JP2005141520A - Method for setting character recognizing parameter in portable terminal and portable terminal - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To support mail and Internet connection, applications on WEB, and inputs such as registration, and to simplify the operation of a portable terminal. <P>SOLUTION: A parameter such as a character type or character line extracting parameter required for character recognition is automatically set in accordance with a driven application by using an OCR input as an input means to the application without cares by the user. Consequently time and labor for manually selecting an item to be inputted (recognized) can be eliminated and highly accurate character recognition processing can be realized. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a technique for selecting and recognizing characters by selecting a character string in a photographed image using an image input means, specifically, a mobile terminal or a mobile phone having a camera.

  Conventionally, it has been proposed that character recognition is performed on image data such as a telephone number, a URL, and an e-mail address taken from a camera, and that the recognition result is used as an input to support character input and simplify the operation ( For example, see Patent Document 1).

  As a URL input method for connecting to the Internet, a method using a barcode, a two-dimensional barcode, OCR, or the like has been proposed (see, for example, Patent Document 2).

JP 2002-152696 A (3rd to 6th terms, FIG. 7)

Japanese Patent Laid-Open No. 2003-198674 (term 4, FIG. 3)

  In the character recognition processing by the former OCR application, for example, after the user selects an item to be input such as a telephone number, URL, mail address, etc., a photograph is taken with a camera and character recognition processing is performed. However, there is a problem that it is necessary to manually select a recognition target, and it takes time to select a corresponding recognition target from a large number of recognition targets.

  In the latter case, when an OCR is input for information on an active application, for example, a URL, the types of characters to be recognized are limited to letters, numbers, and symbols. However, since these pieces of information are not used for character recognition processing, Depending on the camera shooting environment and characters to be recognized, the recognition accuracy may vary greatly, and OCR input may not function at all.

  In addition, applications and registration on the WEB are now possible, and the number of WEB sites that support these is increasing. However, since there are a plurality of input items in the application or registration input form, at present, OCR input by a mobile terminal has not been supported, and all of them must be manually input.

  SUMMARY OF THE INVENTION The present invention has been made to solve the above-described problems, and an object of the present invention is to support the input of e-mail, Internet connection, application on the WEB, registration, and the like by a mobile terminal, and simplify the operation.

  In order to solve the above problems, an image input unit (or a camera or an imaging unit) for inputting an image, a display unit for displaying image data input by the input unit, and an operation unit for a user to perform input operations such as buttons And using a character recognition unit that recognizes characters in the image and a mobile information terminal or mobile phone having a control unit for controlling the whole, using an OCR input as an input means to an application such as a translation application or a browser, We propose a method for automatically setting parameters necessary for character recognition, such as character types and character line extraction parameters, without the user being aware of it according to the application being used.

  It is possible to support the input of e-mail, Internet connection, application on the WEB, registration, and the like by a portable terminal, and the operation can be simplified.

  Hereinafter, embodiments of the present invention will be described with reference to FIGS. 1 to 5.

FIG. 1 is a configuration diagram illustrating an outline of a mobile terminal or mobile phone 100 (or simply referred to as a mobile terminal or a mobile device) having an image input unit according to the present embodiment.
A character recognition target image such as a business card, a magazine, or a signboard is input from the image input unit 110, line candidates are extracted by the character recognition unit 150, and character line candidates (images) are displayed on the display unit 120. A user selects a character line to be recognized from character line candidates extracted by operating the operation unit 130, and the recognition unit 150 recognizes the selected character line.

  The character recognition unit 150 extracts a character line to be recognized from the image data obtained by the character recognition parameter setting unit 151 and the image input unit 110 that set parameters for character recognition according to the running application. A character line extraction unit 152 that performs character recognition from the extracted character line data into character-like rectangles, a character identification unit 154 that performs character recognition on a character rectangle basis and converts the character rectangles into character codes, It comprises a knowledge collation unit 155 that collates a character dictionary with a word dictionary or a notation dictionary provided in advance and uses the collation result as a character recognition result.

  In the present embodiment, the GPS (Global Positioning System) 160 is provided to obtain the location information of the terminal itself, but this is not necessarily an essential element in the present invention.

  The functions of each unit and unit such as the character recognition unit 150, the operation unit 130, the image input unit 110, and the display unit 120 are controlled by a control unit 140 including a CPU, a memory, and the like. Each unit described above and below can be expressed as means, mechanism, and unit, and is basically a function that is processed and controlled by software or hardware, or a combination of software and hardware. It should be noted that it is desirable that an image captured, acquired, input, or the like is stored in a memory such as a control unit or a memory card provided in a portable terminal so as to be used for character recognition described later.

FIG. 2 is a diagram for explaining main character recognition processing from the character line extraction of the basic character recognition unit 150 using the portable terminal of FIG. 1 to obtaining a recognition result. The processing and control described below are mainly performed by the control unit and control unit 140, but the description is omitted.
A user uses an image input means 110 such as a CCD or an image sensor provided in the portable information terminal or the cellular phone 100 to shoot an image of a business card, magazine, or signboard that is a character recognition target, A character line 210 is extracted from the image 200 by the character line extraction unit 152 and displayed on the display unit 120. The user uses the operation unit 130 to select a recognition target character line from the displayed character line 210. The character cutout unit 153 cuts out the recognition target character line 220 for each character, and the character identification unit 154 executes character recognition processing for each character to generate the character recognition network 230. The character recognition network 230 is a network representation of candidate characters for character identification and stored in a memory. The character recognition network 230 includes incorrect character identification candidate characters (such as “|” and “/” in the drawing) and unnecessary character strings (“T”, “e”, and “direct” in the drawing) before and after the character string to be recognized. Etc.) may be included. Therefore, collation between the character recognition network 230 and the notation dictionary 240 provided in advance is performed by the knowledge collation unit 155, and a character string at a position where the notation and the collation match is extracted, so that the correct recognition result 250 can be easily and quickly shortened. It can be obtained with time-consuming processing.

  FIG. 2 shows an example in which a plurality of phone number notation patterns are stored in advance as the notation dictionary 240. The notation dictionary is compared with the extracted character string, and a phone number that matches one of the notations 241 is shown. An example in which only a portion is extracted as a recognition result character string 231 is also shown. In the above description, the description dictionary is used, but it is also possible to perform collation processing using a word dictionary.

  The finally recognized recognition result 250 is stored in a memory in the control unit 140 of the portable terminal, and is displayed on the display unit 120, and the text-based editing of characters is performed in accordance with the content that the user subsequently operates. Is possible.

  FIG. 3 is a diagram for explaining processing from application activation to execution.

The user selects, confirms and activates the application 300 displayed on the display unit 120 of the mobile phone 100 using the operation unit 130 (step S100). Examples of applications include a mail application, a browser, a translation application, a text editor, and the like.
Next, the user selects an input mode using the operation unit 130 (step S200). As the input mode 310, direct (hand) input, selection from a list, OCR input, barcode input, two-dimensional barcode input, GPS input, and the like are conceivable. The input item needs to be input when each application is executed, such as an email address in the case of a mail application and a URL in the case of a browser.
The controller 140 determines the presence / absence of image input (step S300) and the presence / absence of GPS input (step S700) according to the input mode selected by the user. For example, when selecting directly from a list or selecting from a list, the user inputs or selects an input item using the operation unit 130 (step S900), so there is no need for image input or GPS input. On the other hand, in the case of OCR input or barcode input, image input (step S400) by the image input unit 110 is necessary in order to use image data for input assistance. In the case of GPS input, GPS 160 is required. In the case of GPS input, recognition processing is not performed, and position information obtained from the GPS 160 is used as input information and set as an input item (step S800). For example, when setting the current station name as an input item in a route guidance application that supports destination guidance, it is conceivable that the station name is automatically searched and set based on the location obtained from the GPS 160.

  A recognition process (step S500) is performed on the image data obtained by the image input unit 110. If the input is OCR input, the character recognition unit 150 performs character recognition processing, and if the input is barcode input, performs barcode recognition processing. The recognition result is set as an input item (step S600). If there is an error in the recognition result, the user must use the operation unit 130 to correct it. It is also possible to make it recognize again using the image input unit 110.

  When the input item setting is completed, processing corresponding to the activated application is executed (step S1000). For example, in the case of a mail application, it is conceivable to send a mail to the mail address input by the image input means 110 and register the mail address in the phone book. In the case of a translation application, it is conceivable to translate the recognized word and display the result on the display unit 120.

  FIG. 4 is a diagram for explaining processing of the character recognition parameter setting unit 151 in the character recognition unit 150, and describes a character recognition parameter setting method using a portable terminal alone.

  The control unit 140 checks whether or not the input mode selected by the user is OCR input (step S501), and does nothing if it is not OCR input. In the case of OCR input, character line parameters (S502), character type parameters (S503), and knowledge dictionary parameters (S504) are set according to the application.

  The character line parameters are parameters used for the character line extraction unit 152, and indicate parameters such as the maximum width / height and minimum width / height of the character line to be extracted. The character type parameter is a parameter used for the character identification unit 153 and indicates a character type such as a number, an alphabetic character, or a Chinese character. The knowledge dictionary parameter is a parameter used for the knowledge matching unit 154, and indicates a word dictionary such as English words and URLs.

  For example, when the active application is an English word translation application, the character line extraction parameter is set to an optimum English word extraction parameter for extracting a short character line. The character type parameter is set to the character type of English characters and some symbols used for English words. The knowledge dictionary parameter is set in the knowledge dictionary for English words.

Thus, by setting the following table in advance as the “activation application-character recognition parameter table”, it is possible to automatically set the character recognition parameters according to the application being activated.
Application: English word translation Character line parameters: Word extraction parameters (width: x, height: y, etc)
Character type parameter: Alphabet + symbol ("-", "'", etc)
Knowledge dictionary parameters: English words Character recognition parameters can also be set by using GPS 160 together. For example, when a translation application is used overseas, the character type parameter can be automatically switched to English, French, German, Chinese, etc. according to the location obtained from the GPS 160. Further, the translation language of the translation application can be switched according to the location obtained from the GPS as described above.

  Based on the character recognition parameters set above, character recognition processing (step S505) such as character line extraction processing, character extraction processing, character identification processing, and knowledge matching processing is performed.

  FIG. 5 is a diagram for explaining a character recognition parameter setting process using WEB, and a character recognition parameter setting method through communication.

  An example will be described in which the OCR input of a portable terminal is used for inputting a present application form field (name field, address field) published on a certain WEB.

  The user selects the application 300 displayed on the display unit 120 of the mobile phone 100 using the operation unit 130, and then activates the browser (step S1000) and accesses the designated WEB site (step S2000). When inputting in the present application form field published on the accessed WEB site, the user selects the input mode 310 as described above, and the control unit 140 checks whether or not the input is an OCR input (step S100). S3000). If it is not OCR input, it is input directly or by input means such as barcode input or two-dimensional barcode input (step S4000). In the case of OCR input, the application form is downloaded from the WEB site (step S5000). The application form is information on what items are input in which order, for example, information such as inputting a name first and an address second.

  Next, it is checked whether or not there is an item that needs to be input (step S6000). If the input item needs to be set, the image input unit 110 is used to input the image data to the location corresponding to the item. Obtain (step S7000). Character recognition processing (step S8000) is performed on the obtained image data by the character recognition unit 150. At this time, since the items recognized from the application form are already known, the character recognition parameter setting unit 151 sets the parameters for character recognition based on this information. For example, since it is known that the first item to be entered from the application form is a name, a parameter for name recognition is set. The difference from the character recognition parameter setting method by the mobile terminal alone is that the character recognition parameter is associated with the attribute of the input item, not the active application.

As described above, by setting the following table in advance as the “input attribute-character recognition parameter table”, it is possible to automatically set the character recognition parameter in accordance with the input attribute.
Input attribute: Name Character line parameter: Name extraction parameter (width: α, height: β, etc)
Character type parameter: Kanji + Katakana + Hiragana Knowledge dictionary parameter: Name The recognition result is set as an input item (step S9000). If there is an error in the recognition result, the user must use the operation unit 130 to correct it. It is also possible to make it recognize again using the image input unit 110.

  The above process is repeated for the number of input items indicated on the application form, and when there are no more input items, the user sends the application form with all input items set to the WEB site using the operation unit 130. (Step S10000) and the application is completed.

  As described above, the character recognition parameters are automatically set based on information obtained from the WEB site without the user's awareness, so there is no need to manually input items to be read and high-precision character recognition processing. This can be realized, and the input operation can be greatly simplified.

It is explanatory drawing which shows schematic structure of the mobile phone with a camera as an Example. It is a flowchart of the character recognition process by a portable terminal. It is a flowchart until application activation / execution. It is a flowchart of a character recognition parameter setting process. It is a flowchart of the character recognition parameter setting process using WEB.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 ... Mobile terminal (mobile phone), 110 ... Image input part, 120 ... Display part, 130 ... Operation part, 140 ... Control part, 150 ... Character recognition part, 151 ... Character recognition parameter setting part, 152 ... Character line extraction part 153 ... Character extraction unit, 154 ... Character identification unit, 155 ... Knowledge verification unit, 200 ... Input image, 210 ... Extracted character line, 220 ... Recognized character line, 230 ... Character recognition network, 231 ... Target for verification Character line, 240 ... notation dictionary, 241 ... notation to be verified, 250 ... recognition result, 300 ... application selection screen, 310 ... input mode selection screen

Claims (8)

An image input unit for inputting an image;
A display unit for displaying the input image;
A character recognition unit for recognizing character data included in the input image;
A portable terminal having a control unit for controlling each unit,
The mobile terminal according to claim 1, wherein the control unit automatically sets a parameter for character recognition processing according to a running application. The mobile terminal according to claim 1,
The control unit automatically sets a parameter for character recognition based on an input form downloaded from a WEB site. The mobile terminal according to claim 1 or 2,
The mobile terminal according to claim 1, wherein the control unit automatically sets a target character recognition item and a recognition order based on an input form downloaded from a WEB site. The mobile terminal according to claim 1, wherein
It further has a GPS for acquiring location information,
The portable terminal according to claim 1, wherein the control unit automatically sets a parameter for character recognition based on position information obtained from the GPS. A method for setting a character recognition parameter for image data input to a mobile terminal,
Launching the application; and
Inputting an image;
Automatically setting a parameter for character recognition processing according to the type of the application;
A method for setting a parameter for character recognition in a portable terminal characterized by comprising: A method for setting a parameter for character recognition in the mobile terminal according to claim 5,
The step of automatically setting the character recognition processing parameter sets the character recognition parameter automatically based on an input form downloaded from a WEB site. A method for setting a parameter for character recognition in a mobile terminal according to claim 5 or 6,
The step of automatically setting the parameter for character recognition processing includes automatically setting a target character recognition item and a recognition order based on an input form downloaded from a WEB site. How to set parameters for use. A method for setting a parameter for character recognition in the mobile terminal according to claim 5,
The step of automatically setting the parameter for character recognition processing further includes the step of acquiring position information by GPS, and automatically setting the parameter for character recognition based on the position information obtained from the GPS. To set parameter for character recognition in portable terminal.

Images