KR20050089368A - An apparatus and a method for character recognition - Google Patents

Abstract

The present invention relates to an apparatus for recognizing a word by sensing a character in an image of a printed document and a method for utilizing the recognized word.

Character recognition apparatus according to an embodiment of the present invention is a light source for irradiating light; A display unit configured to display an image in which light emitted from the light source is reflected from the ground; An image detector for detecting an image with light reflected from the ground; And a character recognition unit that recognizes the sensed image as a character.

The present invention can replace the keyboard input method by easily recognizing a word that is difficult to input through the keyboard when using the electronic dictionary function by easily inputting the character recognition device.

Description

AN APPARATUS AND A METHOD FOR CHARACTER RECOGNITION}

The present invention relates to a character recognition apparatus and method for recognizing a word of a printed document, and more particularly, to a device for recognizing a word by detecting a character in an image of a printed document and a method of utilizing a recognized word. .

Generally, the following two methods are representative of the printed character recognition method.

The first method is Optical Character Recognition, which scans the entire printed page with a flatbed scanner, analyzes the bitmap image of the scanned page on a computer, and then Extract it and recognize it. However, the conventional optical character recognition method by the scanner has a problem in that the entire document is written, even if only one word is to be recognized.

The second method is a method in which a user holds a pen-shaped device in his hand and scans and inputs a desired word with a pen tip, and sequentially scans images of small areas and combines the scanned areas by image processing. The pen type conventional character recognition device may have a small field of view of the camera sensor, so the optical path from the character to the camera sensor is short, making the device easy to manufacture, but it is impossible to scan characters larger than the pen tip diameter. It is a hassle to move the strings by hand. In addition, in the conventional pen type conventional character recognition apparatus, since the character currently being scanned is hidden under the pen, there is no way for a user to check whether a desired character is properly scanned.

On the other hand, as a code recognition device, a barcode scanner that recognizes a bar code recorded in an industrial product, and a magnetic code that records a bank and account code of a personal check in numbers of special fonts that have been magnetically processed are recorded. There is a code scanner. Such a device for recognizing codes is only suitable for character recognition for automatically extracting and recognizing words recorded in a complicated printed document because it can only scan patterns or specially processed magnetic codes for easy recognition. In addition, the barcode scanner inputs the entire code at the same time without depending on the sequential scanning method, and because the optical path from the barcode to the sensor is long, the size of the device is large, so it is not suitable for use in an electronic dictionary.

In general, when the user inputs a procedure of a word that the user does not know, the electronic dictionary searches for the input word and informs the user of its meaning and pronunciation. In this case, the electronic dictionary may be implemented as an independent device such as an electronic notebook or a free electronic dictionary provided by an Internet portal site connected through software or a communication terminal in a computer. An input method used in an electronic dictionary or an electronic dictionary provided in a computer software or portal site is performed by directly inputting a word to be searched by a user using a keyboard (keypad).

However, the conventional electronic dictionary requires a user to directly enter a word that he does not know through an electronic organizer or a keypad or keyboard provided in a computer, so that the user may use the function of the dictionary if the Chinese character is difficult to enter or cannot enter a word. There is no problem. In addition, when producing a conventional electronic dictionary, a manufacturer has to input a meaning and pronunciation of a word directly and make a database so that a lot of time and effort is required.

On the other hand, the educational site or software that provides the foreign language learning related services only sends or displays the pronunciation or meaning to the user when providing the foreign language learning related services. There is a drawback that is not available at. For example, if a user solves a problem described in his or her textbook separately from an educational site or software, the user has to check the process of checking and scoring an answer.

In addition, when a user uses a foreign language learning-related textbook, there is a disadvantage in that the pronunciation of the textbook can be confirmed as audio only by purchasing a pronunciation tape or CD of the textbook at an additional cost.

Conventional interactive teaching methods use software that can only be performed on computers or require special pens and hardware devices. Therefore, there is an urgent need for a means of maintaining the convenience and portability of printed textbooks that can be easily used, and at the same time linking them with software having various interactive functions and databases when needed.

The present invention has been made to solve the above problems of the prior art, and an object thereof is to provide a character recognition apparatus and method for facilitating character input in an electronic dictionary.

Another object of the present invention is to provide a character recognition apparatus and method for scanning a printed foreign language document and recognizing a word to transmit a pronunciation to a user.

In addition, the present invention is to provide a character recognition apparatus and method for allowing the language-related software to transmit the fingerprint printed on the page in accordance with the language-related software in recognition of the page described in the language textbook to the speaker For that purpose.

In addition, the present invention is to provide a character recognition apparatus and method that can recognize the answer number selected by the user when the user solves the multiple-choice questions printed on the textbook to determine the correct answer in the software and feedback to the user do.

It is also an object of the present invention to provide a character recognition apparatus and method that can easily produce the contents described in a printed encyclopedia, language dictionary, etc. as an electronic dictionary.

In addition, an object of the present invention is to provide a character recognition apparatus and method that allows a user to immediately check the contents of a document currently being scanned.

In order to achieve the above object, a character recognition device according to an embodiment of the present invention is a light source for irradiating light; A display unit configured to display an image in which light emitted from the light source is reflected from the ground; An image detector for detecting an image with light reflected from the ground; And a character recognition unit that recognizes the sensed image as a character.

In addition, the character recognition apparatus according to another embodiment of the present invention is a light source for irradiating light; A first display unit configured to display an image in which light emitted from the light source is reflected from the ground; A first detector configured to detect image data with light reflected from the ground; A second sensing unit sensing a position of light irradiated by the light source; A button unit having a predetermined button and generating a control signal according to the click of the button; A character recognition unit that recognizes the image data as a character; And a second display unit which displays a character recognized by the character recognition unit.

In addition, the character recognition method according to another embodiment of the present invention maintains a character recognition device for recognizing an image recorded on a predetermined page as a character and an electronic dictionary for performing a dictionary search function for a predetermined input character step; Receiving an image recorded on the ground from the text recognition device as text data; And inputting the input text data as a search word of the electronic dictionary.

In addition, the character recognition method according to another embodiment of the present invention comprises the steps of maintaining a character recognition device for recognizing a character recorded on a predetermined page, a predetermined display device and a voice transmission device for transmitting a predetermined voice; Receiving an image recorded on the ground from the text recognition device as text data; Displaying the text data on the display device; Converting the text data into voice data; And transmitting the voice data to the voice transmitting device.

Hereinafter, a character recognition apparatus and method according to the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram of a character recognition apparatus according to the present invention, wherein the character recognition apparatus includes a light source 110, a display unit 120, an image sensing unit 130, a character recognition unit 140, a display unit 150, and a speaker. 160.

Referring to FIG. 1, the light source 110 may be implemented as an illumination element such as an LED lamp, and irradiates light onto the ground (document) to be scanned. The display unit 120 displays an image in which light emitted from the light source 110 is reflected from the ground, which is currently being scanned. The display unit 120 is a means for allowing the user to directly check the contents of the portion currently being scanned from the contents written on the ground to be scanned. The display unit 120 may be implemented as a convex lens for enlarging the image. The image detector 130 may be implemented as an image sensor, and detects an image with light reflected from the ground. The character recognizer 140 recognizes the image detected by the image detector 130 as a character. The display unit 150 displays the characters recognized by the character recognition unit 140. The voice transmitter 160 converts the text data recognized by the character recognizer 140 into voice data and transmits the converted voice data. The path setting unit 170 transmits the light reflected from the ground to the display 120 and transmits the light reflected from the ground to the image sensing unit 130. The path setting unit 170 may be implemented as a transflective mirror that transmits the light reflected from the ground to the display unit 120 and reflects the light reflected from the ground. The path setting unit 170 further includes path reducing means for reducing the path of the light reflected by the plate beam splitter. The path reducing means may be implemented with a prism that refracts the path of light reflected by the transflective mirror.

2 is a block diagram of a character recognition apparatus according to an exemplary embodiment of the present invention. The character recognition apparatus including a mouse function includes a button unit 200, a light source 210, a first display unit 220, and a detection unit 230. ), A text recognition unit 240, a second display unit 250, and a voice transmitter 260.

Referring to FIG. 2, the button unit 200 includes a plurality of buttons required to perform a mouse function, and generates a control signal according to the click of the button. The button unit 200 may include a button for selecting a character recognition device or a mouse function operation, and may generate a control signal for performing a function selected by a user. The light source 210 may be implemented as an illumination element such as an LED lamp, and irradiates light onto the ground to be scanned. The first display unit 220 displays the content of the portion currently being scanned on the ground to be scanned. The sensing unit 230 includes a first sensing sensor 231 and a second sensing sensor 232, and detects a surface irradiated with light by the light source 210 to perform an operation selected by the button unit 200. Scanning to sense image data or to detect the position of the light irradiated by the light source (210). The first detection sensor 231 is an image sensor having a high resolution (eg, 640 × 80 resolution) and detects image data by scanning a surface irradiated with light by the light source 210. The second detection sensor 232 is a low resolution (eg, 16 × 16 resolution) image sensor that is used in a general optical mouse. The detector 230 may be implemented as one image sensor that combines the functions of the two sensors 231 and 232 without using the first and second sensors 231 and 232. The character recognizer 240 recognizes a character from the image data sensed by the detector 230. The second display unit 250 may be implemented by a display means such as an LCD, and displays the text data recognized by the character recognition unit 240. The voice transmitter 260 converts the text data recognized by the character recognizer 240 into voice and transmits the voice data.

3A is a plan view illustrating an upper housing of an optical mouse including a function of a character recognition device according to a specific embodiment of the present invention. The upper housing of the optical mouse including a function of the character recognition device includes a left button 301 and a scroll wheel ( 302, a right button 303, a first display means 310, and a second display means 320.

2 and 3A, the left button 301, the scroll wheel 302, and the right button 303 are the same as the button functions of a conventional mouse, respectively, and are included in the button unit 200 shown in FIG. 2. Can be. The first display unit 310 is a cylindrical lens window, and the first display unit illustrated in FIG. 2 is a means for allowing a user to directly check an image described in a document currently detected by the detector 230. Play the same role as 220. The second display means 320 may be implemented as a display element such as an LCD, and the same role as the second display unit 240 shown in FIG. 2 as a means for displaying a character recognized by the character recognition unit 240. Do this. The speaker 330 is a means for transmitting a character recognized by the character recognition unit 240 as a voice, and performs the same role as the voice transmitter 260 shown in FIG. 2.

3B is a cross-sectional view of the optical mouse including the function of the character recognition apparatus according to the exemplary embodiment of the present invention, taken on the basis of AA ′ shown in FIG. 3A.

Referring to FIG. 3B, an optical mouse including a function of a character recognition device according to a specific embodiment of the present invention includes a mouse button 301, an upper window 310, a light source 311, a semi-transmissive mirror 312, and a lower window. 313, optical mouse image sensor 314, upper window 316, prism 317, lens 318, image sensor 319, LCD 320, character recognition circuit 321, speaker 330 It includes. The mouse button 301 is a component provided to perform a function of a general optical mouse and is represented on the left button 301, the scroll wheel 302, and the right button 303. The upper window 310 may be embodied as a cylindrical lens window. The upper window 310 is a type of display means for directly showing an image of a document currently being scanned to a user when the optical mouse performs a function of a character recognition device. The image described in the document located under the window 313 is shown. The upper window 310 may be implemented as a convex lens that can be enlarged and shown to the user when the image described in the document is very small. The light source 311 may be implemented as an illumination element such as an LED, and irradiates light onto a document to be scanned. Light irradiated by the light source 311 is reflected on the ground (document), partially transmitted through the transflective mirror 312 to the upper window 310, and partially reflected by the transflective mirror 312. The transflective mirror 312 transmits the light irradiated by the light source 311 to the upper window 310 by transmitting a portion 316 of the light 314 reflected on the ground, and transmitting the portion of the light reflected on the ground. Reflect 315 back to the prism 317. The light 315 reflected by the transflective mirror 312 is rotated 180 degrees by the prism 317 and then passes through the lens 318 to form an image on the image sensor 319. The light 316 passing through the transflective mirror 312 allows the user of the character recognition device to view characters written in the document to be currently recognized through the upper window 310. The optical mouse image sensor 314 may be provided separately from the image sensor 319 as an image sensor of a general optical mouse, or may be included in the image sensor 319. The prism 317 changes the path of the light 315 reflected by the transflective mirror 312 by 180 degrees, making it easier to use the sized inner housing of the mouse. In the embodiment of the present invention, by using the transflective (translucent) mirror 312, the user can easily select the character of the document to be recognized by the document recognition device, and through the prism 317 the optical path in the document recognition device Can be reduced. The image sensor 319 detects an image recorded in the document through the image formed through the lens 318. The image sensor 319 may selectively perform the function of the optical mouse image sensor 314 and the function of the image detection sensor. The character recognition circuit 321 includes interface means for performing an interface with a computer (communication terminal) for recognizing an image sensed by the image sensor 319 as a character in the case of an optical mouse combined character recognition device. The interface means is an interface circuit for transmitting a sensed image to a computer, and includes a USB controller chip, an oscillator, and the like. The character recognition circuit 321 may include a microprocessor, a memory, or the like for recognizing an image sensed by the image sensor 319 as a character when the device is an independent device instead of an optical mouse combined character recognition device. The LCD 320 displays characters recognized by the character recognition circuit 321 when the LCD 320 is an independent device instead of an optical mouse combined character recognition device.

3C is a bottom view of the lower housing of the optical mouse including the function of the character recognition device according to the exemplary embodiment of the present invention, wherein the lower housing of the optical mouse includes a lower window 313, a light source guide 340, and a selection button 350. , 360).

3A to 3C, the lower window 313 is a window that is drilled in a predetermined size in the lower part of the optical mouse and is located in the area of the lower window 313 of the document being scanned by the user and is currently being scanned. The image can be viewed directly in the upper window 310. The light source guide 340 guides light emitted from the lens 318 and the light source 311 in front of the image sensor 319 of the optical mouse. The light source guide 340 guides the light emitted from the light source 311 to be accurately irradiated onto the document to be scanned. The selection buttons 350 and 360 are buttons for selecting a function of the character recognition device such as a general optical mouse function, a word recognition function, a language learning function, etc. by the user, and may be included in the button unit 200 shown in FIG. 2. have. The selection buttons 350 and 360 may be included in buttons 301 and 302 of the upper housing as shown in FIG. 3A instead of being included in the lower housing of the optical mouse as shown in FIG. 3C.

4A and 4B are diagrams illustrating a word recognition method applied to the present invention, and show a method of recognizing characters scanned by scanning an image located at a cross-shaped cursor in the center of an image sensor as a word.

4A and 4B, reference numeral 410 corresponds to a content that a user can identify in the upper window 310 of a part of the document to be scanned by the lower window 313, and reference numeral 411 denotes a portion of the document to be scanned. A cross-shaped cursor in the center of the upper window 310 or the lower window 313 is shown. The user moves the document recognition device such that the cursor of the upper window 310 is located at the center of the word to be scanned among the contents of the document to be scanned through the upper window 310. In this case, the document recognition device recognizes a document by software performed on a computer connected to the mouse when used with a mouse, and a micro process included in the document recognition circuit 321 when the document recognition device is not a mouse. The document is recognized by software running in the processor. The document recognition apparatus estimates the orientation angle θ of the character string, finds the character strings 413, and separates the words on the character string 413 into a bounding box 412 ( word segmentation). The image in the bounding box 412 under the cursor located in the center performs character segmentation again and then recognizes the word using a word recognition process (for example, an artificial neural network).

FIG. 5 is a flowchart illustrating a character recognition method according to an exemplary embodiment of the present invention. An electronic dictionary function of inputting a recognized character as a search word and outputting the meaning and pronunciation of the word is performed.

1 to 5, in operation 510, the image detector 130 detects an image of a document to be scanned through a text recognition device, and the display 120 is currently scanned through a document recognition device. Display an image of the document so that the user can see it immediately. In operation 520, the character recognition unit 140 recognizes the image detected by the image detection unit 130 as a character, and the display unit 150 allows the user to check the character recognized by the character recognition unit 140. Mark it. In operation 530, the character recognition unit 140 recognizes the scanned character as a word, and inputs the recognized word as a search word according to the execution of the electronic dictionary software being executed in the internal memory. In operation 540, the character recognition unit 140 searches for the meaning and pronunciation of the word in the electronic dictionary database stored in the internal memory of the input word according to the execution of the electronic dictionary software. In operation 550, the character recognition unit 140 outputs the meaning and pronunciation of the searched word to the display unit 150. The character recognition unit 140 may transmit the meaning and pronunciation of the searched word as a voice through the voice transmitter 160.

The character recognition method according to the embodiment of the present invention as disclosed in FIG. 5 has an advantage that a character recognition device may input a word which is difficult for a user to input with a keypad when using the electronic dictionary function.

6 is a flowchart illustrating a text recognition method according to another embodiment of the present invention, in which text recorded in a document is recognized, converted into voice data, and sent.

1 and 6, in step 610, the image detector 130 detects an image of a document requested to be scanned by the document recognition device, and the display 120 displays an image currently being detected. do. In operation 620, the character recognition unit 140 recognizes a character from the image detected by the image sensing unit 130, and in operation 630, the display unit 150 generates the character recognized by the character recognition unit 140. Is displayed. In operation 630, the voice transmitter 160 converts the text data recognized by the character recognition unit 140 into voice data, and transmits the converted voice data in operation 640.

Character recognition method according to an embodiment of the present invention as shown in Figure 6, if a character recorded in the document is recognized, for example in the case of language textbooks without having to purchase a separate audio tape or CD to the textbook through the text recognition device After the words described are recognized, the words may be transmitted by voice. In addition, the user recognizes the current contents of the textbook (page number, correct answer) using the character recognition device, and then informs the language software running on the computer of the contents of the current user's learning. You can easily communicate with each other. For example, the user may inform the user of the data, voice, etc. corresponding to the page recognized by the character recognition method according to the present invention through the monitor and the speaker of the computer, and the answer checked when the user solves the multiple choice question on the page. Can be easily scored by software.

Embodiments of the invention also include computer-readable media containing program instructions for performing various computer-implemented operations. The computer readable medium may include program instructions, data files, data structures, etc. alone or in combination. Program instructions recorded on the media may be those specially designed and constructed for the purposes of the present invention, or they may be of the kind well-known and available to those having skill in the computer software arts. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tape, optical media such as CD-ROMs, DVDs, and magnetic disks, such as floppy disks. Magneto-optical media, and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. The medium may be a transmission medium such as an optical or metal wire, a waveguide, or the like including a carrier wave for transmitting a signal specifying a program command, a data structure, or the like. Examples of program instructions include not only machine code generated by a compiler, but also high-level language code that can be executed by a computer using an interpreter or the like.

As described above, the present invention has been described by way of limited embodiments and drawings, but the present invention is not limited to the above-described embodiments, which can be variously modified and modified by those skilled in the art to which the present invention pertains. Modifications are possible. Accordingly, the spirit of the present invention should be understood only by the claims set forth below, and all equivalent or equivalent modifications thereof will belong to the scope of the present invention.

According to the present invention, when using the electronic dictionary function, it is possible to replace the keyboard input method by easily recognizing a word that is difficult to input through the keyboard with a character recognition device.

In addition, according to the present invention, as a link between the printed language textbook and the language learning software on the computer, the fingerprint printed on the textbook may be transmitted through a speaker provided in the computer.

In addition, according to the present invention, by the character recognition device recognizes the answer number selected when the user solves the multiple-choice questions printed on the textbook, it is easy to grade the correct answer, it is possible to provide appropriate feedback to the user.

In addition, according to the present invention, by using a character recognition device to recognize the content recorded in the dictionary, the meaning and pronunciation of the word can be confirmed by the speaker or the electronic dictionary can be easily produced.

1 is a view showing the configuration of a character recognition apparatus according to the present invention.

2 is a diagram illustrating a configuration of a character recognition apparatus according to an embodiment of the present invention.

3A is a plan view illustrating an upper housing of an optical mouse including a function of a character recognition device according to a specific embodiment of the present invention.

FIG. 3B is a cross-sectional view of the optical mouse including the function of the character recognition apparatus according to the exemplary embodiment of the present invention, taken along the line A-A 'as shown in FIG. 3A.

3C is a bottom view of the lower housing of the optical mouse including the functions of the character recognition apparatus according to the specific embodiment of the present invention.

4A and 4B illustrate a word recognition method applied to the present invention.

5 is a flowchart illustrating a character recognition method according to an embodiment of the present invention.

6 is a flowchart illustrating a character recognition method according to another embodiment of the present invention.

<Description of the symbols for the main parts of the drawings>

110, 210, and 311: light source 120: display unit

130: image detection unit 140, 240, 321: character recognition unit

150: display unit 160, 260: speaker

200: button portion 220: first display portion

230: detection unit (231: first detection sensor, 232: second detection sensor)

301: left mouse button 302: scroll wheel

303: right mouse button 310: upper window

312: transflective mirror 313: lower window

314: light reflected off the ground 315: light reflected off the transflective mirror

316: light through the transflective mirror 317: prism

318: lens 320: LCD

330: speaker 340: light source guide

350, 360: Function selection button 410: Image seen from the top window

411: Cursor in center of top window

412: bounding box for separating a word in a string

413: string θ: posture angle of the string

Claims (13)

In the character recognition apparatus, A light source for irradiating light; A display unit configured to display an image in which light emitted from the light source is reflected from the ground; An image detector for detecting an image with light reflected from the ground; And Character recognition unit for recognizing the detected image as a character Character recognition apparatus comprising a. The method of claim 1, And the display unit is a convex lens for enlarging the image. The method of claim 1, And a display unit for displaying a character recognized by the character recognition unit. The method of claim 1, And a path setting unit for transmitting the light reflected from the ground to the display unit and transmitting the light reflected from the ground to the image sensing unit. The method of claim 4, wherein And the path setting unit is a transflective mirror that transmits the light reflected from the ground to the display unit and reflects the light reflected from the ground. The method of claim 5, Characterization apparatus further comprises a path reducing means for reducing the path of light reflected from the transflective mirror. The method of claim 6, And the path reducing means is a prism that refracts the path of light. In the character recognition apparatus, A light source for irradiating light; A first display unit configured to display an image in which light emitted from the light source is reflected from the ground; A first detector configured to detect image data with light reflected from the ground; A second sensing unit sensing a position of light irradiated by the light source; A button unit having a predetermined button and generating a control signal according to the click of the button; A character recognition unit that recognizes the image data as a character; And A second display unit which displays a character recognized by the character recognition unit Character recognition apparatus comprising a. The method of claim 8, And the button unit generates a control signal for selecting a sensing result of the first sensing unit or the second sensing unit. The method according to claim 1 or 8, And a voice transmitter for transmitting a character recognized by the character recognizer as a voice. In the character recognition method, Maintaining a character recognition device for recognizing an image recorded on a predetermined sheet of paper as a character and an electronic dictionary for performing a dictionary search function for a predetermined input character; Receiving an image recorded on the ground from the text recognition device as text data; And Inputting the input text data as a search word of the electronic dictionary Character recognition method comprising a. In the character recognition method, Maintaining a character recognition apparatus for recognizing a character recorded on a predetermined sheet of paper, a predetermined display apparatus, and a speech transmitting apparatus for transmitting a predetermined speech; Receiving an image recorded on the ground from the text recognition device as text data; Displaying the text data on the display device; Converting the text data into voice data; And Transmitting the voice data to the voice transmitter; Character recognition method comprising a. A computer-readable recording medium for recording a program for executing the method of claim 11 or 12 on a computer.