefly explained. Among the post-processing character string data “I want to remember the lost city”, “Lost”, “Memory”, “Thought” Is a recognition result for handwritten input, and there are character code groups each consisting of a plurality of candidates as processing targets, and “city” is input as inserted text character string data. Only character codes exist.

  That is, as the post-processing character string data, a combination of character string (character code) candidate strings as shown in FIG. 14 is obtained. Then, the post-processing unit 34 determines a candidate character string having the highest chain probability based on the post-processing dictionary for chain probability. As a method for calculating the chain probability, a conventionally known method may be used, and the description thereof is omitted here.

  As described above, according to the second embodiment, when a configuration for inserting a text character into a character recognition result editing (confirmation) region (input character buffer region) for handwritten character input is provided, A text character string generated by another application or a text character string stored in the system character string storage area can be inserted.

  Accordingly, an information processing apparatus equipped with a handwritten character recognition function capable of functioning a post-processing for a character string in which a character string including a recognition result for handwritten character input and a separately inserted text character string functions without any problem. Can be realized.

  More specifically, after inputting a character string by handwriting input, for example, inputting a kanji character string by hiragana-kanji conversion created by a word processing application, and further inputting these characters even when inputting a character string by handwriting input Post-processing (for example, chain probability calculation processing in the second embodiment) can be executed with the processed character string as one processing target. As a result, it is possible to efficiently post-process the character string intended by the user in an appropriate processing unit, and to provide a character recognition environment with a higher recognition rate.

  Although the embodiments have been described in detail above, the present invention can take an embodiment as, for example, a system, an apparatus, a method, a program, or a storage medium, and specifically includes a plurality of devices. The present invention may be applied to a system that is configured, or may be applied to an apparatus that includes a single device.

  In the present invention, a software program (in the embodiment, a program corresponding to the flowchart shown in the figure) that realizes the functions of the above-described embodiment is directly or remotely supplied to the system or apparatus, and the computer of the system or apparatus Is also achieved by reading and executing the supplied program code.

  Accordingly, since the functions of the present invention are implemented by computer, the program code installed in the computer also implements the present invention. In other words, the present invention includes a computer program itself for realizing the functional processing of the present invention.

  In that case, as long as it has the function of a program, it may be in the form of object code, a program executed by an interpreter, script data supplied to the OS, or the like.

  As a recording medium for supplying the program, for example, floppy (registered trademark) disk, hard disk, optical disk, magneto-optical disk, MO, CD-ROM, CD-R, CD-RW, magnetic tape, nonvolatile memory card ROM, DVD (DVD-ROM, DVD-R) and the like.

  As another program supply method, a client computer browser is used to connect to an Internet homepage, and the computer program of the present invention itself or a compressed file including an automatic installation function is downloaded from the homepage to a recording medium such as a hard disk. Can also be supplied. It can also be realized by dividing the program code constituting the program of the present invention into a plurality of files and downloading each file from a different homepage. That is, a WWW server that allows a plurality of users to download a program file for realizing the functional processing of the present invention on a computer is also included in the present invention.

  In addition, the program of the present invention is encrypted, stored in a storage medium such as a CD-ROM, distributed to users, and key information for decryption is downloaded from a homepage via the Internet to users who have cleared predetermined conditions. It is also possible to execute the encrypted program by using the key information and install the program on a computer.

  In addition to the functions of the above-described embodiments being realized by the computer executing the read program, the OS running on the computer based on an instruction of the program is a part of the actual processing. Alternatively, the functions of the above-described embodiment can be realized by performing all of them and performing the processing.

  Furthermore, after the program read from the recording medium is written in a memory provided in a function expansion board inserted into the computer or a function expansion unit connected to the computer, the function expansion board or The CPU or the like provided in the function expansion unit performs part or all of the actual processing, and the functions of the above-described embodiments are realized by the processing.

It is a figure which shows an example of the usage pattern of the character recognition apparatus of Embodiment 1 of this invention. It is a figure which shows the hardware constitutions of the information portable apparatus which functions as a character recognition apparatus of Embodiment 1 of this invention. It is a figure which shows the function structure of the portable information device of Embodiment 1 of this invention. It is a figure which shows the post-process continuation permission determination table of Embodiment 1 of this invention. It is a figure for demonstrating the specific example of the character recognition process of Embodiment 1 of this invention. It is a figure which shows an example of the screen for handwritten character input which the handwritten character input utility program of Embodiment 1 of this invention produces | generates. It is a flowchart which shows the character string post-process continuation determination process of Embodiment 1 of this invention. It is a flowchart which shows the handwritten character recognition process of Embodiment 1 of this invention. It is a figure which shows the function structure of the portable information device of Embodiment 2 of this invention. It is a figure for demonstrating the specific example of the character recognition process of Embodiment 2 of this invention. It is a figure for demonstrating the specific example of the character recognition process of Embodiment 2 of this invention. It is a figure for demonstrating the specific example of the character recognition process of Embodiment 2 of this invention. It is a flowchart which shows the insertion process of Embodiment 2 of this invention. It is a figure which shows the example of a combination of the character string of the post-processing object of Embodiment 2 of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Portable information device 2 Display part 3 Indicator 4 Camera part 4a CCD
5 Coordinate input part 6 CPU
7 RAM
8 ROM
9 Media I / F
10 Keyboard 11 External I / F
12 Network I / F
13 System bus

Claims (10)

An information processing apparatus that inputs handwritten character trajectory data, recognizes characters, and performs post-processing on the character recognition results,
Input means for inputting the locus data;
Character recognition means for recognizing characters corresponding to the trajectory data input by the input means;
Post-processing means for post-processing candidate character strings of character recognition results by the character recognition means;
Instruction means for instructing execution of various processes including input processing by the input means and processing for the character recognition result candidates;
Based on the processing type instructed to be executed by the instruction means, a new character input after the instruction by the instruction means is continued to the post-processing character string that has already been post-processed by the post-processing means. An information processing apparatus comprising: determination means for determining whether or not. The information processing apparatus according to claim 1, further comprising: an inserting unit that inserts a character whose character code corresponding to the newly input character is determined. When the processing type is an input process by the input unit, the determination unit determines a candidate character newly input as a character recognition result by the character recognition unit with respect to trajectory data input by the input unit. The information processing apparatus according to claim 1, wherein the post-processing means continues the post-processing character string that has already been subjected to post-processing. An operation screen including a candidate display area for displaying candidate characters as a result of character recognition by the input means, the instruction means, and the character recognition means; an application screen generated by an application operating on the information processing apparatus; The information processing apparatus according to claim 1, further comprising display means for displaying a switching operation screen for switching a plurality of types of applications operating on the information processing apparatus. The input means includes an input area for inputting the trajectory data,
The instruction means includes
A changing unit for changing candidate characters on the candidate display area;
A conversion unit for converting a candidate character on the candidate display area to a kana-kanji character;
A deletion unit for deleting the candidate character before the cursor displayed on the candidate display area is located;
A blank character insertion portion for inserting a blank character on the candidate display area;
5. The information processing apparatus according to claim 4, further comprising: a candidate character string determination unit that determines a candidate character string on the candidate display area and inserts the candidate character string on the application screen. . The determination means includes an input process for the input area, a timer event process that occurs when a non-input time for the input area has elapsed for a predetermined time, a deletion process by the deletion unit, and a switching by the switching operation screen. In the case of any of the processes, a candidate character newly input as a character recognition result by the character recognition unit for the trajectory data input to the input area is already subjected to post-processing by the post-processing unit. The information processing apparatus according to claim 5, wherein the post-processing character string is continued. An operating unit that controls the information processing apparatus further includes an insertion unit that inserts characters stored in a storage area secured in the internal memory;
In the case where the type of processing is insertion processing by the insertion means, the determination means determines a character newly input by the insertion processing as a post-processing character that has already been post-processed by the post-processing means. The information processing apparatus according to claim 1, wherein the information processing apparatus is continued in a row. The information processing apparatus according to claim 1, wherein the post-processing unit calculates a chain probability between candidate character strings of the character recognition result. A method for controlling an information processing apparatus that inputs handwritten character trajectory data, recognizes characters, and performs post-processing on the character recognition results.
An input step for inputting the trajectory data;
A character recognition step for recognizing characters corresponding to the trajectory data input in the input step;
A post-processing step of post-processing the candidate character string of the character recognition result by the character recognition step;
An instruction process for instructing execution of various processes including an input process by the input process and a process for the candidate character recognition result;
Based on the processing type instructed to be executed by the instruction process, new characters input after the instruction by the instruction process are continued to the post-processing character string that has already been post-processed in the post-processing process. And a determination step of determining whether or not the information processing apparatus is controlled. A program that realizes control of an information processing apparatus that inputs trajectory data of handwritten characters, recognizes characters, and performs post-processing on the character recognition results,
A program code of an input process for inputting the locus data;
A program code of a character recognition step for recognizing characters corresponding to the trajectory data input in the input step;
A program code of a post-processing step of post-processing a candidate character string of a character recognition result by the character recognition step;
A program code of an instruction process for instructing execution of various processes including an input process by the input process and a process for the candidate character recognition result;
Based on the processing type instructed to be executed by the instruction process, new characters input after the instruction by the instruction process are continued to the post-processing character string that has already been post-processed in the post-processing process. And a program code of a determination step for determining whether or not.

Images