JP2013077243A - Character input device, character input system and character input method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a character input device, character input system and character input method capable of efficiently inputting a characters without being affected by a failure of character recognition.SOLUTION: A character input device 1 includes an acquisition part 103 for acquiring an image, a character recognition processing part 107 for extracting a character candidate group on the basis of character recognition processing to the image acquired by the acquisition part 103, a language processing part 109 for extracting a word candidate group on the basis of the character candidate group extracted by the character recognition processing part 107, and a candidate group display part 112 for displaying the character candidate group extracted by the character recognition processing part 107 and the word candidate group extracted by the language processing part 109 as an input candidate group during character inputting.

Description

  The present invention relates to a character input device having a character input function, a character input system including the character input device, and a character input method in the character input device.

  Conventionally, IME (Input Method Editor) is known as software for inputting characters. When performing kana-kanji conversion in Japanese using general IME, the user needs to know in advance how to read the kanji to be input and the corresponding character code, or input from a large list of kanji You need to find and select the kanji you want. The same applies to other languages such as Chinese.

  Conventionally, IME using online character recognition technology that estimates an input character from a writing order when a user inputs characters by handwriting is known. However, in online character recognition, there are cases where character input cannot be performed with only the appearance of the characters, such as erroneous recognition occurring when the character writing order is different. Also, in online character recognition, input is similar to handwriting, so that complicated character input requires a lot of time for character input.

  On the other hand, Patent Document 1 discloses an apparatus that performs character recognition on characters included in an image and performs kana-kanji conversion using a confirmed character recognition result.

JP 2010-191907 A

  Generally, in character recognition, there is a possibility that character recognition failure occurs, that is, erroneous recognition occurs. In the apparatus of Patent Document 1, when a character recognition failure occurs, a problem that the character input cannot be performed efficiently occurs because the confirmed erroneous recognition result is used for kana-kanji conversion.

  Therefore, the present invention has been made in view of the above, and provides a character input device, a character input system, and a character input method capable of efficiently inputting characters without being influenced by a failure in character recognition. Objective.

  In order to solve the above problems, a character input device of the present invention is based on an acquisition unit that acquires an image, a character recognition processing unit that extracts a character candidate group based on a character recognition process on the image, and a character candidate group. The language processing means for extracting the word candidate group and the candidate group display means for displaying the character candidate group or the word candidate group as an input candidate group at the time of character input.

  The character input method of the present invention is a character input method in a character input device, in which an acquisition unit acquires an image, and a character recognition processing unit is a character candidate based on a character recognition process for the image. A character recognition processing step for extracting a group; a language processing unit for extracting a word candidate group based on the character candidate group; and a candidate group display unit for inputting a character candidate group or a word candidate group. A candidate group display step for displaying as a candidate input group at the time.

  According to such a character input device and character input method of the present invention, the acquisition unit acquires an image, and the character recognition processing unit extracts a character candidate group based on the character recognition processing for the acquired image, The language processing means extracts a word candidate group based on the extracted character candidate group. And a candidate group display means displays the extracted character candidate group or word candidate group as an input candidate group at the time of character input. Thus, in the present invention, a character candidate group or a word candidate group based on the character recognition processing for the acquired image is displayed as an input candidate group at the time of character input. For example, the character candidate group includes a plurality of character candidates as a result of character recognition processing, and the word candidate group includes a plurality of word candidates composed of combinations of the plurality of character candidates. By selecting from the character candidate group or the word candidate group, it is possible to input characters efficiently without being affected by the failure of character recognition.

  Further, in the present invention, based on the character information or word information associated with the image acquired by the acquisition means, further comprising a candidate group extraction means for extracting a character candidate group or word candidate group, the candidate group display means, You may further display the character candidate group or word candidate group extracted by the candidate group extraction means.

  According to the present invention, the candidate group extraction unit extracts the character candidate group or the word candidate group based on the character information or the word information associated with the image, and the candidate group display unit extracts the extracted character candidate group or the word The candidate group is further displayed. As described above, the character candidate group or the word candidate group based on the character information or the word information associated with the image is further displayed as the input candidate group at the time of character input, so that the character input can be performed efficiently.

  In the present invention, an imaging unit may be further provided, and the acquisition unit may acquire an image captured by the imaging unit.

  According to the present invention, an image captured by the imaging unit can be used as a target for character recognition processing. For example, if a user can capture an image that includes a character or a word, the user can capture the image and acquire it as an image including the character or word, thereby using the character or word as an input candidate group when inputting characters. Therefore, it is possible to input characters efficiently.

  In the present invention, image storage means for storing an image may be further provided, and the acquisition means may acquire the image stored by the image storage means.

  According to the present invention, since the image stored by the image storage means can be the target of character recognition processing, for example, if the user stores an image containing characters or words by the image storage means, the image storage means Since characters or words can be used as input candidate groups when inputting characters, it is possible to efficiently input characters.

  In the present invention, the touch panel, the image display means for displaying the image acquired by the acquisition means on the touch panel, and a partial area of the image displayed by the image display means based on the touch operation on the touch panel are shown. Area extraction means for extracting area information, and the area extraction means may output an image corresponding to a partial area specified by the extracted area information to the character recognition processing means.

  According to this invention, the image display means displays an image on the touch panel. Then, the area extracting means extracts area information indicating a partial area of the image based on a touch operation on the touch panel, and an image corresponding to the partial area specified by the extracted area information is converted into a character recognition processing means. Output to. Thereby, for example, the user selects a partial area including a necessary character by a touch operation while viewing an image displayed on the touch panel, and sets an image corresponding to the selected partial area as a target of character recognition processing. Therefore, it is possible to input characters efficiently.

  Further, in the present invention, the area extracting means moves the image based on a single touch operation on the touch panel after fixing a frame for extracting a partial area, and based on the multi touch operation on the touch panel. The image may be enlarged or reduced. The single touch operation refers to an operation in which a user moves a finger on the touch panel while touching the touch panel. The multi-touch operation refers to an operation in which a user moves each finger on the touch panel while keeping two or more fingers in contact with the touch panel.

  According to the present invention, the user can efficiently select a partial region in an image with a simple operation.

  In the present invention, the area extracting means moves the frame for extracting a partial area based on the single touch operation on the touch panel after fixing the image, and based on the multi touch operation on the touch panel. The frame may be enlarged or reduced.

  According to the present invention, the user can efficiently select a partial region in an image with a simple operation.

  The character input system of the present invention is a character input system including a character input device and a server that can communicate with each other via a network. The character input device includes an acquisition means for acquiring an image, and character recognition for the image. Character recognition processing means for extracting a character candidate group based on processing, client transmission means for transmitting the character candidate group to the server, client receiving means for receiving the word candidate group from the server, character candidate group or word candidate group A candidate group display means for displaying a character candidate group from a character input device, and a server candidate means received by the server receiver means. Based on the language processing means for extracting the word candidate group and the server transmitting means for transmitting the word candidate group extracted by the language processing means to the character input device , Characterized in that it comprises a.

  According to such a character input system of the present invention, a character input group is extracted based on character recognition processing on an image by a character input device, and a word candidate group is extracted based on the character candidate group at a server. To do. In this way, since the word candidate group is extracted on the server side, the processing load on the character input device can be reduced.

  The character input system of the present invention is a character input system including a character input device and a server that can communicate with each other via a network. The character input device includes an acquisition means for acquiring an image, and the image to the server. Client transmitting means for transmitting; client receiving means for receiving a character candidate group or word candidate group from the server; and candidate group display means for displaying the character candidate group or word candidate group as an input candidate group at the time of character input. A server receiving means for receiving an image from the character input device, a character recognition processing means for extracting a character candidate group based on a character recognition process for the image received by the server receiving means, and a character recognition processing means. The language processing means for extracting the word candidate group based on the character candidate group extracted by the above, and the character candidate group extracted by the character recognition processing means Characterized in that the provided word candidates extracted by the language processing unit, a server sending means for sending the character input device, the.

  According to such a character input system of the present invention, a server extracts a character candidate group based on a character recognition process for an image, and extracts a word candidate group based on the character candidate group. Thus, since extraction of a character candidate group and a word candidate group is performed by the server side, the processing load in a character input device can be reduced.

  ADVANTAGE OF THE INVENTION According to this invention, the character input device, character input system, and character input method which can perform character input efficiently, without being influenced by failure of character recognition can be provided.

1 is a schematic configuration diagram of a character input device 1 according to a first embodiment of the present invention. It is a hardware block diagram of the character input device 1 and the server 2. It is a flowchart which shows operation | movement of the character input device 1 which concerns on 1st Embodiment of this invention. It is FIG. (1) which shows the Example of the character input device 1 which concerns on 1st Embodiment of this invention. It is FIG. (2) which shows the Example of the character input device 1 which concerns on 1st Embodiment of this invention. It is a structure schematic diagram of the character input system 3A which concerns on 2nd Embodiment of this invention. It is a structure schematic diagram of the character input system 3B which concerns on 3rd Embodiment of this invention.

  DESCRIPTION OF EMBODIMENTS Hereinafter, preferred embodiments of a character input device, a character input system, and a character input method according to the present invention will be described in detail with reference to the accompanying drawings. In the description of the drawings, the same elements are denoted by the same reference numerals, and redundant description is omitted.

(Configuration of the character input device 1 according to the first embodiment)
First, the configuration of the character input device 1 according to the first embodiment of the present invention will be described with reference to FIG. As the first embodiment, the character input device 1 does not communicate with other devices, and has a stand-alone configuration that completes character recognition processing and language processing within the device itself. FIG. 1 is a schematic configuration diagram of the character input device 1. The character input device 1 is a mobile phone terminal including an imaging unit 101 and a touch panel 104 described later. The character input device 1 may be a notebook PC (Personal Computer), a PC, or the like.

  FIG. 2 is a hardware configuration diagram of the character input device 1. As shown in FIG. 2, the character input device 1 physically includes a CPU 201, a main storage device such as a ROM 202 and a RAM 203, and a communication module 204 for transmitting / receiving data to / from other external devices, which will be described later. The image forming unit 101, the input / output device 205 such as the touch panel 104, the auxiliary storage device 206, and the like are configured as a normal mobile phone terminal. Each function of the character input device 1 to be described later operates the communication module 204 and the input / output device 205 under the control of the CPU 201 by reading predetermined computer software on hardware such as the CPU 201, the ROM 202, and the RAM 203. It is realized by reading and writing data in the ROM 202, RAM 203 and auxiliary storage device 206.

  Returning to FIG. 1, the character input device 1 functionally corresponds to an imaging unit 101 (corresponding to “imaging unit” in the claims) and an image storage unit 102 (corresponding to “image storage unit” in the claims). ), An acquisition unit 103 (corresponding to “acquisition means” in the claims), a touch panel 104 (corresponding to “touch panel” in the claims), and an image display unit 105 (“image display means” in the claims). Equivalent), region extraction unit 106 (corresponding to “region extraction means” in claims), character recognition processing unit 107 (corresponding to “character recognition processing means” in claims), character recognition dictionary 108, language A processing unit 109 (corresponding to “language processing means” in claims), a vocabulary dictionary 110, and a candidate group extraction unit 111 (corresponding to “candidate group extraction means” in claims) ) Configured with considerable) to "candidate group display means" in the scope of and candidate group display section 112 (claims. Hereinafter, each component of the character input device 1 will be described in detail.

  The imaging unit 101 is specifically a camera and captures still images and moving images. Hereinafter, still images and moving images are collectively referred to as images.

  The image storage unit 102 stores one or more images. Specifically, the image storage unit 102 stores an image captured by the imaging unit 101, an image received from an external device via the communication module 204, and the like. The image storage unit 102 may store an image in advance.

  The acquisition unit 103 acquires an image. Specifically, the acquisition unit 103 acquires an image captured by the imaging unit 101, acquires an image stored by the image storage unit 102, and acquires an image from an external device via the communication module 204. To do.

  The touch panel 104 is an apparatus having both the function of an output device that displays an image or the like on the touch panel and the function of an input device that accepts a touch operation on the touch panel. The touch operation is a contact operation with a user's finger or the like on the touch panel. The touch operation includes a single touch operation and a multi-touch operation. The single touch operation refers to an operation in which the user moves one finger on the touch panel while touching the touch panel. The multi-touch operation refers to an operation in which a user moves each finger on the touch panel while keeping two or more fingers in contact with the touch panel.

  The image display unit 105 displays the image acquired by the acquisition unit 103 on the touch panel 104.

  The area extraction unit 106 extracts area information indicating a partial area of the image displayed by the image display unit 105 based on a touch operation on the touch panel 104, and corresponds to the partial area specified by the extracted area information. The image to be output is output to a character recognition processing unit 107 described later. Here, the area information is, for example, a set of XY coordinate values on a pixel indicating a partial area of the image. The area extraction unit 106 fixes a frame for extracting a partial area, moves an image based on a single touch operation on the touch panel 104, and moves the image based on a multi-touch operation on the touch panel 104. May be enlarged or reduced. In addition, the region extraction unit 106 fixes the image, moves a frame for extracting a partial region based on a single touch operation on the touch panel 104, and moves the frame based on the multi-touch operation on the touch panel 104. May be enlarged or reduced. Details of the operation of the area extracting unit 106 will be described in “Example of Character Input Device 1 According to First Embodiment” described later.

  The character recognition processing unit 107 extracts a character candidate group based on the character recognition processing for the image. Specifically, the character recognition processing unit 107 first inputs an image acquired by the acquisition unit 103 or an image output by the region extraction unit 106. Next, the character recognition processing unit 107 detects a character region that is a region where characters exist in the image. Next, the character recognition processing unit 107 divides the characters in the detected character region in units of characters, and executes character recognition processing for each character on the characters existing in the divided regions. As the character recognition processing executed by the character recognition processing unit 107, an existing character recognition processing technology is applied. The character recognition processing unit 107 may supplementarily use the character recognition dictionary 108 when executing the character recognition processing. As a result of executing the character recognition process, the character recognition processing unit 107 outputs one or more character candidates as a result of the character recognition process for each character, and extracts a character candidate group including one or more character candidates.

  The language processing unit 109 extracts word candidate groups based on the character candidate groups extracted by the character recognition processing unit 107. For example, the language processing unit 109 may extract a word candidate group using a weighted finite state transducer (Weighted Finite State Transducer) expressing a symbol string conversion and a set of weights as state transitions. Specifically, the language processing unit 109 inputs the character candidate group extracted by the character recognition processing unit 107, calculates a weight value for the transition to the character candidate, and a character string based on the combination of the character candidate and the weight value Create transition data. Next, the language processing unit 109 sequentially performs the state transition based on the character string transition data, accumulates the weight value in each state transition, calculates the cumulative weight value for each state transition, and results of one or more state transitions Is output based on the cumulative weight value. Then, the language processing unit 109 extracts a word candidate group including one or more word candidates based on the cumulative weight value. The language processing unit 109 may supplementarily use the vocabulary dictionary 110 when extracting word candidate groups. Note that the language processing unit 109 may further receive a character candidate group from an external device, and extract the word candidate group based on the received character candidate group.

  The candidate group extraction unit 111 extracts a character candidate group or a word candidate group based on character information or word information associated with the image acquired by the acquisition unit 103. For example, the image acquired by the acquisition unit 103 may be registered with characters or words in advance. The user can freely perform editing operations such as addition and deletion of characters and words registered in association with images. The acquisition unit 103 acquires a character or word registered in association with an image, and extracts the acquired character or word and a combination thereof as a character candidate group or a word candidate group.

  The candidate group display unit 112 displays the character candidate group extracted by the character recognition processing unit 107 or the word candidate group extracted by the language processing unit 109 as an input candidate group at the time of character input. In addition, the candidate group display unit 112 selects the character candidate group or the word candidate group extracted by the candidate group extraction unit 111 from the character candidate group extracted by the character recognition processing unit 107 or the word candidate extracted by the language processing unit 109. In addition to the group, it may be further displayed as an input candidate group when inputting characters. Here, the input candidate group at the time of character input is, for example, when performing kana-kanji conversion using IME, a plurality of conversion candidates are displayed, and conversion is performed by the user selecting one of these conversion candidates. Although it is fixed, it indicates the plurality of conversion candidates displayed.

(Operation of the character input device 1 according to the first embodiment)
Next, a character input method that is an operation performed by the character input device 1 according to the first embodiment will be described with reference to FIG. FIG. 3 is a flowchart showing a character input method in the character input device 1.

  First, the acquisition unit 103 acquires an image (step S01, acquisition step). Next, the character recognition processing unit 107 extracts a character candidate group based on the character recognition processing for the image acquired in step S01 (step S02, character recognition processing step). Next, the language processing unit 109 extracts a word candidate group based on the character candidate group extracted in step S02 (step S03, language processing step). Next, the candidate group display unit 112 displays the character candidate group extracted in step S02 or the word candidate group extracted in step S03 as an input candidate group at the time of character input (step S04, candidate group display step).

(Example of the character input device 1 according to the first embodiment)
Next, an example of the character input device 1 according to the first embodiment will be described with reference to FIGS. 4 and 5. FIG. 4 and FIG. 5 are diagrams showing an example of displaying a search site and inputting a search input character using the character input device 1. Here, when inputting an input character for search, a character candidate group and a word candidate group are extracted from the characters included in the image and displayed as an input candidate group, and the word selected by the user from the input candidate group is used for the search. It is input as an input character.

  FIG. 4A is a diagram illustrating a state in which a search site is displayed on the touch panel 104 of the character input device 1. A caret C indicating a character insertion position is displayed in the search text box. In the state shown in FIG. 4A, when the user instructs the character input device 1 to switch the IME, for example, when the user presses a button corresponding to the switching operation of the IME, the user is shown in FIG. Transition from the state to the state shown in FIG.

  FIG. 4B is a diagram illustrating a state in which a menu M1 that can select an acquisition method of an image to be subjected to character recognition processing is displayed at the bottom of the touch panel 104. In the state shown in FIG. 4B, when the user selects either “camera” or “gallery” from the menu M1, the state shown in FIG. 4B transitions to the state shown in FIG. 4C. . In the state shown in FIG. 4B, when the user selects “camera” from the menu M1, the imaging unit 101 is activated, and an image captured by the imaging unit 101 is displayed by the image display unit 105 in FIG. ) In the image display field P. On the other hand, when the user selects “Gallery” from the menu M1 in the state shown in FIG. 4B, a screen prompting the user to select a specific image from the images stored in the image storage unit 102 is displayed (not shown). (Illustrated). Therefore, when the user selects a specific image, the selected image is displayed in the image display field P of FIG.

  In the state shown in FIG. 4C, an image is displayed in the image display column P, and a frame F for extracting a partial region of the image is displayed on the image. Displays a menu M2 for selecting a mode for selecting a partial region of the image. In the present embodiment, the frame F is rectangular, but is not limited thereto. For example, the frame F may be a circle or a polygon.

  A case where the user selects “Fix frame” from the menu M2 in the state shown in FIG. 4C will be described. In this case, the frame F is fixed on the image display field P. When the user performs a single touch operation on the touch panel 104, for example, when performing a single touch operation that moves to the right on the image, the image moves to the right according to the amount of movement. Note that the position of the frame F remains fixed, that is, the absolute position of the frame F on the touch panel 104 does not change before and after the single touch operation. Further, when the user performs a multi-touch operation on the touch panel 104, for example, a multi-touch operation in which two fingers are separated from each other on the image, the image is enlarged according to the amount of movement. Conversely, for example, when a multi-touch operation is performed such that two fingers approach each other on the image, the image is reduced according to the amount of movement. Note that the position and size of the frame F remain fixed when the image is enlarged or reduced.

  Next, a case where the user selects “fix image” from the menu M2 in the state shown in FIG. 4C will be described. In this case, the image is fixed on the image display field P. When the user performs a single touch operation on the touch panel 104, for example, when performing a single touch operation that moves in the right direction on the frame F, the frame F moves in the right direction according to the amount of movement. Note that the position of the image remains fixed, that is, the absolute position of the image on the touch panel 104 does not change before and after the single touch operation. Further, when the user performs a multi-touch operation on the touch panel 104, for example, a multi-touch operation in which two fingers are separated from each other on the frame F, the frame F is enlarged according to the movement amount. Conversely, for example, when a multi-touch operation is performed such that two fingers approach each other on the frame F, the frame F is reduced according to the amount of movement. Even when the frame F is enlarged or reduced, the position and size of the image remain fixed.

  In the state shown in FIG. 4C, when the user selects either “Fix frame” or “Fix image” from the menu M2 and finishes the touch operation, the state shown in FIG. Transition to the state shown in a). In FIG. 5A, an image corresponding to the frame F is processed by the character recognition processing unit 107 and the character recognition dictionary 108, and the extracted character candidate group and word candidate group are input by the candidate group display unit 112. It is a figure which shows the state currently displayed on the input candidate list | wrist L as an input candidate group. As shown in FIG. 5A, a frame F is positioned at “boiled” in the image, and character candidate groups and word candidate groups of “boiled”, “expected”, “boiled” and “boiled” are input candidate lists. L is displayed.

  If the user selects one input candidate displayed in the input candidate list L in the state shown in FIG. 5A, the state transitions to the state shown in FIG. As shown in FIG. 5B, when the user selects “boiled” from the input candidate list L, the selected word “boiled” is inserted at the position of the caret C in the search text box. . At this time, the character input device 1 may register the word selected by the user in association with the image acquired by the acquisition unit 103. Thereby, when the acquisition unit 103 acquires the image next time, the candidate group extraction unit 111 extracts the character candidate group or the word candidate group registered in association with the image, and the extracted character candidate group Alternatively, the word candidate group is displayed by the candidate group display unit 112. Therefore, the user can select a character candidate group or a word candidate group registered in advance in association with an image, and the user's operation can be simplified.

(Operation and effect of the character input device 1 according to the first embodiment)
Then, the effect | action and effect of the character input device 1 which concern on 1st Embodiment are demonstrated.

  According to the character input device 1, the acquisition unit 103 acquires an image, the character recognition processing unit 107 extracts a character candidate group based on the character recognition processing for the acquired image, and the language processing unit 109 extracts A word candidate group is extracted based on the character candidate group. And candidate group display part 112 displays the extracted character candidate group or word candidate group as an input candidate group at the time of character input. Thus, in the character input device 1, a character candidate group or a word candidate group based on the character recognition process for the acquired image is displayed as an input candidate group at the time of character input. For example, the character candidate group includes a plurality of character candidates as a result of character recognition processing, and the word candidate group includes a plurality of word candidates composed of combinations of the plurality of character candidates. By selecting from the character candidate group or the word candidate group, it is possible to input characters efficiently without being affected by the failure of character recognition.

  In addition, when inputting characters, there is no need for complicated input operations such as writing input according to the stroke order, re-translation of kana input and romaji input results. Can be achieved. For example, by preparing an environment for multilingual recognition, it is not only possible to understand the meaning at first glance like a language in another country, but even in situations where it is difficult to input itself, candidates for general character input can be used. Characters can be entered simply by selecting from the group. That is, according to the character input device 1, the user can input characters without prior knowledge only by estimation from the appearance of the characters.

  Further, according to the character input device 1, the candidate group extraction unit 111 extracts a character candidate group or a word candidate group based on character information or word information associated with the image, and the candidate group display unit 112 is extracted. The character candidate group or word candidate group is further displayed. As described above, the character candidate group or the word candidate group based on the character information or the word information associated with the image is further displayed as the input candidate group at the time of character input, so that the character input can be performed efficiently.

  Further, according to the character input device 1, an image captured by the imaging unit 101 can be a target of character recognition processing. For example, if a user can capture an image that includes a character or a word, the user can capture the image and acquire it as an image including the character or word, thereby using the character or word as an input candidate group when inputting characters. Therefore, it is possible to input characters efficiently.

  In addition, according to the character input device 1, the image stored in the image storage unit 102 can be a target of character recognition processing. For example, the user stores an image including characters or words in the image storage unit 102. If it does, since the said character or word can be utilized as an input candidate group at the time of character input, character input can be performed efficiently.

  Further, according to the character input device 1, the image display unit 105 displays an image on the touch panel 104. Then, the area extraction unit 106 extracts area information indicating a partial area of the image based on the touch operation on the touch panel 104, and recognizes an image corresponding to the partial area specified by the extracted area information. The data is output to the processing unit 107. Thereby, for example, the user selects a partial area including a necessary character by a touch operation while viewing an image displayed on the touch panel 104, and selects an image corresponding to the selected partial area as a target of character recognition processing. Therefore, it is possible to input characters efficiently.

  Further, according to the character input device 1, the area extraction unit 106 moves an image based on a single touch operation on the touch panel 104 after fixing a frame for extracting a partial area, and on the touch panel 104. Since the image is enlarged or reduced based on the multi-touch operation, the user can efficiently select a partial region in the image with a simple operation.

  In addition, according to the character input device 1, the area extraction unit 106 moves the frame for extracting a partial area based on the single touch operation on the touch panel 104 after fixing the image, and on the touch panel 104. Since the frame is enlarged or reduced based on the multi-touch operation, the user can efficiently select a partial region in the image with a simple operation.

  Further, according to the character input device 1, the image is analyzed, and characters and character candidates and words and word candidates are generated based on the analysis result, and the user selects and uses them as a plurality of character input candidates. This eliminates the need for complicated input operations such as handwriting input, re-converting Kana input and Romaji input results, improving character input efficiency for specific applications, and means for inputting foreign languages that users do not understand Can be used as That is, according to this embodiment, the user can input characters without prior knowledge only by estimation from the appearance of the characters from the character input system.

(Configuration of the character input system 3A according to the second embodiment)
Next, the configuration of the character input system 3A including the character input device 1A and the server 2A according to the second embodiment of the present invention will be described with reference to FIG. As a second embodiment, the character input device 1A communicates with the server 2A via a network and performs a process related to character input in the character input system 3A. FIG. 6 is a schematic configuration diagram of the character input device 1A and the server 2A. Note that the character input device 1A, the acquisition unit 103A, the character recognition processing unit 107A, the character recognition dictionary 108A, the language processing unit 109A, the vocabulary dictionary 110A, and the candidate group display unit 112A shown in FIG. Have the same configuration or function as the component without the symbol A in FIG. Therefore, in the second embodiment, only functional differences will be described for those components.

  In FIG. 6, the character input device 1A and the server 2A can communicate with each other via a network such as the Internet. The server 2A is a server device such as a PC. Similarly to the character input device 1 in the first embodiment, the server 2A has the hardware configuration shown in FIG. 2, and each function of the server 2A reads predetermined computer software on hardware such as the CPU 201, ROM 202, and RAM 203. Thus, the communication module 204 and the input / output device 205 are operated under the control of the CPU 201, and data is read and written in the ROM 202, the RAM 203, and the auxiliary storage device 206.

  The character input device 1A is functionally equivalent to the acquisition unit 103A, the character recognition processing unit 107A, the character recognition dictionary 108A, the candidate group display unit 112A, and the client transmission unit 120A (corresponding to “client transmission unit” in the claims). ), And a client receiving unit 121A (corresponding to “client receiving means” in the claims).

  The acquisition unit 103A acquires an image. The character recognition processing unit 107A extracts a character candidate group by appropriately using the character recognition dictionary 108A based on the character recognition processing for the image acquired by the acquisition unit 103A. The client transmission unit 120A transmits the character candidate group extracted by the character recognition processing unit 107A to the server 2A. The client reception unit 121A receives a word candidate group from the server 2A. The candidate group display unit 112A displays the character candidate group extracted by the character recognition processing unit 107A or the word candidate group received by the client reception unit 121A as an input candidate group at the time of character input.

  The character input device 1A may include the imaging unit 101, the image storage unit 102, the touch panel 104, the image display unit 105, and the region extraction unit 106 of the character input device 1 according to the first embodiment.

  Subsequently, the server 2A functionally includes a language processing unit 109A, a vocabulary dictionary 110A, a server reception unit 122A (corresponding to “server reception means” in the claims), and a server transmission unit 123A (claims). Equivalent to “server transmission means” in FIG.

  Server reception unit 122A receives a character candidate group from character input device 1A. Based on the character candidate group received by the server reception unit 122A, the language processing unit 109A extracts the word candidate group by appropriately using the vocabulary dictionary 110A. The server transmission unit 123A transmits the word candidate group extracted by the language processing unit 109A to the character input device 1A. Note that the server reception unit 122A may further receive a character candidate group from an external device other than the character input device 1A.

(Operation and effect of the character input system 3A according to the second embodiment)
Next, the operation and effect of the character input system 3A according to the second embodiment will be described.

  In the character input device 1 according to the first embodiment, a word candidate group is extracted by the language processing unit 109 in the character input device 1. On the other hand, in the character input system 3A according to the second embodiment, the word candidate group is extracted by the language processing unit 109A in the server 2A. Thereby, it is not necessary to perform extraction of a word candidate group in the character input device 1A, and the processing load of the character input device 1A is reduced.

  In the character input device 1 according to the first embodiment, the vocabulary dictionary 110 is held in the character input device 1. On the other hand, in the character input system 3A according to the second embodiment, the vocabulary dictionary 110A is held in the server 2A. This eliminates the need to maintain a vocabulary dictionary in the character input device 1A, and reduces the memory capacity required for the character input device 1A. Further, in the character input device 1A, it is not necessary to read a vocabulary dictionary, and the processing speed is improved.

(Configuration of the character input system 3B according to the third embodiment)
Next, the configuration of a character input system 3B including a character input device 1B and a server 2B according to a third embodiment of the present invention will be described with reference to FIG. As a third embodiment, the character input device 1B communicates with the server 2B via a network and performs a process related to character input in the character input system 3B. FIG. 7 is a schematic configuration diagram of the character input device 1B and the server 2B. It should be noted that the character input device 1B, server 2B, acquisition unit 103B, character recognition processing unit 107B, character recognition dictionary 108B, language processing unit 109B, vocabulary dictionary 110B, candidate group display unit 112B, client transmission unit 120B, Each of the client reception unit 121B, the server reception unit 122B, and the server transmission unit 123B has the same configuration or function as the component of the reference numeral A having the same reference number in the second embodiment. Therefore, in the third embodiment, only functional differences will be described for those components.

  The character input device 1B functionally includes an acquisition unit 103B, a candidate group display unit 112B, a client transmission unit 120B, and a client reception unit 121B.

  The acquisition unit 103B acquires an image. The client transmission unit 120B transmits the image acquired by the acquisition unit 103B to the server 2B. The client reception unit 121B receives a character candidate group or a word candidate group from the server 2B. The candidate group display unit 112B displays the character candidate group or the word candidate group received by the client reception unit 121B as an input candidate group at the time of character input.

  The character input device 1B may include the imaging unit 101, the image storage unit 102, the touch panel 104, the image display unit 105, and the region extraction unit 106 of the character input device 1 according to the first embodiment.

  Subsequently, the server 2B functionally includes a character recognition processing unit 107B, a character recognition dictionary 108B, a language processing unit 109B, a vocabulary dictionary 110B, a server reception unit 122B, and a server transmission unit 123B.

  Server receiving unit 122B receives an image from character input device 1B. The character recognition processing unit 107B extracts a character candidate group by appropriately using the character recognition dictionary 108B based on the character recognition processing for the image received by the server reception unit 122B. The language processing unit 109B extracts the word candidate group by appropriately using the vocabulary dictionary 110B based on the character candidate group extracted by the character recognition processing unit 107B. The server transmission unit 123B transmits the character candidate group extracted by the character recognition processing unit 107B or the word candidate group extracted by the language processing unit 109B to the character input device 1B. The server reception unit 122B may further receive an image for character recognition from an external device other than the character input device 1B. The server reception unit 122B further receives a character candidate group from an external device other than the character input device 1B, and the language processing unit 109B uses the vocabulary dictionary 110B based on the character candidate group received by the server reception unit 122B. May be used as appropriate to further extract word candidate groups.

(Operation and effect of the character input system 3B according to the third embodiment)
Then, the effect | action and effect of the character input system 3B which concern on 3rd Embodiment are demonstrated.

  In the character input device 1 according to the first embodiment, extraction of character candidate groups is executed by the character recognition processing unit 107 in the character input device 1, and extraction of word candidate groups is executed by the language processing unit 109. . On the other hand, in the character input system 3B according to the third embodiment, extraction of character candidate groups is executed by the character recognition processing unit 107B in the server 2B, and extraction of word candidate groups is executed by the language processing unit 109B. Yes. Thereby, it is not necessary to perform extraction of the character candidate group and the word candidate group in the character input device 1B, and the processing load of the character input device 1A is reduced.

  In the character input device 1 according to the first embodiment, the character recognition dictionary 108 and the vocabulary dictionary 110 are held in the character input device 1. On the other hand, in the character input system 3B according to the third embodiment, a character recognition dictionary 108B and a vocabulary dictionary 110B are held in the server 2B. Thereby, it is not necessary to hold the character recognition dictionary and the vocabulary dictionary in the character input device 1B, and the memory capacity required for the character input device 1B is reduced. Further, in the character input device 1B, it is not necessary to read the character recognition dictionary and the vocabulary dictionary, and the processing speed is improved.

  DESCRIPTION OF SYMBOLS 1, 1A, 1B ... Character input device 2, 2A, 2B ... Server, 3, 3A, 3B ... Character input system, 101 ... Imaging part, 102 ... Image storage part, 103, 103A, 103B ... Acquisition part, 104 ... Touch panel, 105 ... Image display unit, 106 ... Area extraction unit, 107, 107A, 107B ... Character recognition processing unit, 108, 108A, 108B ... Character recognition dictionary, 109, 109A, 109B ... Language processing unit, 110, 110A, 110B: Vocabulary dictionary, 111: Candidate group extraction unit, 112, 112A, 112B ... Candidate group display unit, 120, 120A, 120B ... Client transmission unit, 121, 121A, 121B ... Client reception unit, 122, 122A, 122B ... Server Receiving unit, 123, 123A, 123B... Server transmitting unit.

Claims (10)

An acquisition means for acquiring an image;
Character recognition processing means for extracting a character candidate group based on the character recognition processing for the image;
Language processing means for extracting a word candidate group based on the character candidate group;
A candidate group display means for displaying the character candidate group or the word candidate group as an input candidate group at the time of character input;
A character input device comprising: Further comprising candidate group extraction means for extracting a character candidate group or a word candidate group based on character information or word information associated with the image acquired by the acquisition means;
The candidate group display means further displays the character candidate group or the word candidate group extracted by the candidate group extraction means.
The character input device according to claim 1. Further comprising an imaging means,
The acquisition unit acquires an image captured by the imaging unit.
The character input device according to claim 1, wherein the character input device is a character input device. It further comprises image storage means for storing an image,
The acquisition unit acquires the image stored by the image storage unit;
The character input device according to claim 1, wherein the character input device is a character input device. A touch panel;
Image display means for displaying the image acquired by the acquisition means on the touch panel;
Area extracting means for extracting area information indicating a partial area of the image displayed by the image display means based on a touch operation on the touch panel;
Further comprising
The area extraction means outputs an image corresponding to the partial area specified by the extracted area information to the character recognition processing means.
The character input device according to claim 1, wherein the character input device is a character input device. The area extracting means moves the image based on a single touch operation on the touch panel after fixing a frame for extracting the partial area, and based on the multi touch operation on the touch panel. Enlarge or reduce the image,
The character input device according to claim 5. The region extracting means moves the frame for extracting the partial region based on a single touch operation on the touch panel after fixing the image, and based on the multi touch operation on the touch panel. Enlarge or reduce the frame,
The character input device according to claim 5. A character input system including a character input device and a server that can communicate with each other via a network,
The character input device includes:
An acquisition means for acquiring an image;
Character recognition processing means for extracting a character candidate group based on the character recognition processing for the image;
Client transmission means for transmitting the character candidate group to the server;
Client receiving means for receiving word candidate groups from the server;
A candidate group display means for displaying the character candidate group or the word candidate group as an input candidate group at the time of character input;
With
The server
Server receiving means for receiving the character candidate group from the character input device;
Language processing means for extracting the word candidate group based on the character candidate group received by the server receiving means;
Server transmission means for transmitting the word candidate group extracted by the language processing means to the character input device;
A character input system comprising: A character input system including a character input device and a server that can communicate with each other via a network,
The character input device includes:
An acquisition means for acquiring an image;
Client transmission means for transmitting the image to the server;
Client receiving means for receiving character candidate groups or word candidate groups from the server;
A candidate group display means for displaying the character candidate group or the word candidate group as an input candidate group at the time of character input;
With
The server
Server receiving means for receiving the image from the character input device;
Character recognition processing means for extracting the character candidate group based on character recognition processing for the image received by the server receiving means;
Language processing means for extracting the word candidate group based on the character candidate group extracted by the character recognition processing means;
Server transmitting means for transmitting the character candidate group extracted by the character recognition processing means or the word candidate group extracted by the language processing means to the character input device;
A character input system comprising: A character input method in a character input device,
An acquisition step in which the acquisition means acquires an image;
A character recognition processing means for extracting a character candidate group based on the character recognition processing for the image;
A language processing step for extracting a word candidate group based on the character candidate group;
Candidate group display means, the candidate group display step of displaying the character candidate group or the word candidate group as an input candidate group at the time of character input,
A character input method comprising:

Images