JP2011186994A - Character input device and character input method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To improve the operability of a character input regarding a character input device and a character input method. <P>SOLUTION: An image processing unit generates mouth-shaped image data by inputting mouth-shaped images. A mouth-shaped image collation database registers vowel information including vowels and syllabic nasals and mouth-shaped image data corresponding to each vowel information. A word conversion dictionary database registers word candidates associated with the vowel information. A control unit performs character input control. The control unit also converts mouth-shaped image data which are generated from mouth shape changes which are image-inputted into vowel information, and retrieves word candidates associated with converted vowel information. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a character input device and a character input method for inputting characters.

With the widespread use of mobile phones, opportunities to input text using mobile phones and create texts are increasing, and operability and efficiency of character input are required.
In a device having few input keys such as a cellular phone, it is generally performed by changing a character by pressing one key a plurality of times. In addition, voices are input using a microphone mounted on a mobile phone, and characters are input by voice recognition.

  When inputting characters by pressing one key multiple times as described above, the number of keys can be reduced to reduce the size of the device. However, if the same key is not pressed multiple times, the desired character cannot be input. Increases and the user operation becomes complicated.

For example, in a mobile phone having only a numeric keypad of “0 to 9”, when it is desired to input the character “ko”, the operation is troublesome, such as pressing “2” of the numeric keypad five times.
On the other hand, when inputting characters by voice recognition using a microphone, the complexity of key operations can be eliminated. However, there have been inconveniences such as the occurrence of poor recognition due to surrounding noise, or the fact that a character to be input can be heard by another person in order to make a voice.

  On the other hand, in recent years, devices have been developed that input characters by the movement of the mouth (lip) using a camera mounted on a mobile phone. In this method, an image of a user's mouth is captured by a camera, and character input is performed by mouth shape recognition. For example, Patent Documents 1 to 3 have been proposed as conventional techniques for performing character input by mouth shape recognition.

JP 2005-309952 A JP 2005-108079 A JP 2002-269544 A

  When character input is performed by mouth shape recognition, the input character that can be discriminated from the movement of the mouth becomes vowel information, so consonant information is input separately using other means. In Patent Documents 1 and 2 described above, a mouth shape image is input by a camera mounted on a mobile phone, vowel information of an input character is analyzed by mouth shape, and consonant information is input by key operation. Is going.

  However, in Patent Documents 1 and 2, a consonant is input to the vowel one character at a time, so it is necessary to link the movement of the mouth and the key input operation every time one character is input. Easy to miss. For this reason, it is not easy to input characters accurately and smoothly, and it cannot be said that operability is good.

  In Japanese Patent Application Laid-Open No. 2004-260260, characters can be designated by eye and face movements for people who have difficulty in key operations. However, it is necessary to link the mouth and eye and face movements every time a character is input. Therefore, the character input operation is also complicated in this case.

The present invention has been made in view of these points, and an object of the present invention is to provide a character input device that improves operability in character input by mouth shape recognition.
Another object of the present invention is to provide a character input method that improves operability in character input by mouth shape recognition.

  In order to solve the above problems, a character input device is provided. The character input device includes an image processing unit that inputs a mouth shape image to generate mouth shape image data, vowel information including vowels and sound repellent, and mouth shape image data corresponding to each of the vowel information. A mouth shape image matching database for registering words, a word conversion dictionary database for registering word candidates related to the vowel information, and a control unit for performing character input control. In addition, the control unit converts the mouth shape image data into the vowel information, and searches for the word candidates related to the converted vowel information.

  It becomes possible to improve operability.

It is a figure which shows the structural example of a character input device. It is a figure which shows the external appearance structure of a mobile telephone. It is a figure which shows the functional block of a mobile telephone. It is a figure which shows the example of a display of a character input screen. It is a figure which shows the example of a screen display of the state which captured the image of the mouth of "A". It is a figure which shows the example of a screen display of a word candidate. It is a figure which shows the screen display at the time of selecting an input character. It is a figure which shows the screen display at the time of confirming an input character. It is a figure which shows the registration data of mouth shape image collation DB. It is a figure which shows the registration data of word conversion dictionary DB. It is a figure which shows the example of a screen display of a word candidate. It is a figure which shows the screen display in the case of selecting an input character. It is a figure which shows the screen display at the time of confirming an input character. It is a figure which shows the example of a display of a character input screen. It is a figure which shows the flowchart of a Japanese character input operation. It is a figure which shows the example of a display of a character input screen. It is a figure which shows the registration data of alphabet conversion dictionary DB. It is a figure which shows the example of a screen display of an alphabet character candidate. It is a figure which shows the screen display at the time of selecting an input character. It is a figure which shows the screen display which fixes an input character. It is a figure which shows switching of the full-width / half-width of the alphabet. It is a figure which shows the screen display which fixes an input character. It is a figure which shows the flowchart of an alphabet character input operation | movement. It is a figure which shows the external appearance structure of a mobile telephone.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a diagram illustrating a configuration example of a character input device. The character input device 1 includes a user interface unit 10, an image processing unit 20, a mouth shape image collation database 31, a word conversion dictionary database 32a, and a control unit 40. For example, the character input device 1 is used in a communication device such as a mobile phone where character input is performed. Applicable.

  The user interface unit 10 performs key operations and screen display. The image processing unit 20 inputs an image of the mouth shape when sound is generated, and generates mouth shape image data. The mouth shape image collation database 31 registers vowel information including vowels and repellent sounds and mouth shape image data corresponding to individual vowel information. The word conversion dictionary database 32a registers word candidates related to vowel information. The control unit 40 performs character input control.

  Here, the control unit 40 collates one or a plurality of mouth shape image data generated from the mouth shape change inputted as an image with vowel information registered in the mouth shape image collation database 31, and collectively Convert to vowel information.

  For example, if there is only one change in the mouth shape input as an image, one mouth shape image data is generated. Then, the corresponding mouth shape image data is collated with the vowel information registered in the mouth shape image collation database 31 and converted into one corresponding vowel information.

  Alternatively, for example, if there are three different changes in the mouth shape input as an image, three mouth shape image data are generated. Then, the three corresponding mouth shape image data are collated with the vowel information registered in the mouth shape image collation database 31, respectively, and converted into the corresponding three vowel information.

After the mouth shape image data is converted into vowel information, word candidates related to the converted vowel information are searched using the word conversion dictionary database 32a and displayed on the screen.
At this time, if there is a corresponding word candidate, it is determined by the user, but if there is no corresponding word candidate among the word candidates including the converted vowel information, the user inputs consonant information, A related word candidate is searched from the vowel information and the input consonant information. The detailed operation will be described later with a specific example.

  Next, the configuration and operation when the character input device 1 is applied to a mobile phone will be described in detail. FIG. 2 is a diagram showing an external configuration of the mobile phone, and FIG. 3 is a diagram showing functional blocks of the mobile phone. The cellular phone 1a includes a user interface unit 10, an image processing unit 20, a memory unit 30, and a control unit 40.

  The user interface unit 10 includes a screen display unit (LCD: Liquid Crystal Display) 11 and an input operation unit 12, and the input operation unit 12 includes a numeric keypad 12a, a selection key 12b, and a confirmation key 12c. The image processing unit 20 includes a camera 21 and an image data generation unit 22.

  The memory unit 30 corresponds to a flash memory or the like, and includes a mouth shape image collation database (DB) 31, a word conversion dictionary DB 32a, and an alphabet conversion dictionary DB 32b. The control unit 40 includes a CPU (Central Processing Unit) 41, a RAM (Random Access Memory) 42, and a ROM (Read Only Memory) 43.

  The camera 21 inputs an image. When inputting characters by mouth shape recognition, an image of the user's mouth area is input. The image data generation unit 22 converts the input image into data. When an image of the mouth area is input, the mouth shape image is converted into data to generate mouth shape image data.

  The mouth shape image collation DB 31 stores vowel information including vowels (“A”, “I”, “U”, “E”, “O”) and sound repellent (“N”), and mouths corresponding to individual vowel information. Register shape image data.

  The word conversion dictionary DB 32a registers word candidates related to vowel information. The alphabet conversion dictionary DB 32b registers alphabet candidates related to vowel information. The word conversion dictionary DB 32a is used in the Japanese (kana) character input mode, and the alphabet conversion dictionary DB 32b is used in the alphabet input mode.

  From the generated mouth shape image data, the control unit 40 recognizes the vowel and “n” shape in units of words using the mouth shape image matching DB 31. Then, based on the obtained vowels and vowel information of “n”, word candidates related to the corresponding vowel information are extracted from the word conversion dictionary DB 32a.

  The screen display unit 11 displays the extracted word candidates. The user uses the selection key 12b to select a word to be input from the displayed word candidates. The selection key 12b is a key for selecting a desired word from a plurality of word candidates based on a user operation. The confirmation key 12c is a key for confirming a selected word and determining a word to be input by a user operation.

  If there is no word to be input among the displayed word candidates, the user inputs the first character of the consonant information of the word to be input using the numeric keypad 12a. The numeric keypad 12a is a key for inputting consonant information based on a user operation.

  The control unit 40 searches the word candidate dictionary previously extracted from the word conversion dictionary DB 32a by adding the first character consonant information, extracts the word candidate as a result, and the screen display unit 11 is extracted. Display the word candidates. If the word to be input is not displayed even in this state, the word candidate is further searched by inputting the second character consonant information.

  Next, the character input operation of the mobile phone 1a will be described in detail with a specific example. First, as an example of Japanese character input, a case where the word “love” is input will be described.

  FIG. 4 is a diagram showing a display example of the character input screen. On the screen 11-1 when creating the mail text, a confirmed input character display section 11a (corresponding to the mail text) for displaying the confirmed input characters, a key operation guidance section 11b for guiding the user to perform key operations, and A candidate character display portion 11c for displaying input character candidates is displayed.

  First, the user holds the mobile phone 1a and moves the mouth of “A” toward the camera 21 to input a mouth shape image. FIG. 5 is a diagram illustrating a screen display example in a state in which an image of the mouth “A” is captured. The vowel corresponding to the inputted mouth shape image data is collated from the mouth shape image collation DB 31 to recognize that the vowel “a” is inputted.

  In the screen 11-2, the candidate character display unit 11c displays hiragana uppercase “a”, katakana uppercase “a”, hiragana lowercase “a”, katakana as characters recognized from the “a” mouth image. A lowercase “A” is displayed. In addition, “a” is displayed in the input character display portion 11d before confirmation that displays the input character before confirmation.

  Next, the user moves the mouth of “I”. FIG. 6 is a diagram illustrating a screen display example of word candidates. The user moves the mouth of “I” and stops the mouth movement toward the camera 21. Then, the vowel corresponding to the inputted mouth shape image data is collated from the mouth shape image matching DB 31 to recognize that the vowel “i” is inputted. Then, word candidates related to “ai” are extracted from the word conversion dictionary DB 32a.

  On the screen 11-3, the extracted word is displayed on the candidate character display unit 11c. In the case of FIG. 6, six words are extracted and displayed. Further, “ai” is displayed in the input character display portion 11d before confirmation.

  FIG. 7 is a diagram showing a screen display when an input character is selected. Here, since the character “love” is displayed on the candidate character display portion 11c of the screen 11-4, the character “love” is selected by pressing down the selection key 12b twice on the screen 11-4. .

  FIG. 8 is a diagram showing a screen display when an input character is confirmed. After selecting the input character as shown in FIG. 7, the character “love” is expanded in the mail text of the confirmed input character display portion 11a on the screen 11-5 by pressing the confirm key 12c once.

  Next, the mouth shape image collation DB 31 and the word conversion dictionary DB 32a will be described. FIG. 9 is a diagram showing registration data in the mouth shape image collation DB 31. Mouth shape image matching DB 31 corresponds to vowel information including vowels of a (a), i (i), u (u), e (e), o (o) and n (n), and each vowel information. Mouth shape image data is registered.

  Mouth shape image data, which is mouth shape image data when a character is pronounced, is divided into system standard data and user registration data. The system standard data is system standard mouth shape image data registered and set in advance in the mouth shape image matching DB 31. The user registration data is mouth shape image data registered by an individual user who inputs his mouth shape with a camera.

  In this way, assuming a plurality of users using the mobile phone 1a, in addition to the system standard, the user himself / herself can register mouth shape image data, and each mouth shape image data of the plurality of users is used as vowel information. It was set as the structure preserve | saved correspondingly. Thereby, it is possible to improve the accuracy of collation between the vowel information and the mouth shape image data.

  Here, when the mouth shape image data generated by inputting an image through the camera 21 is converted into the vowel information of “A” to “O” or “n”, it matches the input mouth shape image data. Vowel information to be detected is detected from the mouth shape image matching DB 31. Then, the mouth shape image data is converted into vowel information, and the corresponding vowel character is displayed on the candidate character display unit 11c.

  FIG. 10 is a diagram showing registration data in the word conversion dictionary DB 32a. The example of the registration data for displaying a word candidate (Japanese) from the vowel information of a vowel and "n" is shown. Registered items include “vowel information”, “reading”, “key”, and “display”. It should be noted that the corresponding characters described in “Display” are displayed by pressing the number shown in the “Key” column with the numeric keypad 12a.

  Next, a case where the word “lower” is input will be described. The user holds the mobile phone 1a, moves the mouths of “ka” and “i” toward the camera 21, and inputs a mouth shape image. Since the vowel of the word “lower” is “ai”, the mouth shape of “ka” is converted to the vowel “a”, and the mouth shape of “i” is converted to the vowel “i”. The screen operation is the same screen operation up to the above-described FIG.

  Since the character “lower” to be input is not displayed in the candidate character display portion 11c of FIG. 6, the user inputs “ka”, which is the first consonant of “lower”. Press the “2” key once.

  FIG. 11 is a diagram illustrating a screen display example of word candidates. The control unit 40 searches the word conversion dictionary DB 32a for words related to the vowel “ai” and the first character consonant “ka”, and displays the searched word candidates on the candidate character display unit 11c on the screen 11-6. . In the case of the figure, six word candidates are displayed.

  FIG. 12 is a diagram showing a screen display when an input character is selected. Since the character “lower” to be input exists in the candidate character display portion 11c, the selection key 12b is pressed four times downward on the screen 11-7 for selection.

  FIG. 13 is a diagram showing a screen display when an input character is confirmed. After selecting an input character as shown in FIG. 12, by pressing the enter key 12c once, a “lower” character is expanded in the mail text of the confirmed input character display portion 11a on the screen 11-8.

  If the word candidate to be input is not displayed when the word candidate of FIG. 11 is displayed, the second character consonant information is input with the numeric keypad 12a to search the word conversion dictionary DB 32a, and the candidate character display unit A character to be input is displayed in 11c.

  Next, a modified example of the screen display will be described. FIG. 14 is a diagram showing a display example of the character input screen. As shown in FIG. 5, the mouth shape image d <b> 1 input by the camera 21 is displayed in the empty space on the screen 11-2 a when the mouth movement is input. By performing such display, the user can perform an operation while confirming the movement of his / her mouth, so that the recognition rate of vowels and “n” can be improved.

Next, a character input operation will be described using a flowchart. FIG. 15 is a flowchart of the Japanese character input operation.
[S1] Using the camera 21, the user inputs an image of mouth movement character by character.

[S2] The mouth shape image data is converted into vowel information using the mouth shape image collation DB 31 from the input mouth movement.
[S3] It is determined whether or not the mouth shape has moved during a certain time (whether or not it has changed). If there is no movement of the mouth for a certain time, the process goes to step S5, and if the mouth moves for a certain time, the process goes to step S4.

[S4] The next character is input from the camera 21. Return to step S2.
[S5] From the m vowel information analyzed from the mouth movement, n word candidates related to the vowel or “n” are displayed on the candidate character display unit 11c using the word conversion dictionary DB 32a.

[S6] Wait for key input.
[S7] The input key is recognized. If it is the selection key 12b, go to step S8a, if it is the numeric keypad 12a, go to step S8b, and if it is the enter key 12c, go to step S8c. If the character input is to be terminated, the process is terminated.

[S8a] When the selection key 12b is input, the cursor is moved in the corresponding direction, and the process returns to step S6 to enter a key input waiting state.
[S8b] When the numeric keypad 12a is input, the related word candidate is searched from the word conversion dictionary DB 32a by adding the input consonant information to the corresponding word, and the search result is displayed on the candidate character display unit 11c. To do. Returning to step S6, a key input waiting state is entered.

[S8c] When the confirmation key 12c is input, the confirmation character is displayed on the confirmation input character display portion 11a.
[S8c-1] The word conversion dictionary DB 32a updates the registration order so that the word conversion dictionary DB 32a is displayed at a higher position on the screen in the descending order of the frequency of words to be determined.

  As described above, the mobile phone 1a to which the character input device 1 is applied captures an image of mouth movement at the character or word level in Japanese input, and analyzes the vowel and “n” of the corresponding word. Then, word candidates related to the vowel input by the movement of the mouth and the vowel information of “n” are searched from the word conversion dictionary DB 32a and displayed on the display.

  If a word to be input is included in the word candidates of only the vowel information displayed on the display, the input character is selected and confirmed by the selection key 12b and the confirmation key 12c. If the word candidate displayed on the display does not include the word to be input, for example, the corresponding consonant is input with the numeric keypad 12a of “0-9”, and additional consonants are sequentially input until the candidate appears. Search.

  In this way, only the vowel information is input in units of words, and the word conversion dictionary DB 32a is first searched only with the vowel information. Become.

  With such a configuration, an operation of inputting consonants in accordance with the vowels character by character becomes unnecessary. Accordingly, since it is not necessary to link the mouth-shaped image input and the key input operation every time a character is input, the operation is facilitated, and the characters can be input accurately and smoothly.

Next, alphabetic character input will be described. As an example of alphabet input, a case where “g” is input will be described.
FIG. 16 is a diagram showing a display example of a character input screen. On the screen 11-1a, the character input mode is switched to the alphabet input mode by pressing the character switch key 11b-1 shown in the key operation guidance unit 11b.

  Here, in order to input the letter “g”, the user inputs the movement of the mouth from the camera 21 with the mouth shape as “ji”, and the corresponding mouth shape image data from the mouth shape image matching DB 31 described above. The vowel information “i” corresponding to is obtained.

  Next, in order to obtain alphabet information corresponding to “i”, a candidate alphabet is extracted using the alphabet conversion dictionary DB 32b and displayed on the candidate character display unit 11c.

  FIG. 17 is a diagram showing registration data in the alphabet conversion dictionary DB 32b. An example of registered data for displaying alphabet candidates from vowel information is shown. Registered items include “vowel information”, “reading”, “key”, and “display”. Note that “display” includes full-size (full-width lowercase letters), full-size (full-size uppercase letters), half-small (half-size lowercase letters), and half-large (half-size uppercase letters).

  FIG. 18 is a diagram showing a screen display example of alphabet character candidates. On the screen 11-2a, the alphabet corresponding to the vowel “i” is displayed as “b”, “e”, “g”, “p”, “t”, “z” and a plurality of candidates.

  FIG. 19 is a diagram showing a screen display when an input character is selected. By pressing “4” of the numeric keypad 12a corresponding to “g” to be input on the screen 11-3a, only “g” is displayed on the candidate character display portion 11c. Alternatively, in the state of the screen 11-2a, the selection key 12b is pressed three times to move the cursor downward and select “g”.

  FIG. 20 is a diagram showing a screen display for confirming an input character. After selecting the input character as shown in FIG. 19, the character “g” is expanded in the mail text of the confirmed input character display portion 11a on the screen 11-4a by pressing the confirm key 12c.

  FIG. 21 is a diagram illustrating switching between full-width / half-width alphabets. In the state of FIG. 19, the character of “display” in the alphabet conversion dictionary DB 32 b is switched by pressing the full half size button 11 b-2. As a result, the alphabet can be switched between “full-width / half-width” and “uppercase / lowercase”. The screen 11-5a is an example of a state where the full half size button 11b-2 is pressed once, and indicates that the letter “g” is “G”.

  FIG. 22 is a diagram showing a screen display for confirming an input character. By selecting the input character as shown in FIG. 21 and then pressing the confirmation key 12c, the character “G” is expanded in the mail text of the confirmed input character display portion 11a on the screen 11-6a.

Next, a character input operation will be described using a flowchart. FIG. 23 is a flowchart of the alphabet character input operation.
[S11] The user uses the camera 21 to input an image of mouth movement character by character.

[S12] The mouth shape image data is converted into vowel information from the input mouth movement using the mouth shape image matching DB 31.
[S13] It is determined whether or not the mouth shape moves during a certain time. If the mouth does not move for a certain time, the process goes to step S15. If the mouth moves for a certain time, the process goes to step S14.

[S14] The next character is input from the camera 21. Return to step S12.
[S15] From the m vowel information analyzed from the movement of the mouth, n related alphabet candidates are displayed on the candidate character display unit 11c using the alphabet conversion dictionary DB 32b.

[S16] It is in a key input waiting state.
[S17] The input key is recognized. In the case of the selection key 12b, the process proceeds to step S18a, in the case of the numeric keypad 12a, the process proceeds to step S18b, and in the case of the confirmation key 12c, the process proceeds to step S18c. If the character input is to be terminated, the process is terminated.

[S18a] When the selection key 12b is input, the cursor is moved in the corresponding direction, and the process returns to step S16 to enter a key input waiting state.
[S18b] When the numeric keypad 12a is used, based on the key information (corresponding to the “key” in the alphabet conversion dictionary DB 32b), alphabet candidates are specified and input and displayed on the candidate character display unit 11c. The process returns to step S16 and waits for key input.

[S18c] When the confirmation key 12c is input, the confirmation character is displayed on the confirmation input character display portion 11a.
[S18c-1] The alphabet conversion dictionary DB 32b updates the registration order so that the alphabetical conversion dictionary DB 32b is displayed at a higher position on the screen in the descending order of the alphabet frequency to be determined.

  As described above, the mobile phone 1a to which the character input device 1 is applied has an alphabet input mode, takes an image of mouth movement for each alphabet, and analyzes the vowel information of the corresponding alphabet. Then, one alphabetic character related to vowel information input by mouth movement from the alphabet conversion dictionary DB 32b is displayed on the display.

  If the alphabet to be input is included in the alphabet candidates displayed on the display, the input character is confirmed by the selection key 12b and the confirmation key 12c (or the alphabet is directly specified by the numeric keypad 12a).

  When the alphabet to be input is not included in the candidates displayed on the display, the alphabet candidate is specified and input by inputting the corresponding alphabet with, for example, the numeric keypad 12a of “0-9”. Thus, not only Japanese input but also alphabet input can be performed.

  FIG. 24 is a diagram showing an external configuration of a mobile phone. In the mobile phone 1a, the input operation is performed with a key such as a numeric keypad. However, a mobile phone 1b having an input unit such as the touch panel 110 may be used. Since the basic configuration is the same as that of the mobile phone 1a, the description is omitted.

  As described above, the character input device 1 includes the mouth shape image collation DB 31 for registering mouth shape image data corresponding to individual vowel information, and the word conversion dictionary DB 32a for registering word candidates related to vowel information. Thus, the mouth shape image data is converted into vowel information, and word candidates related to the converted vowel information are searched.

  In conventional character input by mouth shape recognition, characters are input in units of character in conjunction with vowel input by mouth shape recognition and consonant input by key operation etc., so operability is poor and character input efficiency Was low.

  On the other hand, in the character input device 1, the mouth shape image data is first converted into vowel information by mouth shape recognition, and only the vowel input is input first, for example, in units of words, and the word conversion dictionary DB 32a. Is used to search for word candidates related to the input vowel information. When the desired word does not exist, the consonant information is input by key operation, the word candidate is searched again, and the desired word is input.

  As described above, when the desired word is not displayed only by the vowel information, the convolution information is sequentially input to improve the search accuracy. The trouble of inputting characters by performing operations is eliminated, and the operability can be significantly improved compared to conventional character input.

  Moreover, even in a portable terminal device having only a numeric keypad of “0-9”, the mouth shape recognition character input function of the character input device 1 facilitates operability including alphabets by a minimum key operation even in crowds. Character input becomes possible.

  The processing functions of the character input device 1 can be realized by a computer. In that case, a program (character input control program) describing the processing contents of the functions that the character input device 1 should have is provided. By executing the program on a computer, the above processing functions are realized on the computer. The program describing the processing contents can be recorded on a computer-readable recording medium.

  The entire computer is controlled by a CPU. A RAM, a hard disk drive (HDD), a communication interface, a graphic processing device, and an input / output interface are connected to the CPU via a bus.

  The RAM temporarily stores at least a part of an OS (Operating System) program to be executed by the CPU and a program for performing character input control. The RAM stores various data necessary for processing by the CPU. The HDD message stores the OS and application programs.

  The communication interface is connected to the network. The communication interface transmits / receives data to / from other computers via a network. The graphic processing apparatus is connected to a monitor. The graphic processing device displays an image on a monitor screen in accordance with a command from the CPU.

  A keyboard and a mouse are connected to the input / output interface. The input / output interface transmits signals sent from the keyboard and mouse to the CPU via the bus. The input / output interface can be connected to an external storage medium interface capable of writing information to the external storage medium and reading information from the external storage medium.

  The character input device 1 can be realized by executing a character input control program describing the processing contents of each function by a computer. That is, processing contents corresponding to the user interface unit 10, the image processing unit 20, the mouth shape image collation database 31, the word conversion dictionary database 32a, and the control unit 40 of FIG. 1 are described as a program. Here, the described program can be recorded on a computer-readable recording medium.

  Examples of the computer-readable recording medium include a magnetic storage device, an optical disk, a magneto-optical recording medium, and a semiconductor memory. Magnetic storage devices include hard disk devices, flexible disks (FD), and magnetic tapes. Optical discs include DVD, DVD-RAM, CD-ROM / RW, and the like. Magneto-optical recording media include MO (Magneto-Optical disc).

  When distributing the program, for example, a portable recording medium such as a DVD or a CD-ROM in which the program is recorded is sold. It is also possible to store the program in a storage device of a server computer and transfer the program from the server computer to another computer via a network.

  In addition, at least a part of the above processing functions can be realized by an electronic circuit such as a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), or a PLD (Programmable Logic Device).

  The computer that executes the program stores, for example, the program recorded in the external storage medium or the program transferred from the server program in its own storage device. Then, the computer reads the program from its own storage device and executes processing according to the program. The computer can also read a program directly from an external storage medium and execute processing according to the program. Further, each time the program is transferred from the server computer, the computer can also execute processing according to the sequentially received program.

  As mentioned above, although embodiment was illustrated, the structure of each part shown by embodiment can be substituted by the other thing which has the same function. Moreover, other arbitrary structures and processes may be added. In addition, although the example which applied the character input device 1 to the mobile telephone was shown above, it is applicable not only to a mobile telephone but to the whole apparatus which mounts a camera and performs character input.

DESCRIPTION OF SYMBOLS 1 Character input device 10 User interface part 20 Image processing part 31 Mouth shape image collation database 32a Word conversion dictionary database 40 Control part

Claims (6)

An image processing unit for inputting mouth shape images and generating mouth shape image data;
A mouth shape image collation database for registering vowel information including vowels and repellent sounds, and the mouth shape image data corresponding to each vowel information;
A word conversion dictionary database for registering word candidates related to the vowel information;
A control unit for performing character input control,
The control unit converts the mouth shape image data into the vowel information, and searches for the word candidates related to the converted vowel information.
A character input device characterized by that.   The said control part searches the said word candidate from the said vowel information and the input consonant information, when there is no applicable word candidate in the said word candidate containing the said vowel information. The character input device according to 1. An alphabet conversion dictionary database for registering alphabet candidates related to the vowel information;
When the alphabet character input mode is set, the control unit converts the mouth shape image data into the vowel information, and searches for the alphabet candidates related to the converted vowel information.
The character input device according to claim 1.   The character input device according to claim 1, wherein the control unit displays a mouth shape at the time of image input on a screen. In the character input method,
Mouth shape image data is generated by inputting the mouth shape image,
Register vowel information including vowels and repellent sounds and the mouth shape image data corresponding to each vowel information in a database,
Register word candidates related to the vowel information in the database;
The mouth shape image data is converted into the vowel information, and the word candidates related to the converted vowel information are searched.
Character input method characterized by this. In a character input control program for causing a computer to execute character input,
In the computer,
Mouth shape image data is generated by inputting the mouth shape image,
Register vowel information including vowels and repellent sounds and the mouth shape image data corresponding to each vowel information in a database,
Register word candidates related to the vowel information in the database;
The mouth shape image data is converted into the vowel information, and the word candidates related to the converted vowel information are searched.
A character input control program for executing a process.

Images