JP2000222522A - Recognition and processing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a recognition and processing device which quickly recognizes an inputted holograph to automatically start an application program or an external device corresponding to a recognition result character string. SOLUTION: The recognition range of a holograph inputted by a holograph input means 102 is designated by a recognition range designation means 201, and the holograph in the designated range is subjected to character recognition by a character recognition means 202, and its contents are analyzed in a recognition result analysis means 206 through a postprocessing means 204 to determine an application to be started, and the application is started by an application automatic start means 208. If the analysis result is related to an external device, data are transferred to the external device by an external device automatic operation means 209.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a recognition processing device for recognizing handwriting information input with a pen as characters and analyzing the contents of a character string obtained as a result.

[0002]

2. Description of the Related Art Conventionally, a recognition processing device in this field performs a character recognition process for obtaining a recognition candidate character in which a candidate rank is added to an input handwriting, as in an online handwritten character recognition device described in Japanese Patent Application Laid-Open No. 5-189621. After going,
A recognition processing apparatus has been devised that includes a post-processing unit that changes the order of recognition target character candidates based on predetermined information and obtains a more natural character recognition result. Many post-processing means have been proposed, including a method using a word dictionary or phrase information, and a method based on chain information indicating the ease of chaining between character strings.

[0003]

However, the above-described conventional recognition processing apparatus merely displays a character string as a result of recognition and post-processing of the input handwriting, and converts the character string into another application program. In this case, the user had to pass a character string to the application program via the buffer area.

In the above-described conventional recognition processing apparatus, all input handwritings are recognized at one time, and when a large amount of handwriting is input, the time required to recognize all the handwritings is long.

Furthermore, the above conventional recognition processing apparatus does not consider using the result of recognizing input handwriting as a character in an external device.

An object of the present invention is to provide a recognition processing device which can quickly recognize a handwriting input by a user and can use the handwriting with another application program or an external device.

[0007]

In order to achieve the above object, a recognition processing apparatus according to the present invention has a display device and a handwriting input device, and a handwriting obtaining means for obtaining a handwriting input by a user; Recognition range specifying means for specifying a range to be recognized for handwriting obtained by the means, character recognition means for character recognition of handwriting within the range specified by the recognition range specifying means, and characters obtained by the character recognition means And a recognition result analyzing means for extracting a predetermined keyword from the column and analyzing the content.

Further, the recognition processing device according to the present invention is characterized in that the recognition processing device is provided with an automatic application starting means for automatically starting an application program according to the analysis result obtained by the recognition result analyzing means.

Further, the recognition processing device according to the present invention includes an external device automatic operation means for automatically transferring data to an external device according to the analysis result obtained by the recognition result analysis means and issuing a command. Features.

According to a first aspect of the present invention, there is provided a recognition processing device having a display device and a handwriting input device, wherein the handwriting obtaining device obtains a handwriting input by a user and the handwriting obtaining device. Recognition range specifying means for specifying a range to be recognized for the handwriting, character recognition means for character recognition of handwriting within the range specified by the recognition range specification means, and a predetermined character string obtained by the character recognition means. A recognition result analyzing means for extracting a keyword and analyzing the content is provided, and has an effect of promptly recognizing input handwriting data and analyzing the content of information included in a character string of the recognition result.

According to a second aspect of the present invention, in the first aspect of the present invention, the recognition result analyzing means includes a keyword dictionary in which the degree of association between the keyword and the application corresponding to the keyword is described. It is determined that there is provided application relevance degree calculating means for checking whether or not there is a keyword described in the keyword dictionary in the given character string and calculating the degree of relevance between the character string and the application. It has an effect of calculating how much the character recognition result of handwriting is related to various applications by using a keyword dictionary.

According to a third aspect of the present invention, in the second aspect of the present invention, the application association degree calculating means includes:
The character string candidate group and the recognition evaluation value of each character string are obtained from the character recognition means, and a predetermined operation is performed on the degree of association with the application described in the keyword dictionary and the recognition evaluation value, thereby obtaining the character string candidate group. The degree of association with the application is determined even if the character string obtained at the character recognition stage is different from the user's intention. Has an action.

According to a fourth aspect of the present invention, in the keyword dictionary according to the second aspect of the present invention, the user can change the content of the keyword and the degree of association with the application corresponding to the keyword. There is an effect that the user can register his / her favorite keywords or frequently used keywords.

According to a fifth aspect of the present invention, there is provided a post-processing means for automatically correcting an erroneously recognized portion by collating a character string obtained by the character recognition means in the first aspect of the invention with a word dictionary. Even if a part of a character string is erroneously recognized, if there is a candidate in the word dictionary that matches the entire character string as a word by comparing it with the word dictionary, it will It has the effect of automatically correcting the recognition.

According to a sixth aspect of the present invention, in the first aspect of the present invention, there is provided an automatic application starting means for automatically starting the application program according to the analysis result obtained by the recognition result analyzing means. Since the content of the recognition result character string is analyzed and the application program corresponding to the content is automatically started, there is no need for the user to transfer the character string to the application program via the buffer area. Have.

According to a seventh aspect of the present invention, in the automatic application starting means according to the sixth aspect, a start candidate application for presenting an application program to a user in ascending order of relevance obtained by the application relevance calculating means. When the application to be activated is different from the user's intention, the application to be activated can be confirmed by presenting the application in descending order of the degree of association. Has an action.

According to an eighth aspect of the present invention, the application program according to the sixth aspect is an address book application for registering a name, an address, and a telephone number, and is obtained by the recognition result analyzing means. When the analyzed result includes at least one of a name, an address, and a telephone number, the application automatic start unit automatically starts the address book application and converts the analysis result into a format that can be used as data by the address book application. It has the effect of registering.

According to a ninth aspect of the present invention, the application program according to the sixth aspect is a schedule management application for registering a date, a place, and a schedule, and the application program is obtained by the recognition result analyzing means. When the analysis result includes the date and time, the application automatic start unit automatically starts the schedule management application, converts the analysis result into a format that can be used as data by the schedule management application, and registers it.

According to a tenth aspect of the present invention, the application program according to the sixth aspect is a memo pad application for storing a memo, and the analysis result obtained by the recognition result analyzing means is a memo pad. In the case where a keyword indicating that the memo pad is included, the automatic application starting means automatically starts the memo pad application, converts the analysis result into a format that can be used as data by the memo pad application, and registers the result.

According to an eleventh aspect of the present invention, in the first aspect of the present invention, the analysis result is converted into an external device corresponding to the analysis result obtained by the recognition result analysis means so that the external device can use it. External device automatic start-up means for delivering the data and issuing a command, analyzing the content of the character string of the recognition result, and automatically delivering the analysis result to the external device in accordance with the content, and executing the command. It has the remote control action of issuing.

According to a twelfth aspect of the present invention, the external device according to the eleventh aspect is a video deck, and the analysis result obtained by the recognition result analysis means is a keyword relating to operation of the video deck. If you include
The external device automatic activation means has a function of converting the analysis result into a format usable by the video deck, transferring the result to the video deck, and issuing a command.

According to a thirteenth aspect of the present invention, the external device according to the eleventh aspect is a facsimile. When the analysis result obtained by the recognition result analysis means includes a keyword relating to a facsimile, There is an effect that the external device automatic activation means converts the analysis result into a format usable by the facsimile, delivers it to the facsimile, and issues a command.

[0023]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing a configuration of a recognition processing apparatus according to one embodiment of the present invention. 1 includes a central processing unit (hereinafter, referred to as a CPU) 101, a handwriting input device 102, a display device 103, and a main memory 10.
4. External device interface 105, file device 10
6. Read-only memory (hereinafter referred to as ROM) 107
Consists of The CPU 101 executes a program stored in the ROM 107 using the main memory 104.

FIG. 2 is a processing block diagram showing the flow of processing from handwriting input to application activation in the recognition processing apparatus according to one embodiment of the present invention. The handwriting input device 102 is a device that acquires handwriting input by a user using a pen as stroke data. Here, the stroke data is continuous information of coordinate values of the pen point on the handwriting input surface, which are detected during a period from when the pen comes into contact with the handwriting input surface until the pen leaves.

The user designates a part of the handwriting to be recognized in the handwriting input to the handwriting input device 102 by the recognition range designating means 201. The recognition range is specified by surrounding the handwriting with a line in the range specification mode. When the specification of the recognition range is completed, the recognition range specifying unit 201 extracts the stroke data within the recognition range from the entire stroke data, and delivers the extracted stroke data to the character recognition unit 202. Extraction of stroke data within the recognition range is described in IPSJ National Conventions, Vol. 44
TH, NO. 2, PP. 2.339-2.340 (19
92) A method such as the algorithm described in the above.

The character recognizing means 202 is provided by IEICE Transactions D-II, VOL. J73-D-II, NO.
7, PP. The stroke data extracted by the recognition range designating unit is divided into several partial lines by a method such as the algorithm described in 973-981. Next, character recognition is performed on each partial line using the recognition dictionary 203. First, stroke data constituting a partial line is cut out in units of characters by a method or the like in the character cutout device described in JP-A-6-124364. Next, the cut-out character unit stroke data is compared with the recognition dictionary 203 and converted into a character code. Dictionary for recognition 20
Reference numeral 3 describes the correspondence between segment strings and character codes obtained by further subdividing the stroke data according to the direction. Next, a segment sequence in the dictionary similar to the segment sequence obtained from the cut-out input stroke data is searched by a technique such as a dynamic programming method, and the candidate ranking is assigned in descending order of similarity. Then, a character code corresponding to the segment sequence in the dictionary having the highest similarity with the input segment sequence is adopted as a character recognition result. When all character unit stroke data are converted into character codes, the character code string obtained as a result of the conversion is passed to the post-processing means 204.

The post-processing means 204 automatically corrects an erroneously recognized portion by comparing the delivered character code string with the word dictionary 205. In the word dictionary 205, character strings in units of words and phrases are registered, and recognition is performed using a technique such as a language processing unit described in JP-A-6-124364 so that the continuation between characters becomes more appropriate. The candidate order is exchanged, and the character string of the correction result is delivered to the recognition result analysis means 206.

The recognition result analyzing means 206 includes the post-processing means 2
In order to determine whether the character string passed from the application 04 is relevant to the application, the keyword dictionary 20 is used.
7, the evaluation value for the character string is calculated by the application association degree calculation means. (Table 1) is an example of the structure of the keyword dictionary 207, and evaluation values are set for each item in accordance with the type of character or character string included in the character string. Here, the items are the type of application and the type of external device.

[0030]

[Table 1]

After calculating the evaluation values for all the character strings, the evaluation values for all the character strings are added for each item. As a result, it is determined that the application corresponding to the item having the highest evaluation value is started. However, when the highest evaluation value does not exceed the predetermined threshold, the memo pad application is activated. When the application to be started is determined, the character string is delivered to the application automatic starting unit 208.

The application automatic start unit 208 converts the transferred character string into a format that can be used as data by the application, starts the application, and transfers the data. The display device 103 presents the user's input handwriting and the screen display of the application activated by the application automatic activation means 208 to the user.
When the recognition result analysis unit 206 determines that the character string is related to an external device, the recognition processing device of the present invention transfers data to the external device through the external device automatic operation unit 209.

The operation of the recognition processing apparatus according to one embodiment of the present invention will be described below with reference to FIGS. FIG. 3 is a flowchart illustrating the operation of the recognition processing device according to an embodiment of the present invention. FIG. 4 is a diagram showing a handwriting input surface of the recognition processing device according to one embodiment of the present invention. In FIG. 4, the recognition processing device includes a handwriting input surface 401, a mode selection switch 402, and a stylus pen 403. FIG. 5 is a diagram showing an example of a screen display when a person's name, a telephone number, a date and time of a meeting, a place, and contents are input to the recognition processing device according to the embodiment of the present invention. FIG. 6 is a diagram illustrating an example of a screen display when a character recognition range is specified for a handwriting input to the recognition processing device according to an embodiment of the present invention. FIG. 7 is a flowchart illustrating a flow of keyword search performed by the recognition processing device according to an embodiment of the present invention. FIG. 8 is a diagram illustrating an example of a screen display when the address book application is activated in the recognition processing device according to the embodiment of the present invention. FIG. 9 is a diagram showing an example of a screen display when a recording reservation of a VCR and a facsimile transmission destination number are input to the recognition processing device according to the embodiment of the present invention, and a character recognition range is designated. FIG. 10 is a diagram showing an example of a screen display when recording reservation data is transferred to a video deck in the recognition processing device according to an embodiment of the present invention. FIG. 11 is a flowchart showing the operation of the recognition processing device in the case where the activation candidate application presenting means for presenting applications in order of application relevance is added to the recognition processing device in the recognition processing device according to one embodiment of the present invention. It is.

In FIG. 3, when the power is first turned on,
In step S301, the apparatus is initialized, and the process proceeds to step S302. In step S302, a process of accepting handwriting input by the stylus pen 403 in the handwriting input device 102 is performed. The handwriting input when the pen leaves the input surface of the handwriting input device 102 is registered as stroke data, and the process proceeds to step S303. If the user presses the range switch of the mode selection switch 403 to notify the end of handwriting input in step S303, the process proceeds to step S304. In step S303, if the end of handwriting input is not notified from the user and there is a subsequent handwriting input, the process returns to step S302 to perform handwriting input processing. FIG. 5 shows an example of handwriting input.

In step S304, the user presses the pen on the input surface of the handwriting input device 102, draws a line so as to surround the recognition range, and detects the coordinates of the pen until the pen is released from the input surface. Is displayed on the display device 103, stroke data within the designated recognition range is extracted from all the stroke data obtained before step S304, and the process proceeds to step S305. Once the pen touches the input surface again after the pen is released, it is regarded as a retry of the recognition range designation, the recognition range designation line that has already been input is erased, and the process returns to the recognition step S304 to designate the recognition range. Again. When the end of the recognition range designation is notified from the user by pressing the recognition switch of the mode selection switch 402 in step S305, step S305 is performed.
Move to 306. FIG. 6 shows a state in which the recognition range is designated by a thick line 601 so as to recognize the personal name and the telephone number from the handwriting in FIG.

In step S306, the handwriting specified to be recognized in FIG. 6 includes a partial line "Ichiro Matsushita", a partial line "tel", and a partial line "06-1234".
5678 ". Next, character recognition is performed on each partial line using the recognition dictionary 203.

In step S307, step S306
Is performed using the word dictionary 205 for all the character strings obtained as a result of the character recognition in, and the erroneously recognized portion is automatically corrected. In the example of FIG. 6, the three partial lines obtained in step S306 are “Ichiro Matsushita” as a result of character recognition and post-processing.
, A character string “tel”, and “06-
1253-5678 ".

In step S308, step S307
With respect to one character string from among the character strings obtained in (1), it is checked whether or not there is a keyword related to the application using the keyword dictionary 207 as shown in (Table 1). The keyword related to the application here is a word or a symbol (for example, a name of a prefecture, a city or a county, a postal code, etc.) or a telephone number representing an address in the case of an address book application. In the case of a schedule management application, words and symbols representing dates, words and symbols representing time, and the like are used.

The keyword search performed in step S308 will be described with reference to FIG. When the keyword search is started, one keyword is extracted from the keyword dictionary in step S701. Next, in step S702, it is checked whether the keyword extracted in step S701 exists in the character string. If the keyword exists in the character string, the process proceeds to step S703, the evaluation value for the keyword is added to the integration result of the evaluation value for each item obtained so far, and the process proceeds to step S704. If the keyword does not exist in the character string in step S702, the process proceeds to step S702.
The process moves to 704. In step S704, it is checked whether a keyword that is not yet a search key remains in the keyword dictionary.
Returning to 1, the next keyword is extracted. If the search has been completed for all the keywords, the keyword search ends.

In the example of FIG. 6, the evaluation value of the character string "Ichiro Matsushita" obtained in step S307 is calculated as follows. First, in (Table 1), the keyword “/” at the top is extracted, and it is checked whether or not this keyword is in the character string “Ichiro Matsushita”. Since the keyword “/” does not exist in this character string, the evaluation value of each item is 0. Next, the keyword ":" is extracted, and the evaluation value 0 is obtained in the same manner. This is performed for all keywords in the keyword dictionary. As a result, an evaluation value of 0 is obtained for all items for the character string "Ichiro Matsushita".

When the calculation of the evaluation value for one character string is completed, the flow shifts to step S309. In step S309, it is checked whether a keyword search has been performed for all character strings. If there is a character string for which a keyword search has not yet been performed, the process returns to step S308, and a keyword search is performed for the character string. If a keyword search has been performed for all character strings, step S3
Move to 10. In the example of FIG. 6, "tel" and "06-
Since the keyword search has not yet been performed on the character string “1234-5678”, the keyword search is started on these character strings. From (Table 1), for the character string "tel", "telephone, TEL, tel, Te"
The evaluation value is obtained only by the keyword “l”. As a result, the evaluation value of the character string “tel” is 7 in the “address book” item, 0 in the “schedule” item, 1 in the “memo pad” item, and “video deck”. 0 for the item and 2 for the "facsimile" item. In addition, the character string "06-1234-5678"
, An evaluation value is obtained only by the keyword “−”. The evaluation value is 3 in the “address book” item, 1 in the “schedule” item, 2 in the “memo pad” item, and in the “VCR” item. 1 and 3 in the “Facsimile” item.
Since there are two keywords "-" in this character string,
The evaluation value for this character string is 6,
The item “schedule” is 2, the item “memo pad” is 4, the item “video deck” is 2, and the item “facsimile” is 6.

In step S310, an application to be started is determined. The evaluation values for all the character strings obtained in step S308 are added for each item. As a result, it is determined that the application corresponding to the item having the highest evaluation value is started. However,
If the highest evaluation value does not exceed the predetermined threshold, all the character strings are regarded as having low association with the specific application, and it is determined that the memo pad application is to be started. Here, from the evaluation values described in (Table 1),
It is assumed that the threshold value for determining the start application is 10. Once the application to launch is determined,
It moves to step S311. In the example of FIG. 6, “Ichiro Matsushita”, “tel”, and “06-1234-5678”
Add the evaluation value for each character string for each item to determine the application to be started. The total of the evaluation values for the three character strings is 13 for the “address book” item, 2 for the “schedule” item, 5 for the “memo pad” item, 2 for the “video deck” item, and 8 for the “facsimile” item. . The item with the highest evaluation value is “address book”, and since the evaluation value exceeds the threshold, the application to be started is determined as the address book application.

In step S311, step S310
The character string is converted into a format that can be used as data by the application in order to transfer the character string to the application determined in step S312, and the process proceeds to step S312. Step S31
In step 2, the application program determined in step S310 is started, and the converted data is received and registered in step S311. In the example of FIG. 6, "Ichiro Matsushita"
Is not included in the character string, the name is determined to be a name. The character string "tel" is determined to be a character string indicating that the following character string is a telephone number, and is not directly treated as data. "06-1234
The character string "-5678" is composed of numbers and symbols, and the character string immediately before it is "te
l ", and the fact that the area code and the local area code are included in the character string are determined to be telephone numbers. These character strings are converted into a format usable by the address book application, and the address book application is activated. FIG. 8 shows an example of a screen display when the address book application is activated. Since these character strings are not related to the external device, the process returns to step S302 and waits for the next input.

The above operation will be described using different specific examples. FIG. 9 is a diagram showing an example of a screen display when a recording schedule of a video deck, a facsimile transmission destination number and the like are input to the recognition processing device according to the embodiment of the present invention, and a character recognition range is designated. From the handwriting within the range specified in FIG. 9, “7/24” is obtained by character recognition and post-processing.
Three character strings of "4ch" and "19:00 to 20:55" are obtained (step S307).

An evaluation value is obtained from the keyword dictionary 207 for each character string (step S308).
For the character string “7/24”, “Address book”
The evaluation value is 1 for the item, 4 for the “schedule” item, 2 for the “memo pad” item, 3 for the “video deck” item, and 0 for the “facsimile” item. For the character string “4ch”, the “address book” item is 0, the “schedule” item is 0, the “memo pad” item is 3, and the “video deck”
An evaluation value of 7 for the item and 0 for the “fax” item is obtained. For the character string “19:00 to 20:55”, the “address book” item is 0, the “schedule” item is 14, the “memo pad” item is 4, the “VCR” item is 12, and the “facsimile” Item is 0.

Next, the evaluation values for the three character strings are added for each item. As a result, "Address Book" item is 1,
"Schedule" item is 18, "Notepad" item is 9,
The “video deck” item is 22 and the “facsimile” item is 0. The item with the highest evaluation value is “VCR”, and its evaluation value exceeds the threshold value. Therefore, these character strings are determined to be character strings related to the VCR, and are used to transfer data to the VCR. It is determined to start the application (step S310).

Next, data for transfer to the VCR is created from the three character strings (step S311), and an application 1001 for transferring data to the VCR is activated (step S312). Application 100 for transferring data to VCR
FIG. 10 shows an example of a screen display when the application 1 is activated. Since the three character strings relate to the external device, that is, the video deck (step S313), the recognition processing device transfers data to the video deck through the external device automatic operation means 209 (step S314).

As shown in FIG. 11, "FAX: 06-1
When the range is specified so that the recognition process is performed on the handwriting “234-5678”, “FAX: 06-1” is obtained from this handwriting.
A character string "234-5678" is obtained, and an evaluation value is calculated for this character string by the recognition result analysis means 206. Since this character string includes the keywords “FAX”, “:” and “−”, the calculation results of the evaluation values are “address book” item “9”, “schedule” item “7”, and “memo pad”
The item is 5, the “VCR” item is 6, and the “Facsimile” item is 13. Thus, it is determined that the application for transferring data to the facsimile is started. Next, the character string "FAX: 06-1234-567
8 ”, the input handwriting data (FIG. 11) is converted into image data to be transmitted by facsimile, and the external device automatic operation means 2 is extracted.
In step 09, the image data is sent to the destination via facsimile. The handwriting to be converted into image data is only handwriting within the range surrounded by the recognition range designation line 601 or handwriting other than the range surrounded by the recognition range designation line 601.
Alternatively, there are three types of all handwriting regardless of the range surrounded by the recognition range designation line 601. Which handwriting is to be converted is left to the judgment of the user.

The application relevance calculating means obtains a character string candidate group and a recognition evaluation value of each character string from the character recognizing means, and determines a predetermined relation between the relevance and the recognition evaluation value of the application described in the keyword dictionary. By calculating the degree of association between the character string candidate group and the application by performing the above operation, even when character recognition different from the user's intention is performed in the character recognition stage, the degree of association with the application is determined by the user. It can be obtained in a way that meets your intentions. For example, let us consider a case in which a user has input a handwriting “memo”, but is recognized as “× 〒” by character recognition and post-processing. At this time, when the post-processing is performed, a plurality of recognition candidates are listed together with the recognition evaluation value, and the character string having the highest recognition evaluation value is adopted as the recognition result. In this example, it is assumed that a recognition candidate group having a recognition evaluation value of 8 for a recognition candidate “× 〒” and a recognition evaluation value of 7 for a recognition candidate “memo” is given for an input handwriting. Next, an application relevance evaluation value for each candidate character string is calculated by performing a predetermined operation on the recognition evaluation values of these candidate character string groups and the application relevance in the keyword dictionary. here,
It is assumed that the product of the recognition evaluation value of the candidate character string and the application relevance value in the keyword dictionary is taken. Then, since the keyword “〒” in the keyword dictionary exists in the candidate character string “× 〒”, the candidate character string “× 〒”
And the application relevance value of the keyword “〒”. As a result, an evaluation value of 56 is obtained for the “address book” item and 24 is obtained for the “memo pad” item. Next, "Note"
Since the candidate character string matches the keyword in the keyword dictionary, the product of the recognition evaluation value of the candidate character string “memo” and the application association degree value of the keyword “memo” is calculated. As a result, the evaluation value 70 in the “memo pad” item
Is obtained. As a result, the evaluation value indicating the association with the memo pad is obtained as the highest evaluation value, although the character recognition result different from the user's intention is obtained in response to the input of the handwriting “memo”. It is possible to follow the original intention of

Further, by allowing the user to change the degree of association between the contents of the keyword dictionary and the application corresponding to the keyword, the user can register his or her favorite keywords or frequently used keywords. This makes it possible to reduce the possibility that an application different from the user's intention is started.

Further, by adding the activation candidate application presenting means to the recognition device, it becomes possible to select activation of another application when the application to be activated is different from the user's intention. The operation at this time will be described with reference to FIG. In FIG.
Steps before step S309 and steps after step S312 are the same as those in FIG. Step S30
When the keyword search is completed for all the character strings by 9 the evaluation values obtained for all the character strings in step S1201 are added for each item. From the results, the candidate applications to be launched are listed in descending order of the evaluation value,
The application having the highest evaluation value is set as a candidate to be activated, and the process proceeds to step S1202. Step S12
In step 02, the character string is converted into a format that can be used as data by the application in order to be passed to the activation candidate application, and the flow advances to step S1203. In step S1203, a screen for confirming the user to activate the activation candidate application determined in step S1201 is displayed on the display device 103, and step S1204 is performed.
Move to If the user approves in step S1204 that the presented application is to be activated, the process proceeds to step S312 to activate the application. If the user does not approve, step S12
05, the next candidate application is set as a candidate for starting the application having the next highest evaluation value, and the process returns to step S1202 to convert the character string into a format that can be used by the next candidate application, and then proceeds to step S1203 to start the next candidate application. Confirm with the user. Step S12
If the user does not approve the activation of the application presented in 04, the user selects an application to be activated via an application selection screen 1301 in which application names are listed and presented in descending order of evaluation values as shown in FIG. It is also possible.

As described above, according to an embodiment of the present invention, a recognition range is designated from a handwriting input to a recognition processing device, and an application related to a result obtained by performing character recognition and post-processing is automatically started. By doing so, it becomes possible to quickly use the character string whose input handwriting is recognized by an application or an external device.

[0053]

As described above, in the recognition processing device of the present invention, the character recognition is performed by specifying the range of recognition for the handwriting acquired by the handwriting acquisition means, post-processing is performed, and automatic correction is performed. By extracting a predetermined keyword from the extracted character string and analyzing the content, and automatically starting an application according to the analysis result, the handwriting input by the user can be quickly recognized and used by another application. Is obtained.

Further, character recognition is performed by designating a range to be recognized for the handwriting obtained by the handwriting obtaining means, post-processing is performed, and automatic correction is performed. A predetermined keyword is extracted from the obtained character string and the content is extracted. Is analyzed, and data is automatically transferred to an external device according to the analysis result, whereby an effect is obtained that handwriting input by a user can be promptly recognized and used by the external device.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of a recognition processing device according to an embodiment of the present invention.

FIG. 2 is a processing block diagram illustrating a flow of processing from handwriting input to application activation and operation of an external device of the recognition processing device according to an embodiment of the present invention.

FIG. 3 is a flowchart illustrating the operation of the recognition processing device according to an embodiment of the present invention.

FIG. 4 is a diagram illustrating an example of a handwriting input surface of the recognition processing device according to the embodiment of the present invention;

FIG. 5 is a diagram showing an example of a screen display when a person's name, a telephone number, a date and time of a meeting, a place, and contents are input to the recognition processing device according to the embodiment of the present invention;

FIG. 6 is a diagram illustrating an example of a screen display when a character recognition range is specified for a handwriting input to the recognition processing device according to an embodiment of the present invention.

FIG. 7 is a flowchart showing a flow of a keyword search of the recognition processing device according to the embodiment of the present invention;

FIG. 8 is a diagram illustrating an example of a screen display when an address book application is activated in the recognition processing device according to an embodiment of the present invention.

FIG. 9 is a diagram showing an example of a screen display when a recording reservation of a VCR and a facsimile transmission destination number are input to a recognition processing device according to an embodiment of the present invention, and a character recognition range is designated.

FIG. 10 is a diagram illustrating an example of a screen display when recording reservation data is transferred to a video deck in the recognition processing device according to an embodiment of the present invention.

FIG. 11 is a diagram illustrating an example of a screen display when a facsimile transmission destination number is designated as a character recognition range from a handwriting input in the recognition processing device according to an embodiment of the present invention.

FIG. 12 is a flowchart showing the operation of the recognition processing device when the activation candidate application presenting means for presenting applications in order of application relevance is added to the recognition processing device in the recognition processing device according to one embodiment of the present invention;

FIG. 13 is a diagram showing an example of a screen display of an application selection screen for listing and presenting application names in descending order of evaluation value in the recognition processing device according to the embodiment of the present invention.

[Explanation of symbols]

101 CPU 102 Handwriting input device 103 Display device 104 Main memory 105 External device interface 106 File device 107 ROM 201 Recognition range designation means 202 Character recognition means 203 Recognition dictionary 204 Post-processing means 205 Word dictionary 206 Recognition result analysis means 207 Keyword dictionary 208 Application automatic starting means 209 External device automatic operating means 401 Handwriting input surface 402 Mode selection switch 403 Stylus pen 501 Handwriting input example 601 Recognition range designation line 801 Screen display when address book application starts 1001 Screen display when VCR data transfer application starts 1301 Screen display of application selection screen

Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat II (reference) G06K 9/72 G06F 15/403 310C F-term (reference) 5B029 AA01 BB09 CC25 CC27 EE14 5B064 AA07 AA10 AB04 AB13 AB14 AB15 AB16 AB17 BA05 CA08 DD06 DD20 EA18 FA18 5B075 ND03 NK02 NK32 PP04 PP12 PP24 PQ74 PR06 QM08 QP05 UU09 5E501 AA11 AA12 AA14 AA30 AC33 BA20 CA02 CB06 CC04 FA13

Claims (13)

[Claims] 1. A recognition processing device having a display device and a handwriting input device, comprising: a handwriting acquisition unit for acquiring a handwriting input by a user; and a range for recognizing the handwriting obtained by the handwriting acquisition unit. A recognition range designating means for designating, a character recognition means for recognizing a handwriting within the range designated by the recognition range designating means, and a predetermined keyword extracted from a character string obtained by the character recognition processing means, and the contents are extracted. A recognition processing device comprising: a recognition result analyzing means for analyzing. 2. The apparatus according to claim 1, wherein the recognition result analysis unit includes a keyword dictionary that describes a degree of association between the keyword and an application corresponding to the keyword, and is described in the keyword dictionary in a character string obtained by the character recognition unit. 2. The recognition processing apparatus according to claim 1, further comprising an application relevance calculating unit that checks whether a keyword is present and calculates a relevance between the character string and the application. 3. The application relation degree calculating means obtains a character string candidate group and a recognition evaluation value of each character string from the character recognizing means, and obtains a relation degree between the application described in the keyword dictionary and a recognition evaluation value. 3. A degree of association between the character string candidate group and the application is calculated by performing a predetermined operation on the character string candidate group.
The recognition processing device according to the above. 4. The recognition processing device according to claim 2, wherein the keyword dictionary allows a user to change the degree of association between the content of the keyword and an application corresponding to the keyword. 5. The recognition processing device according to claim 1, further comprising: a post-processing unit that automatically corrects an erroneously recognized portion by collating a character string of a recognition result obtained by the character recognition unit with a word dictionary. Recognition processing device. 6. The recognition processing device according to claim 1, further comprising an application automatic start unit that automatically starts an application program according to the analysis result obtained by the recognition result analysis unit. Recognition processing device. 7. The automatic application activation means,
A recognition processing apparatus, comprising: a launch candidate application presenting means for presenting an application program to a user in ascending order of the degree of association obtained by the application degree of association calculating means. 8. The recognition processing device according to claim 6, wherein the application program is an address book application for registering a name, an address, and a telephone number. 9. The recognition processing device according to claim 6, wherein the application program is a schedule management application for registering a date, a place, and a schedule. 10. The recognition processing device according to claim 6, wherein the application program is a memo pad application for storing a memo. 11. The recognition processing device according to claim 1, wherein data obtained by converting the analysis result into a format usable by the external device is delivered to an external device corresponding to the analysis result obtained by the recognition result analysis means. A recognition processing device comprising external device automatic operation means for issuing a command. 12. The recognition processing device according to claim 11, wherein the external device is a video deck. 13. The recognition processing device according to claim 11, wherein the external device is a facsimile.