JPH04340681A - Character recognition system - Google Patents

Abstract

PURPOSE:To easily execute the set of a recognizing area, the instruction of recognition start and the correction of uncertain characters or the like entirely by a pen touch in character recognition and to considerably improve operability. CONSTITUTION:This system is composed of a main body part A equipped with an image input part 1 to input image information and a character recognition part 2 to execute the character recognition, and a portable part B which can exchange data with this main body part A and is equipped with an online character recognition part 4, and the image information inputted at the main body part A and the recognized result are transmitted to the portable part B. The portable part B integrates a display part 5 and a position input part (tablet) 7 and enables handwriting pen input, and the image information transmitted from the main body part A and the recognized result are displayed at the display part 5. Then, various instructions and operations related to the character recognition can be executed by operating a pen 6 on the tablet 7.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a character recognition system for recognizing, for example, printed characters, particularly characters printed as text.

0002

2. Description of the Related Art Conventionally, methods for recognizing characters have mainly been based on handwritten characters. In particular, by quickly recognizing characters written on slips, it was suitable for inputting the contents of large amounts of slips into computers and the like. Initially, the characters targeted for recognition were mainly alphanumeric characters or kana. but,
Since it became necessary to input information such as addresses, it became possible to recognize kanji.

[0003] Recently, text reader recognition systems that recognize printed characters have been put into practical use in order to streamline office operations. This text reader recognition system recognizes various printed materials that are not in the form of text files. Generally, this system reads the characters of the target printed material as image information using an image scanner, then recognizes them, stores them in the form of a text file, and then uses a word processor in combination to process the text information appropriately. Suitable for creating new documents by adding or deleting.

[0004] The conventional text reader recognition system will be explained in detail below. FIG. 22 schematically shows the configuration of the system, which includes a CRT display section 101, an image scanner 102, a character recognition section 103, a keyboard 104, a mouse 105, and a control section 106 that controls these.

In many cases, general printed matter is printed with image information such as figures and photographs, so in order to recognize the characters in the desired character area, the entire page is first scanned as image information using the image scanner 102. After reading, use the mouse 105 to specify the character area to be recognized.

FIG. 23 is a diagram showing an example of a display screen of the display unit 101 of text read by the image scanner 102. In the figure, 201 and 203 are areas in which characters are printed, 201 is the title of this text, and 203 corresponds to the main text. 205 is an area where a photograph is printed.

[0007] Then, after clicking the area designation icon 206 with the mouse 105, the area 202 is designated. Similarly, after clicking icon 206, area 2
Specify 04. Then, when the recognition icon 207 is clicked, character recognition processing is started. Character recognition processing involves a large amount of processing and requires a high-speed arithmetic unit, so an independent character recognition unit 103 is provided, which performs character recognition processing.

[0008] When the character recognition process is completed, the display unit 101
changes to the display screen shown in FIG. 24. In the figure, 209 and 21
2 is the recognition result, and 208 and 211 are designated character areas. Further, characters 210, 213, and 214 indicate that the recognition results are uncertain and are displayed in reverse. Then, in order to confirm an uncertain character, first move the point of the mouse 105 to the position of the character 210 and click, and the first candidate character in the recognition process will be displayed in the area 216.

Further, in area 218, the original image of character 210 in FIG. 19 is partially displayed, and can be used as a reference by the operator when selecting a recognition candidate. If the character 216 is correct, clicking the icon 215 with the mouse 105 confirms that character as the character displayed at 216.

If the character displayed at 216 is incorrect, if the character 210 is further clicked with the mouse 105, the second candidate character is displayed at 216. If this character is correct, move the icon 215 to the mouse 10
Clicking on 5 confirms this character. Also,
If you are wrong, move the mouse 1 over the character 210 again.
05 to display the third candidate in 216. If this character is correct, clicking the icon 215 with the mouse 105 confirms this character.

[0011] In this system, up to the third candidate for recognition can be selected, so if the third candidate is incorrect, it is necessary to input the correct character manually. Therefore, this time, the correction icon 217 is clicked using the mouse 105. And keyboard 10
From 4 onwards, enter the reading of this kanji in kana. 219 displays the result of inputting a kana.

Next, by pressing the space key or a dedicated conversion key on the keyboard 104, the kana is converted to a kanji as shown in 220, and a plurality of homonym kanji are displayed. Regarding the correct kanji in this display 220,
The kanji can be selected by inputting the number in front of the kanji from the keyboard 104.

[0013] Similarly, character 2 for which the recognition result is uncertain
13 and 214 as well, the correct characters can be input from the keyboard 104. In addition, if the recognition result is not uncertain and is a completely incorrect character, click the correction icon 217 with the mouse 105, move the point of the mouse 105 to the relevant character, click it, and then do the same as above. Kanji can be input by inputting kana from the keyboard 104, converting the kana to kanji, and selecting the correct kanji from the display 220. below,
Similarly, characters 213 and 214 whose recognition results are uncertain can also be corrected or corrected in the same manner as described above.

[0014]

[Problems to be Solved by the Invention] As described above, in the conventional text reader recognition system, processing related to recognition, that is, setting of a recognition area, instruction to start recognition, correction of uncertain characters, and erroneous recognition. All corrections to the written characters were performed using the mouse and keyboard. Therefore, the operation was extremely complicated and inefficient.

Therefore, the present invention enables instructions and operations related to recognition, such as setting a recognition area, instructing to start recognition, and correcting uncertain characters, to be easily performed with a pen touch, thereby significantly improving operability. The purpose is to provide a character recognition method for

[0016]

[Means for Solving the Problems] The character recognition method of the present invention provides a first electronic device having means for inputting image information including at least characters, and a mobile phone capable of exchanging data between the first electronic device and the first electronic device. a second electronic device of the shape;
At least one of the electronic device and the second electronic device has a character recognition means, the second electronic device has a handwriting input means in which a display unit and a tablet are integrated and can input handwritten pen input, and the first electronic device has a character recognition means. When the character recognition means performs character recognition from the image information inputted by the device, before that, an area of the image where the characters exist is designated by the handwriting input means of the second electronic device, and after recognition processing, the recognition result is recognized. The present invention is characterized in that it is possible to select uncertain character candidates and correct erroneously recognized characters using the handwriting input means of the second electronic device.

[0017]

[Operation] The present invention provides, for example, a first electronic device having a means for inputting image information and a character recognition means for performing character recognition, and a portable device capable of exchanging data between the first electronic device and the first electronic device. a second electronic device;
The image information inputted by the electronic device and the recognition result thereof are sent to the second electronic device. The second electronic device has a handwriting input means in which the display unit and the tablet are integrated and can perform handwritten pen input, and displays the image information sent from the first electronic device and the recognition result on the display unit. do,
Instructions and operations related to character recognition can be performed using a pen on the tablet. As a result, instructions and operations related to character recognition on the second electronic device side, such as setting a recognition area, instructions to start recognition, and correcting uncertain characters, can be easily performed with a pen touch. The operability is significantly improved compared to the conventional mouse and keyboard operations.

[0018]

DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 shows a schematic configuration of, for example, a personal complex OA (office automation) device to which the character recognition method of the present invention is applied. Portable unit B as a second electronic device
The main unit A and the portable unit B can exchange data. The main body section A is composed of an image input section 1 for inputting various image information, a character recognition section 2 for performing character recognition, and a control section 3 for controlling these. The mobile unit B also includes an online character recognition unit 4 that performs online character recognition, a display unit 5 that displays image information, etc., and is integrated on the display unit 5, and is used to input information by hand using a pen 6. It is composed of a position input section (transparent tablet) 7 for controlling the camera, and a control section 8 for controlling these operations.

[0020] The personal complex OA equipment with such a configuration is
For example, when using on a desk, main unit A or mobile unit B may be used independently, or main unit A and mobile unit B may be used, for example, mobile unit B may be used as an input device for main unit A. It is also possible to use it in an integrated manner. Further, when going out, only the portable part B can be removed and carried around. Next, the specific configuration of this personal complex OA device will be explained in detail.

First, the main body A is constructed as shown in FIG. That is, the position input unit 11 is for inputting specified position coordinate information, and is composed of, for example, a mouse. The keyboard section 12 is used to input numbers, character/symbol codes, control codes, etc., and includes a plurality of keys. The image input unit 13 (corresponding to the image input unit 1 in FIG. 1) inputs various image information, such as an image scanner or electronic camera that inputs still image data, or a video tape recorder that inputs video data. Consists of etc.

The control unit 14 (corresponding to the control unit 3 in FIG. 1)
It controls the operation of each part of the personal complex OA device, and is mainly composed of, for example, a microprocessor. The bus controller 15 controls information transfer between each section via the bus 16. Bus 16 is
It is used to transfer necessary information between each section of the main body section A.

[0023] The encryption processing section 17 encrypts various information. The image processing unit 18 performs image processing such as composition, recognition, compression/expansion, etc. of image information. The memory unit 19 is a general-purpose memory used when performing a series of processes such as image processing and audio processing. The text data processing unit 20 performs various editing processes on text data.

The database section 21 classifies various information into a plurality of classes and stores them as a database. This database not only contains code information, but also
It is constructed from image information, audio information, etc. The media conversion unit 22 (corresponding to the character recognition unit 2 in FIG. 1) performs processing such as character recognition from the image information input by the image input unit 13 and converting it into a character code, or converts the created document. It performs processing such as converting it into image information and making it into a data format that can be sent by facsimile.

[0025] The audio input unit 23 is for inputting audio information, and is composed of, for example, a microphone. The audio processing unit 24 performs audio processing such as audio recognition, synthesis, and compression/expansion. The timer unit 25 provides time information and time information to the personal complex OA device. The main body interface (I/F) 26 is an interface with the mobile unit B on the main body A side, and is, for example, a wireless interface using infrared rays or radio.

The display section 27 is a display device on the side of the main body section A, and is used to display input image information, display created documents and figures, display operation menus, etc. Consists of etc. The audio output unit 28 outputs audio information,
For example, it consists of a speaker, etc. Image output section 29
is for printing out various image information.
Examples include laser printers and color printers. The communication unit 30 communicates information between the personal complex OA device and other telephones, facsimile machines, personal complex OA devices, terminal devices, etc. via communication lines such as telephone lines or facsimile lines. .

Next, the portable unit B is constructed as shown in FIG. That is, the mobile unit interface (I/F) 31
is an interface with the main body part A on the side of the portable part B, and is, for example, a wireless interface using infrared rays or radio. Control unit 32 (corresponds to control unit 8 in FIG. 1)
controls the operation of each part of the mobile unit B when the mobile unit B is used alone or when the main unit A and the mobile unit B are connected. Constructed as.

The position input unit 33 (corresponding to the position input unit 7 in FIG. 1) is for inputting specified position coordinate information using the portable unit B, and is configured by, for example, a transparent tablet or the like, and is configured with a pen or the like. 34 (corresponding to the pen 6 in FIG. 1) to input figures and characters by hand on the tablet.
placed on top. Note that position coordinate information can also be input by touching the pen 34 on the tablet instead of using the mouse (hereinafter referred to as pointing).

The online character recognition unit 35 (corresponding to the online character recognition unit 4 in FIG. 1) is for inputting characters using the portable unit B. The image input unit 36 inputs various image information such as still images, and is configured by, for example, an image scanner or an electronic camera. The audio input unit 37 is for inputting audio information, and is composed of, for example, a microphone. The bus controller 38 controls information transfer between each unit via the bus 39. The bus 39 is used to transfer necessary information between each section of the portable section B.

The display unit 40 (corresponding to the display unit 5 in FIG. 1) displays characters, figures, images, etc., and is composed of, for example, a liquid crystal display, and the position input unit 33 is placed on the display screen. Ru. The communication section 41 is a mobile section B.
and other telephones, facsimile devices, or via communication lines such as telephone lines or facsimile lines.
It performs information communication with personal complex OA equipment, terminal devices, etc.

[0031] The audio output unit 42 outputs audio information and is composed of, for example, a speaker. A memory card is inserted into the memory card section 43, and necessary information is input/output between the memory card and the memory card. The memory unit 44 is a memory for various processing and data storage. The text data processing unit 45 performs various editing processes on text data.

The personal complex OA device according to this embodiment basically consists of a main body part A and a portable part B, which are equipped with the above-mentioned parts, and effectively utilizes the functions of each of the above-mentioned parts. , exhibiting overall intelligent functionality. In addition, the image input section, position input section, voice input section, display section,
Each input/output section, such as the audio output section, does not necessarily have to be provided in main unit A and mobile unit B, but can be provided only in main unit A or only in mobile unit B, and can be connected to each bus and main unit interface. There is no problem even if the information is transmitted to the necessary section for use via the mobile interface. Furthermore, some characteristic components will be described in detail.

The encryption processing unit 17 encrypts the given communication original text according to the encryption key to generate the encrypted communication, or conversely decrypts the given encrypted communication according to the encryption key. , and execute the process to find the original text. Such encryption processing is described, for example, in Japanese Unexamined Patent Publication No. 63
This is realized by the technology described in No.-85931.

The image processing section 18 includes, for example, an image generation section 51 and an image composition section 52, as shown in FIG.
, and an image compression/expansion section 53. The image generation unit 51 generates a given vector or polygon,
The image synthesis unit 52 generates a line figure indicated by a parameter such as a circular arc, and also performs filling processing, special processing such as clipping processing, and synthesis processing with a natural image. Such image generation processing and image composition processing are described in, for example, Japanese Patent Application Laid-Open No. 63-85
This is realized by techniques such as those described in No. 931. In addition, in FIG. 4, 16a is a general-purpose bus, 167
b is an image bus.

In addition, in FIG. 4, the image input section 13 on the main body section A side is used for image input, but the image input section 36 on the portable section B side is used for image input, and the image input section 36 on the portable section B side is used for image input. The image information is transmitted to the main body A via the bus 39, the mobile interface 31, and the main body interface 26, or the input/output units on the main body A side and the mobile unit B side are used in combination. You may also do so.

The database unit 21 organizes and stores various information such as code information, image information, and audio information.
This is used for various application systems. FIG. 6 shows a schematic configuration of this database section 21, in which an interface section 61 executes command analysis processing, etc.
, a data operation unit 62 that executes search processing of accumulated information (database), etc., a magnetic disk device 63 and an optical disk device 6 as storage media for storing various data.
4 and its additional function section 65. Various data are classified and organized into a plurality of classes according to the type of data, and are registered for each class. The database unit 21 will be explained below in four parts: its logical structure, stored data, physical structure, and additional functions.

The logical structure indicates how various data are stored when the database section 21 is viewed from the application system side. Here, data is handled in a logical structure according to an object-oriented model, such as the structure shown in FIG. 7, for example. The class has several fields (
Attributes) are provided, and predetermined data is stored in each of these columns. A data unit (object) can itself have a complex structure. One class is constructed by an arbitrary number of attributes storing such objects.

In this model, data is stored in the database by specifying a class name and providing values for each of its attributes. Also, when searching a database, specify a class and attribute, and check whether the object stored there satisfies a predetermined condition with the specified object or with an object stored in another attribute. This is done by determining the condition and extracting objects that meet the condition.

This search condition is given as whether the objects are equal, unequal, smaller, larger, etc. At this time, specify search conditions for each of multiple attributes, and apply logical processing (A
(ND, OR, etc.) is also possible. Furthermore, it is also possible to search a database by specifying multiple classes and searching for a given object from among multiple classes based on conditions such as an object with a certain attribute in one class being equal to an object with a certain attribute in another class. It is.

Furthermore, data is deleted from the database basically in the same manner as the above search, but instead of extracting the object, it is performed by deleting the object. Furthermore, data updating is similar, and is performed by changing the value of a specified attribute of the obtained class and storing it.

Furthermore, in each class, information (person name, person in charge code, etc.) who is permitted to read, add, or change data for each attribute is entered, and data protection measures are taken. Note that instead of taking this data protection measure for each attribute, it is also possible to take it for each class. Moreover, the information on a plurality of people may be listed here.

In the class example shown in FIG. 7, the data is shown as a character string, but the data stored in each class may also be a simple bit string. In other words, the data stored in the class is of course strings,
It may be image information or audio information. Furthermore, this data may have a structure.

Now, the data stored in this database section 21 is the "personal schedule" shown in FIG. 7 mentioned above.
For example, starting from the class of FIG. 8(a) to (g)
``Address book,'' ``Individual work and its agents,'' and
Operation history”, “Human resources”, “Conference room”, “Conference room reservation”, “Meeting”
It consists of various classes such as.

As shown in this example, classes consist of classes that are mainly used for personal use and classes that are commonly used by many users. A personal class is provided for each device of this embodiment used by each individual, and a common class is provided for each device of this embodiment that is common to a plurality of users.

Note that the term "common device of this embodiment" does not necessarily mean that its hardware is different from other devices of this embodiment. It goes without saying that the personal device of this embodiment may also serve as a common device of this embodiment. Furthermore, the number of common devices of this embodiment is not limited to one, and a plurality of devices may be provided depending on the hierarchical level of the system.

The data structure of the "Personal Schedule" class shown in FIG. 7 will now be briefly explained. This class indicates that the class name is "Personal Schedule" and that it was created by "△△△△". This class creator "△△△△" is permitted to perform all data operations on this class.

[0047] Furthermore, according to the data protection function added to this class, everyone is permitted to read data, and only "○○○○" and "persons belonging to the engineering department" can add data.
allowed only for Note that this "person belonging to the engineering department" is found by, for example, referring to the "Human Resources" class. In addition, data changes can be made at the “person level”
Only those with a value of ``5'' or higher are allowed. This "person level" relates to personnel classes, and represents positions, for example, (manager: 8) (deputy manager: 7) (section manager: 6) (supervisor: 5).

[0048] Furthermore, this class includes "start time" and "
Attributes such as end time, type, name, and location are set, and data is written to each of them.

[0049]Additional functions include automatic disposal of unnecessary data, for example. This automatic discarding of unnecessary data is performed by providing additional information for the class such as [discarding: allowed/disabled] and [discarding method], and by operating an erase command for each class at predetermined intervals.

Note that, for example, regarding conference information, an object can be deleted by determining whether or not the end time is earlier than the current time. Therefore, there is no need to add any special functionality to erase such objects.

[0051] Another important function of the additional functions is data security. This data protection function prevents data from becoming invalid due to, for example, hardware failure or power outage. Specifically, this data preservation function is achieved by duplicating information, writing it to magnetic tape, etc. In this way, the database unit 21 sorts and organizes various data by class,
It is used for various application systems.

The media conversion unit 22 is activated upon receiving a media conversion request signal output from the control unit 14, and determines, for example, the medium of the source information to be transmitted and the medium of the transmission destination, and determines the medium of the transmission destination. Apply media conversion to the source information that is compatible with . For example, A.D.
When transmitting information given by PCM voice to a destination device that is a G3 standard facsimile device, it is determined that the source information medium is voice information and the destination medium is image information, and the voice information is transmitted. By converting the media in the order of → character code → image information, it becomes possible to send the information to the other party.

The audio input unit 23 converts audio into an electrical signal using, for example, a microphone, and limits the audio to signals in a frequency band of 30 to 3,400 Hz using, for example, a bandpass filter, and converts the audio into an electrical signal using, for example, a 12-bit sampling frequency of 12 KHz. quantize into a digital signal. The audio information obtained in this way is sent to the audio processing section 24 or the like.

The audio processing section 24 performs various processing on audio information, for example, as shown in FIG.
It is composed of a speech recognition section 54, a speech discrimination section 55, a speech synthesis section 56, and a speech compression/expansion section 57, and each component will be explained below. The voice recognition unit 54 recognizes the voice information obtained from the voice bus 16c dedicated to voice information, and outputs a character code to the general-purpose bus 16a as the recognition result. The voice determination unit 55 determines whether the input voice information is by a specific speaker. The speech synthesis unit 56 synthesizes speech information according to data such as documents and outputs the synthesized speech information to the speech bus 16c. Such speech recognition, discrimination, and synthesis processing is described, for example, in Japanese Unexamined Patent Publication No. 63
This is realized by the technology described in No.-85931.

The audio compression/expansion unit 57 compresses and encodes audio information and stores it in the database unit 21, or conversely restores and expands compressed audio information. Such audio information compression/expansion processing is described, for example, in "Computer Audio Processing" (written by Takeshi Yasuin et al., Akiba Publishing, P30 P4).
This is realized by the technology described in 2).

Each component of the audio processing section 24 is controlled by the control section 14 via the general-purpose bus 16a. In addition, in FIG. 5, the audio input section 23 on the main body A side is used for inputting audio, and the audio output section 2 on the main body A side is used for outputting audio.
8, the audio input unit 37 on the mobile unit B side is used for audio input, and the audio output unit 42 on the mobile unit B side is used for audio output.
is used to transmit audio information to the main unit A via the bus 39 on the mobile unit B side, the mobile unit interface 31, and the main unit interface 26, or,
The input/output sections on the main body part A side and the portable part B side may be used in combination.

The audio output unit 28 converts the digital audio information obtained from the audio bus 16c in FIG. 5 into an analog electrical signal, and outputs it as audio from a speaker, for example.

[0058] The online character recognition unit 35, for example,
Recognizes character information input online via a tablet, etc., and outputs a character code as the recognition result. Such online character recognition is realized, for example, by the technique described in Japanese Patent Application Laid-open No. 85931/1983. Next, some characteristic functions that can be realized in this embodiment in such a configuration will be described.

First, the audio information recording function will be explained. The voice obtained from the telephone function of the voice input section 23 and the communication section 30 can be compressed by the voice compression/expansion section 57 and recorded as voice information in the database section 21 via the voice bus 16c.

[0060] When inputting voice or making a call using the telephone function, for example, a handset/receiver 71 as shown in FIG. 9 is used. The handset 71 includes a receiver section (speaker,
72 (corresponding to the audio output section 28) and a speaking section 73 (corresponding to the microphone, the audio input section 23), for example,
The usage status is detected by a hook switch like a normal telephone. If it is in use, a "recording" menu is displayed on the menu screen of the display section 27, for example. Recording is started by touching the "Record" menu. If you want to stop recording midway through, for example, touch the "Stop" menu. Furthermore, when the handset 71 is put down, the hook switch detects the state and automatically ends recording.

[0061] In this case, if the user touches the "Record" menu after determining that there is a need for recording during a call using the telephone function, a problem occurs in which the beginning of the conversation is cut off. Therefore, when the communication unit 30 detects that a call is in progress, the memory unit 19 is used as a temporary recording area, and the voice is always temporarily recorded for a certain period of time. When the "Record" menu is touched, the audio information temporarily stored in the memory section 19 is prefixed, and subsequent audio information is continuously recorded in the database section 21. Through such processing, it is possible to record audio information from a certain period of time before the "recording" menu was touched.

[0062] Recorded audio information can be given a title or keyword to facilitate later retrieval. When recording is stopped, for example, the display section 27
A menu screen as shown in FIG. 10 is displayed on the display screen using a multi-window display. After selecting a title or keyword from the menu, the user inputs the title or keyword using the keyboard section 12. The input title and keyword are added as an audio information header as shown in FIG. 11. The audio information header stores information such as a title, keywords, link information, recording date and time, and recording time as additional information to the audio information.

Furthermore, by touching the "Link" menu, recorded voice information can be linked as a voice memo to media information such as documents and images, or address and schedule information, as shown in FIG. Can be done. Information for connecting links is selected, for example, from a menu. As shown in FIG. 13, if an open document or image information on the current display screen is directly touched, the information will be linked. These link information are also stored in the header as shown in FIG. 11, for example.

Search information can be set in the header by the above menu operations, but if no menu is specified within a preset period of time or if another operation is started, the header will be blank for the time being. Store the code + serial number in the header and record it. After that,
Using the blank code, these temporarily stored audio information can be searched all at once, and titles, keywords, and link information can be added by the same operation.

The recorded audio information is retrieved from the database section 21 by menu selection and keyword search, subjected to decompression processing in the audio compression/expansion section 57, and outputted from the audio output section 28.

Furthermore, if link information exists, voice information can be directly called up from media information such as documents and images, and data such as addresses and schedules. For example, when voice information is linked as a voice memo to a document on the display screen of the display unit 27, an icon 74 as shown in FIG. 14 is displayed on the display screen of the display unit 27. By touching, the linked audio information is output.

Next, the telephone function will be explained. The telephone function is realized using the voice input section 23, the voice output section 28, and the communication section 30. The microphone for voice input and the speaker for output take the form of a telephone handset 71, for example, as shown in FIG. It also functions as a general-purpose audio input/output device, such as inputting audio for audio and outputting synthesized audio. A speaker for audio output is provided in addition to the handset 71, and is used when the handset 71 is in an on-hook state.

[0068] Naturally, control of telephone functions is also performed by the control section 14. The control section 14 has time division control and a control section dedicated to the telephone function, so that the telephone function can be used even when another function such as a word processor is being used. The operation menu for the telephone function is displayed in a multi-window on the display screen of the display section 27 by touching the "Telephone" menu or by detecting with a hook switch that the handset 71 is disconnected, for example. The screen is displayed as shown in FIG. Main operations are executed by touching the operation menu.

[0069] Dial is performed by the following operation. If you want to dial the other party's phone number as is, like a normal phone call, touch the number menu or
The telephone number is keyed in using the ten keys on the keyboard section 12. In the keyboard section 12, the numeric keypad corresponds to a dial number when the operation menu of the telephone function is opened. The telephone number is dialed and the communication department 3
0 starts making a call, the communication section 30 detects this state and releases the keyboard section 12 from the telephone function, making it available for other functions such as a word processor.

[0070] Also, if you touch the alphabetical order menu,
A Japanese syllabary search is performed on the address book information stored in the database section 21. For example, if you touch "A", the page in the A row will open, and you can visually search for the necessary phone number using the "Next page"/"Previous page" menu. When the required page opens, touch the phone number of the person you want to call. The address book information on the display screen of the display unit 27, like the menu, corresponds to the touched coordinates and the information, and the control unit 14 stores the corresponding telephone number in the database unit 21.
, and perform a dial operation on the communication section 30. Of course, the address book information in the database section 21 can also be directly searched by inputting the name of the other party using the keyboard section 12 or online character recognition. Through such operations, the retrieved address book information can be checked on the display screen of the display unit 27, and then touched and dialed in the same manner.

[0071] Redialing is performed by the following operation. History information of previously called telephone numbers is stored in the memory section 19 for a certain period of time, and when the "redial" menu is touched, it is displayed all at once on the display section 27. If the telephone number is a telephone number that exists in the address book information in the database section 21, the name etc. are also referenced from the information in the address book information by searching by telephone number and displayed accordingly. The screens displayed all at once become a menu screen. That is, when a user touches a telephone number to redial, the control section 14 causes the communication section 30 to redial the telephone number. Note that the answering machine function, the message board function, and the like are realized by reading out the response message by the voice synthesis unit 56 and the voice information recording function described above.

Now, when the handset 71 is put down, the hook switch detects the state and terminates the telephone function. When finished, the telephone function operation menu is erased from the display screen and the original function is restored. Of course, operations performed on the original function while using the telephone function remain valid.

Next, the facsimile function will be explained. The facsimile function is realized using the image input section 13, the image output section 29, the image compression/expansion section 53, the media conversion section 22, and the communication section 30. Naturally, the facsimile function is also controlled by the control section 14. The control unit 14 has a control unit dedicated to time division control and facsimile functions, so that, for example,
It is now possible to use the facsimile function even when using another function such as a word processor. By touching the "FAX" menu, the operation menu for the facsimile function is displayed in a multi-window display on the display screen of the display unit 27, for example, as shown in FIG. 16. By touching the operation menu, The main operations are performed.

The receiving operation of the facsimile function is performed as follows. When the communication section 30 detects reception of a facsimile, the control section 14 stores the received facsimile information in the image area of the database section 21 for the time being. The facsimile information header stores information such as the date and time of reception, as shown in FIG. At the same time, the control section 14 displays, for example, "Receiving fax" in a specific area on the display screen of the display section 27 so as not to interfere with the operation currently being performed. The above receiving operation is automatically performed under the control of the control section 14.

The facsimile information stored in the database section 21 is decompressed in the image compression/decompression section 53 as necessary. For example, by touching the "Display" menu, the facsimile information can be displayed on the display section 27, ” menu, the image output unit 29 prints out the image.

The transmission operation of the facsimile function is performed as follows. Image information input from the image input section 13 by touching the "send" menu is compressed by the image compression/expansion section 53 if necessary, and then transmitted by the communication section 30. The compressed image information is also temporarily stored in the memory unit 19 and used for retransmission and confirmation. In addition, the document information created by the document creation function described later is transferred to the media conversion unit 22.
It is also possible to convert the image information into image information and transmit it by the communication unit 30.

Dial or redial of a facsimile destination number can be performed in the same way as a telephone function. In other words, you can dial directly using the number menu or numeric keypad,
One-touch dialing can be performed from the address book information in the database section 21 or from the history of previously called facsimile destination numbers. Note that the details are described in the explanation of the telephone function.

Next, the document creation function will be explained. For example, as shown in FIG. 18, the document creation function includes an input device 81, an editing device 82, a storage device 83, a display device 84,
This is realized by the control device 85.

The input device 81 includes a keyboard section 12, position input sections 11 and 33, image input sections 13 and 36, and voice input sections 23 and 37. The editing device 82 is the text data processing section 20, 45. The storage device 83 includes the memory section 19, the database section 21, a memory card of the memory card section 43, and the like. The display device 84 is the display units 27 and 40. The editing device 82 is the input device 81
It performs applied processing functions according to instructions from
For example, editing a document using a word processor, editing a drawing using coordinate data, editing a drawing using data manipulation on an image basis, etc. The control device 85 includes the control section 14, the control section 32, etc., and includes an input device 81, an editing device 82, etc.
, the storage device 83, and the display device 84.

[0080] A title and keywords can be added to the created file. The case of adding a title will be described below, but the same procedure can be followed when adding a keyword. For example, when a title is selected from the menu items displayed on the display screen of the display device 84, a menu screen as shown in FIG. 19 is displayed.

[0081] At this time, if an arbitrary area of the file has been specified using the mouse or the like before selecting a title from the menu item, the character code will be added to that area when the menu is selected. If a character code exists, the data is stored in the character buffer. If image information exists, it is converted into a character code by character recognition and then stored in the character buffer. Enter that data. If none of them exist, nothing is input to the character buffer. When displaying the menu screen as shown in FIG. 19, character buffer information is also displayed in the title section 91. Next, for example, using the keyboard section 12 or the online character recognition section 35, a title is input or changed for the information in the character buffer.

Finally, by selecting the determination command 92 on the menu screen of FIG. 19, the title in the character buffer is added as the header of the file information. If the cancel command 93 is selected, the above operation is immediately canceled.

Next, a character recognition method according to the present invention will be explained. The printed material is input as image information in the image input section 13 (mainly using an image scanner). After this image information is stored in the memory section 19, the same image information is sent as is to the portable section B via the main body section interface 26 and the portable section interface 31. The control unit 32 of the mobile unit B adds operation icons to this image information and displays a screen as shown in FIG. 20 on the display unit 40. In FIG. 20, 301 and 303 are character information, and 305 is image information.

Then, the area of characters to be recognized is handwritten using the pen 34. In FIG. 20, first, point the area designation icon 306 with the pen 34 (
(touching the icon 306 with the pen 34), and then specifying the area 302 by handwriting with the pen 34. Similarly, after pointing to the icon 306, the area 304 is specified by hand.

Next, the recognition icon 307 is pointed with the pen 34. Here, the pointing information of the icon 306, the areas 302 and 30 specified by hand
4 and the pointing information of the icon 307 are sequentially sent from the mobile unit B to the main unit A via the mobile unit interface 31 and the main unit interface 26. Upon receiving the pointing information of the icon 306 and the information of the area 302 that follows, the control unit 14 of the main body A receives the pointing information of the icon 306 and the information of the area 302 that follows it, and uses the image information stored in the memory unit 19 in advance.
The character information 301 in the designated area 302 is cut out in the form of image information and temporarily stored in the memory unit 19.

Similarly, the same operation as described above is repeated for the next character area 304, and the character information 303 is stored in the memory section 19 in the form of image information. And icon 3
Upon receiving the pointing information (recognition start) of 07, the control unit 14 reads out the previously stored character information (image information format) 301 from the memory unit 19 and sends it to the media conversion unit 22.

[0087] The media conversion unit 22 recognizes characters from input image information and converts the information into character codes. The control unit 14 sends this converted character code to the mobile unit B via the main unit interface 26 and the mobile unit interface 31. Here, characters that are uncertain to be recognized by the media conversion unit 22 are also processed by the mobile unit B along with the second and third candidates.
send to Next, the control unit 14 controls the image generation unit 51 to generate a rectangular line figure corresponding to the area 302 (corresponding to the area 308 in FIG. 21).
, is sent to the portable unit B via the main unit interface 26 and the portable unit interface 31 in the form of image information.

[0088] Similarly, the character information 303 surrounded by the area 304 is also recognized by the main body part A, and the character codes and the first, second and third places of characters whose recognition is uncertain are determined. The candidate, partial image information for the character whose recognition is uncertain, and rectangular image information corresponding to area 304 (corresponding to area 311 in FIG. 21) are sent to portable unit B. The control section 32 of the portable section B displays the information sent from the main body section A on the display section 40, and the display screen is as shown in FIG. 21, for example. At this time, icons for various operations are also displayed simultaneously with the above-mentioned display contents.

In FIG. 21, 309 and 312 are the character information 308 which is the recognition result sent from the main body A.
and 311 are rectangular line figures that divide the character area. 310, 313, and 314 are characters whose recognition is uncertain and are displayed in reverse video. 315 to 321 are icons and displays required for operation, respectively.

Next, a method for correcting characters whose recognition is uncertain will be explained. First, when the operator points the character 310 with the pen 34, the first candidate character for recognition is displayed in the area 316, and at the same time, the character 310 is displayed in the area 316.
8, the original partial image information of the character is displayed. The display in area 318 can serve as a guide for the operator to determine the correct character. If this character is correct, confirm icon 3
15 with the pen 34,
This character is confirmed.

However, the characters displayed in area 316 are
If it's still wrong, add the character 310 to the pen 34.
When pointing with , the second candidate character for recognition is displayed in area 316. If this character is correct, by pointing the icon 315 with the pen 34, this character is confirmed. If it's wrong,
Further, when the character 310 is pointed with the pen 34, a third candidate character for recognition is displayed in the area 316. If this character is correct, by pointing the icon 315 with the pen 34, this character is confirmed.

[0092] If the third candidate character is also incorrect, the operator presses the correction icon 317 on the pen 34.
Pointing to the area 319, pen 3
4. Fill in the correct characters by hand. Thereafter, by pointing the recognition start icon 320 with the pen 34, the control unit 32 transfers the image information of the handwritten character to the online character recognition unit 32.
Send to 5. The online character recognition unit 35 recognizes input character information, and the recognition result is displayed in the area 321. At the same time, the character 310 is also replaced by the recognized result.

[0093] Here, if the recognition result of the online character recognition unit 35 is wrong, point at the character 310 again, point at the icon 317,
19 by handwriting, and by pointing at the icon 320, the recognition result is displayed in the area 321. The operator confirms whether this recognition result is correct. And if I'm wrong again,
Repeat the above operation until the recognition result is correct.

The above operations are not limited to operations on characters whose recognition is uncertain, but can be applied to any of the displayed characters 309 or 312. If an incorrect character is found, first select that character. By pointing and performing the above operations, the character can be corrected. Similarly, characters 31 whose recognition is uncertain
3,314 can also be corrected or corrected by the same operation as above.

[0095] In the above embodiment, when correcting an incorrect character, an online character recognition function is used to directly input a kanji character by hand. However, it is also possible to use a method in which kana characters are input by hand, recognized, and then converted into kanji using a kana-kanji conversion function, as shown in FIG. In this method, since the range of characters to be recognized is limited to kana characters, the amount of processing required by the online character recognition unit 35 is smaller compared to the method of directly recognizing kanji, so it requires less hardware. Simplification and improvement of recognition performance can be expected.

Furthermore, as a method for inputting kanji, for example, a reduced-sized keyboard is displayed on the display section 40 of the portable section B, and kana characters are input by directly pointing with the pen 34. After that, there is also a method of converting to kanji using kana-kanji conversion. This method has the advantage of avoiding recognition uncertainty associated with online character recognition.

[0097] By performing the operations described above, it is possible to specify the desired character area, correct characters that are uncertain in recognition, and correct characters that are incorrect, thereby completing a complete text. Can be done. Then, this text is sent again to the main body part A, stored in the database part 21, and can be used as part of a sentence when creating a later text. In addition, on the mobile unit B, an additional sentence is created using online character recognition for the text created by the above operation, and then again on the main unit A.
It is also possible to send the data to the database unit 21 and store it in the database unit 21.

Furthermore, in the above embodiment, after the image information is read in the main body part A, the image information is once sent to the mobile part B, the recognition character area is specified in the mobile part B, and the area information is used. Then, the main unit A performs character recognition again, and the recognition result is sent to the mobile unit B, where the characters are corrected or corrected. However, after sending the image information from the main unit A, the mobile unit performs all processing such as specifying the recognition area, performing character recognition using the online character recognition unit 35, and modifying and correcting characters that are uncertain about recognition. You can also use B. In addition, the present invention can be implemented with various modifications without departing from the gist thereof.

As explained above, instructions and operations related to text reader recognition, such as cutting out a recognized character area and correcting characters whose recognition is uncertain, have conventionally been performed using a mouse and keyboard. However, according to the above embodiment, writing is performed with the pen 34 on the tablet (position input section 33) of the portable section B while looking at the screen of the display section 40, so
Extremely natural and significantly improved efficiency. In particular, when correcting characters, conventionally the kanji were entered using a keyboard through kana-kanji conversion, but according to the above example, by using the handwritten online character recognition function, complicated keyboard operations are no longer necessary. It is possible to input characters directly by writing.

[0100]

Effects of the Invention As detailed above, according to the present invention, instructions and operations related to recognition, that is, setting of a recognition area,
It is possible to provide a character recognition method in which the command to start recognition and the correction of uncertain characters can all be easily performed with a pen touch, and the operability is significantly improved.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing a schematic configuration of a personal complex OA device to which a character recognition method according to an embodiment of the present invention is applied.

FIG. 2 is a block diagram showing a specific example of the configuration of the main body.

FIG. 3 is a block diagram showing a specific example of the configuration of a portable unit.

FIG. 4 is a block diagram showing in detail the configuration of an image processing section.

FIG. 5 is a block diagram showing in detail the configuration of an audio processing section.

FIG. 6 is a block diagram showing in detail the configuration of a database unit.

FIG. 7 is a diagram showing the structure of data stored in the database section.

FIG. 8 is a diagram showing classes of data stored in the database section.

FIG. 9 is a perspective view showing a handset as an audio input/output unit.

FIG. 10 is a diagram showing an example of a display screen of an audio information recording menu.

FIG. 11 is a diagram showing a configuration example of an audio information header.

FIG. 12 is a diagram illustrating linking recorded audio information with other data.

FIG. 13 is a diagram showing an example of a display screen when directly specifying a link.

FIG. 14 is a diagram showing an example of a voice memo icon.

FIG. 15 is a diagram showing an example of a display screen of a telephone function menu.

FIG. 16 is a diagram showing an example of a display screen of a facsimile function menu.

FIG. 17 is a diagram showing a configuration example of a facsimile information header.

FIG. 18 is a block diagram showing the configuration of a device required to implement a document creation function.

FIG. 19 is a diagram showing an example of a menu screen when adding a title to a file created using a document creation function.

FIG. 20 is a diagram showing an example of a text display screen (image information).

FIG. 21 is a diagram showing an example of a text display screen (recognition results).

FIG. 22 is a block diagram showing the configuration of a conventional text reader recognition system.

FIG. 23 is a diagram showing an example of a text display screen (image information) according to a conventional method.

FIG. 24 is a diagram showing an example of a text display screen (recognition result) according to a conventional method.

[Explanation of symbols]

A...Main unit (first electronic device), B...Mobile unit (second electronic device), 1...Image input section, 2...Character recognition section,
3...Control unit, 4...Online character recognition unit, 5...Display unit, 6...Pen, 7...Position input unit (tablet),
8...Control unit, 11...Position input unit, 12...Keyboard unit, 13...Image input unit, 14...Control unit, 18
...Image processing section, 19...Memory section, 20...Text reader processing section, 21...Database section, 22...
Media converter, 26... Main unit interface, 31
...Mobile unit interface, 32...Control unit, 33...
Position input section (tablet), 34... Pen, 35... Online character recognition section, 36... Image input section, 40...
...Display section, 44...Memory section, 45...Text data processing section.

Claims (1)

[Claims] Claim 1: A first electronic device comprising: a first electronic device having means for inputting image information including at least characters; and a portable second electronic device capable of exchanging data with the first electronic device; At least one of the electronic device and the second electronic device has a character recognition means, the second electronic device has a handwriting input means in which a display unit and a tablet are integrated and can input handwritten pen input, and the first electronic device has a character recognition means. When the character recognition means performs character recognition from the image information inputted by the device, before that, the handwriting input means of the second electronic device specifies an area of the image where the characters exist, and after recognition processing,
A character recognition system characterized in that it is possible to select candidates for characters whose recognition is uncertain and to correct characters that have been erroneously recognized based on the recognition results using handwriting input means of a second electronic device.