KR20020055487A - IMT-2000 utilization a document cognition/a voice recitation means - Google Patents

Abstract

PURPOSE: A method of recognizing and reading a document through an IMT-2000 terminal is provided to photograph a document by using a video camera installed in the IMP-2000 terminal, as recognizing characters of the document, and to read the document through a speaker of the terminal, thereby enabling the old or the visually impaired to immediately confirm the contents of the document. CONSTITUTION: A video camera photographs a document(S1). The camera makes an image inputted through an IMT-2000 terminal a black and white image, and binarizes the image to input a document image(S2). The terminal separates the document inputted as the image, interpreting the image of the document, and separates areas from the interpreted image, separating characters, then recognizes the separated characters to output recognized results on a display(S3). If an error is generated in the results, the terminal corrects the error(S4). If no error is generated, the terminal selectively transmits data to a computer or stores the results on a self database(S5). The terminal recognizes each character of the stored document, and combines the characters as a WAV file(S6). The terminal combines data formed as the WAV file, and reads the data with a voice(S7).

Description

IMT-2000 utilization a document cognition / a voice recitation means}

The present invention relates to a voice reading method through document recognition, which inputs and recognizes a document image using an image camera mounted in the IMT-2000, sends and receives a text document with a computer, and reads it by voice. It relates to document recognition / voice reading method through IMT-2000.

Currently, paper documents are stored and stored in a database on a computer using a scanner. In order to share them, they are shared with each other via a LAN to share data between computers. You are sharing a document via e-mail. And in order to share data in other regions, data can be shared only where a computer or fax machine is installed, but it is very inconvenient to share data with a main computer where a computer or fax machine is not installed. There is a ham.

In the case of a cellular phone or a PCS phone used as a personal communication device, a simple text message can be exchanged with each other, but it is very difficult to share a document between a computer and a personal mobile communication. In order to send a document, a document was sent by connecting with a computer, and it was very difficult to share data between a personal communication device and a computer or a personal communication device.

In addition, in the case of the elderly and the blind who are hard to see, there is a problem in that even if the document is received by the personal communication device, it is not possible to grasp what the contents are.

Accordingly, the present invention is to solve the above problems, the document recognition / audio reading method through the IMT-2000 of the present invention can be input by imaging any document through the image camera installed in the terminal of the IMT-2000 It provides a document recognition / voice reading method through IMT-2000 that allows the document to be shared with the main computer, and to read the contents of the received document or video input through voice. The purpose is.

The document recognition / voice reading method through IMT-2000 according to the present invention for achieving the above object comprises the step S1 of photographing a document using an image camera installed in the terminal of the IMT-2000; S2 step of monochromeizing the image input through the IMT-2000, and binarizes the image to input a document image; In the document input as the image, it is divided into categories, and the image of the text is interpreted therein, the region is separated from the interpreted image, the character is separated from the separated region, and the character is recognized by the separated text. Step S3 of displaying a result on the terminal display panel of the IMT-2000 to indicate that the character is recognized; Step S4 of correcting an error in the result output to the display panel of the terminal of the IMT-2000; Step S5 of selectively transmitting data to a computer or storing a recognition result in its own database when there is no error in the result output to the terminal display panel of the IMT-2000; Step S6 of recognizing each character of the document stored in the database and combining them into a corresponding wave (WAV) file; And S7 step of combining the data formed into the wave (WAV) file and reading it out as voice.

In addition, according to the present invention, the step S6 may include a step S61 of encoding characters of data inputted through a video camera installed in the terminal of the IMT-2000 and converted into a text document; Step S62 of classifying the coded characters; Step S63 of determining which character is the coded character; Step S64 of selecting which character the coded character corresponds to; Step S65 of combining the respective coded character wave (WAV) files; And S66 of expressing the combined wave (WAV) file as a voice through a speaker.

1 is a block diagram of a document recognition / speech reading method through IMT-2000 according to the present invention.

Figure 2 is a flow chart showing the steps by the document recognition / speech reading method through IMT-2000 according to the present invention.

3 is a flowchart showing a voice expression method of the present invention.

* Description of the symbols for the main parts of the drawings *

10: main computer 20: IMT-2000

21: image camera 22: display panel

23: speaker

Hereinafter, a document management system and a method thereof according to the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram of a document recognition / voice reading method through IMT-2000 according to the present invention, Figure 2 is a flow chart showing the steps by the document recognition / voice reading method through IMT-2000 according to the present invention, 3 is a flowchart showing a voice expression method of the present invention.

As shown in FIGS. 1 to 3, the document recognition / voice reading method through IMT-2000 according to the present invention is a document recognition / voice reading method through IMT-2000, which is installed in an IMT-2000 (20) terminal. Step S10 of photographing the document using the camera 21; Step S20 of black and white the image input through the IMT-2000 (20), and binarizing the image to input a document image; In the document input as the image, it is divided by category, and the image of the character is interpreted therein, the region is separated from the analyzed image, the character is separated from the separated region, and the separated character is recognized and confirmed by the user. Step S30 of outputting a result to the terminal display panel 22 of the IMT-2000 20 to indicate that the character is recognized to be recognized; When an error occurs in the result output to the display panel of the terminal of the IMT-2000 (20) is performed to perform the step S40 to correct it.

Step S50 of selectively transmitting data to the main computer 10 or storing the recognition result in its own database when there is no error in the result output to the terminal display panel 22 of the IMT-2000 (20); Step S60 of recognizing each letter of the document stored in the database and combining the same into a corresponding wave (WAV) file; In operation S70, the data formed from the wave (WAV) file is combined and read out as a voice.

Step S60 is a step S61 of encoding the characters of the data input through the image camera 21 installed in the terminal of the IMT-2000 (20) converted into a text document; Step S62 of classifying the coded characters; Step S63 of determining which character is the coded character; Step S64 of selecting which character the coded character corresponds to; Step S65 of combining the respective coded character wave (WAV) files; By performing the S66 step of expressing the combined wave (WAV) file as a voice through the speaker 23, the contents of the document can be notified to the elderly or blind people who cannot see well.

Referring to the operation of the document recognition / speech reading method through the IMT-2000 according to the present invention performing the above steps are as follows.

In order to capture the contents of the paper document desired by the user, the image camera 21 installed in the terminal of the IMT-2000 20 is positioned at the position to be photographed, and the contents of the paper document are photographed.

Monochrome the text image photographed by the image camera 21, and binarize and input the text image, and then interpret the text image, separate regions by category from the analyzed images, and separate characters only from the separated category regions. It recognizes this, and outputs the result to the display panel 22 of the IMT-2000 (20) terminal.

As described above, when an error occurs in the text using a key installed in the IMT-2000 (20) terminal, the user directly corrects the error or photographs it again using the image camera 21 to correct the error. When an error does not occur in the text, the display is displayed on the display panel so that the user knows that there is no error.

Then, the result output on the display panel of the IMT-2000 (20) terminal is wirelessly connected to the server of the operator supplying the main computer 10 or the IMT-2000 (20) determined by the user to the IMT-2000 (20). Allows you to send data stored in the database so that you can continue to enter documents. At this time, the file of the transmitted document is converted into a text document so that it can be transmitted or stored.

Check the code of each character in the data converted into the text document, classify it by code, determine the character that fits the code, select the wave of the determined character code, and combine the selected wave files This is expressed in voice (sound) through a speaker installed in the IMT-2000 (20) terminal so that the elderly or blind people who cannot see well can grasp the contents of the document immediately. In particular, when copying certain contents in the library, the contents are captured and stored through the image camera 21 of the IMT-2000 (20) terminal, and when the capacity of the database is insufficient, the main computer (10) pre-determined by the user. The content can be sent with) so that a large amount of information can be collected.

Looking at the effect of the document recognition / voice reading method through the IMT-2000 according to the present invention, the elderly using the terminal of the IMT-2000 can recognize the text and can be read out by voice Because blind people can read simple documents, they can handle financial transactions, contracts, etc. by themselves without the help of others, and keep the documents as text files to share data with other terminals and computers wirelessly or by wire. The effect of being able to obtain a large amount of information can be obtained.

Claims (2)

In the document recognition method, Step S10 of photographing a document using an image camera installed in a terminal of the IMT-2000; S20 step of black and white, and binarizes the image input through the IMT-2000 and inputs the document image; In the document input as the image, it is divided into categories, and the image of the text is interpreted therein, the region is separated from the interpreted image, the character is separated from the separated region, and the character is recognized by the separated text. Step S30 of displaying the result on the terminal display panel of the IMT-2000 to indicate that the character is recognized; Step S40 for correcting when an error occurs in the result output to the display panel of the terminal of the IMT-2000; Step S50 of selectively transmitting data to a computer or storing the recognition result in its own database when there is no error in the result output to the terminal display panel of the IMT-2000; Step S60 of recognizing each letter of the document stored in the database and combining the same into a corresponding wave (WAV) file; And The document recognition / voice reading method of IMT-2000, comprising the step of combining the data formed by the wave (WAV) file and reading it out as a voice. The method of claim 1, The step S6 includes the step S61 of encoding characters of data inputted through a video camera installed in the terminal of the IMT-2000 and converted into a text document; Step S62 of classifying the coded characters; Step S63 of determining which character is the coded character; Step S64 of selecting which character the coded character corresponds to; Step S65 of combining the respective coded character wave (WAV) files; And The document recognition / speech reading method through IMT-2000, characterized in that it further comprises the step S66 of expressing the combined wave (WAV) file through a speaker.