JP2001111910A - TV video character information extraction conversion device and emergency telop character recognition conversion display device - Google Patents

Abstract

(57) [Summary] [Problem] To extract character information in a TV video that flows instantaneously, it is necessary to store information in advance before the user selects information to be extracted, and the image is stored as it is. And to reduce the storage capacity from being enormous. SOLUTION: A character region is extracted from a TV video image by a telop region extraction unit or a flip region extraction unit, a character region image is compared by an extraction image comparison determination unit, and only a new image is stored in an image storage unit, and character recognition is performed. A TV video character information extraction / conversion device characterized by converting an extracted image into text data using a means.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a TV video character information extracting / converting apparatus which extracts and stores character information on a screen of a digital television or the like, and performs processing as necessary.

[0002]

2. Description of the Related Art Conventionally, when character information in a television image is taken out, information has been searched for while being stored and rewound and played back as a video, or an image containing characters has been stored.

[0003]

In the above-mentioned method, it is necessary to prepare for extracting and storing information in advance, such as setting a video. There was a problem that it took too much time to memorize.

[0004] In order to retrieve instant information without leaking,
Before the user selects the information to retrieve, the information must be stored in advance. However, if the image is stored as it is, the storage capacity may become enormous.

An object of the present invention is to realize a system that extracts only necessary information as much as possible to solve such a problem.

An object of the present invention is to extract a character area for storing necessary information, compare character area images, store only a new image, and convert the new image into text data using character recognition.

[0007]

In order to solve the above-mentioned problems, a TV video character information extracting / converting device extracts a character region from a TV video image by a telop region extracting unit or a flip region extracting unit, and extracts and compares the extracted image. Wherein the character area images are compared, only the new image is stored in the image storage means, and the extracted image is converted into text data using the character recognition means.

[0008]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of a TV video character information extracting and converting device and an emergency broadcast character information converting device of the present invention will be described below with reference to the drawings.

(Embodiment 1) FIG. 1 is a block diagram of a TV video character information extraction / conversion device 110 in a digital TV and an analog TV. 102 is a telop area extraction unit, 103 is a flip area extraction unit, 104 is an extracted image comparison determination unit, 105 is an image storage unit that stores extracted images, 106 is a character recognition target selection unit, 1
Reference numeral 07 denotes character recognition means, reference numeral 108 denotes text data storage means, and reference numeral 109 denotes text data editing / processing means.

The video storage unit 101 stores a video signal input from the outside. Specifically, it is a frame buffer or the like.

[0011] The telop area extraction unit 102 is a video storage unit 1.
The extracted image is extracted, a telop area in the image is extracted, and the extracted image is sent to the extracted image comparison / determination unit 104.

The flip region extracting unit 103 converts the image extracted from the video storage unit 101 into a telop region extracting unit 10.
If no telop area is extracted in step 2, a flip area is extracted from the image extracted from the video storage unit 101, and the extracted image is sent to the extracted image comparison / determination unit 104.

The extracted image comparison / determination means 104 includes a telop area extraction unit 102 or a flip area extraction unit 10
3 is compared with the latest image stored in the image storage unit 105 to determine whether the two images are different.

The image storage means 105 stores an image determined by the extracted image comparison / determination means 104 to be different from the latest image stored in the image storage means 105.

The character recognition target selection means 106 takes out the image stored in the image storage means 105, displays the image at a specific location on the screen with a number attached thereto, and inputs the number to the input means such as a remote controller or a keyboard. By inputting a selection number using, an image to be subjected to character recognition is selected, and the selected image is sent to character recognition means.

The character recognizing means 107 performs character recognition on the image selected by the user by the character recognizing target selecting means 106,
Convert to text data.

The text data storage means 108 stores the text data converted by the character recognizing means 107 together with an identification ID.

The text data editing / processing means 109 includes:
In response to an input from a remote controller, a keyboard, or the like, the text data stored in the text data storage unit 108 is extracted, and the extracted text data is displayed on an editing screen displayed on a TV screen to perform editing. Alternatively, the extracted data is converted into another language, or the difference is detected by pattern matching with another text.

The means for extracting character information on a TV video in the TV video character information extraction and conversion device 110 according to the present embodiment will be described below with reference to the flowchart of FIG.

(Step 201) The image taken out from the video storage unit 101 is taken out, and the telop area extraction unit 10
Send to 2. Identification numbers are assigned in the order in which the telop area extracting means 102 enters the image received, and the telop area in the image is extracted. The values of the identification numbers are added one by one. After a certain time, the value of the identification number is reset.

(Step 202) It is determined whether or not a telop area has been extracted by the telop area extraction means 102. When the telop area is extracted, the extracted image and the identification ID are sent to the extracted image comparison / determination means 104. The identification ID includes the identification number assigned in step 201 and the image attribute. A value of 0 is entered in the image attribute, and a value of 1 is entered when a telop area is extracted. If no telop image has been extracted, the flip area extraction means 103
The original image and the identification ID. At this time, the image attribute is sent as 0.

(Step 203) Using the flip region extracting means 103, the original image sent from step 202 is used to extract a flip region in the image.

(Step 204) When the flip image is extracted in Step 203, the extracted flip image and the identification ID are sent to the extracted image comparison / determination means 104. Identification I
D is composed of an identification number assigned in step 201 and an image attribute having a value of 2 indicating a flip area in the image attribute. If the flip image is not extracted, the original image and the identification ID are discarded without being stored.

(Step 205) The extracted image comparison / determination means 104 compares the image transmitted in step 202 or step 204 with the latest stored image (the image with the largest attached ID) stored in the image storage means 105. . If it is determined that the comparison images are different, step 2
02 or the image transmitted in step 204 is stored in the image storage unit 105 together with the identification ID. If the comparison images are determined to be the same, step 20
2, or delete the image transmitted in step 204 from the data.

(Step 206) In step 205, the extracted image comparison / determination means 104 switches to the image storage means 105
Is stored together with the attached ID (the stored image is deleted after a certain period of time. If the storage capacity exceeds the allowable range, the images are deleted in order from the oldest image).

(Step 207) Numbers are assigned to the images stored in the image storage means 105 (when the user watching the TV inputs a character recognition target image selection input using a remote controller or a keyboard, etc.) Is displayed on the TV screen. The user inputs the number of the selected character recognition target image using a remote controller, a keyboard, or the like. The ID of the image corresponding to the input number is transmitted to the character recognition unit 107, and the selected image is stored in the image storage unit 1.
05. The unselected extracted images are deleted.

(Step 208) The image corresponding to the image ID transmitted in step 207 is taken out from the image storage means 105, and is subjected to character recognition and converted into text data. A text data ID is newly assigned to the text data. The text data and the attached ID are sent to the text data storage means 108.

(Step 209) The text data and the ID transmitted in step 208 are stored in the text data storage means 108.

(Step 210) When requested by the user, the data stored in the text data storage means 108 is taken out, displayed on an editing screen, edited, or changed by a program, and a new ID is assigned to the text. The data is stored in the data storage unit 108.

FIG. 3 shows an original image and an example of the extraction as a specific example of extracting a telop area from the image data sent from the video storage unit 101.

Reference numeral 301 denotes an example of an original image including a telop area, reference numeral 302 denotes an example of a detection area when a high-brightness area of the original image 301 is detected, and reference numeral 303 denotes an edge of the telop area by detecting a high-frequency area of the original image 301. In the figure where the part is extracted, 3
04 is a combination of 302 and 303, 305 is 3
4 shows a telop area extracted from the telop area 04.

The means for extracting a telop area on a TV video in the telop area extraction unit 102 according to the present embodiment will be described below with reference to the flowchart of FIG.

(Step 401) Pixels having a luminance equal to or higher than the threshold value a are extracted from the pixels of the image data sent from the video storage unit 101, and the extracted pixel area is stored in the telop temporary storage unit 411.

(Step 402) The high frequency region of the image data sent from the video storage unit 101 is extracted, and the edge portion of the telop image is extracted.

(Step 403) The output results of steps 401 and 402 are weighted, each pixel is added, an evaluation value is calculated, and a telop area is extracted based on the evaluation value.

FIG. 5 shows a specific example of an original image extracted when a flip area is extracted from the image data sent from the telop area extraction unit 102 and the extracted area.

Reference numeral 501 denotes a specific example of the original image data sent from the telop area extraction unit 102.

Reference numeral 502 denotes an example of an extraction area when a light color area is extracted from 501.

Reference numeral 503 denotes an example of a segment group including 502 light-colored areas when the entire screen is divided into several segments.

Reference numeral 504 denotes a flip region determination target 50.
3 shows the center segment of the segment group extracted in FIG.

Reference numeral 505 denotes a segment which is in contact with the segment angle of 504 when the target segment of 504 is determined to be a flip area, and a flip area is determined by sampling a segment belonging to the segment group of 503 as the next flip area determination target. Indicates the target segment.

Reference numeral 506 denotes a segment which is in contact with the segment 504 and which is surrounded by the segment 505 and which belongs to the segment group 503 when the target segment 505 is determined to be a flip area. This shows the target segment in the case where it is performed.

Reference numeral 507 denotes a flip region determination when a segment belonging to the 503 segment group which is not yet a flip region determination target is sampled as a next flip region determination target when the target segment 506 is determined to be a flip region. Indicates the target segment.

Reference numeral 508 denotes a flip region finally extracted when the flip region determination segment 507 is not determined to be a flip region.

The means for extracting a flip area on a TV video in TV video character information extraction and conversion apparatus 110 according to the present embodiment will be described below with reference to the flowchart of FIG.

(Step 601) Telop area extraction unit 1
In the image data sent from 02, a region of a pixel whose RGB value of each pixel and the sum thereof are equal to or larger than a predetermined threshold is taken.

(Step 602) The screen is 6 × 6 for a total of 48
Divide the segment into segments.

(Step 603) The group of segments in S602 including the region extracted in S601 is extracted.

(Step 604) Four segments located at the center of the segment extracted in S603 are taken out as targets for determining whether or not the region is a flip region, and the average and variance of RGB of each segment are examined (and , RGB, the average value is set as a representative color 1, and the color having the highest frequency after this representative color is set as a representative color 2.)

(Step 605) It is determined whether or not the average / variance value extracted in S604 is equal to or smaller than a threshold value.

(Step 606) If the determination in S605 is equal to or smaller than the threshold value, the segment group in S604 is stored as a flip area in the flip area storage means.

(Step 607) If the difference between the average values of the segments extracted in S604 is equal to or smaller than the threshold value, the area is determined to be the same, and if it is larger than the threshold value, it is determined to be another area. If there are two or more areas, a group number is assigned.

(Step 608) A segment that is in contact with a corner of the flip area extracted in S606 and has not been extracted is selected as a sample.

(Step 609) The average and variance of the pixels of the segment selected in S608 are examined.

(Step 610) The same operation as in steps S605 to S607 is performed.

(Step 611) The average of the pixels of the segment surrounded by the segment newly extracted in S610
Take dispersion.

(Step 612) The operations of S605 to S607 are performed.

(Step 613) It is determined whether the average and variance of all the segments in the segment area extracted in S603 have been checked. If there is any remaining area, the operation from S608 is repeated. When the check is completed, if there are two or more flip areas, group numbers 1, 2, 3 are assigned in order from the one to which the segment belongs. If there is only one flip area, the group number is assigned 0. A value of 2 indicating a flip area is put in the image attribute of the identification ID.

The identification number, image attribute, and group number assigned in S201 are transmitted as identification IDs to the extracted image comparison / determination means 104 together with the extracted image. The operation of the flip area detection ends.

An image extracted by the telop area extraction unit 102 or the flip area extraction unit 103 and compared with the image stored in the image storage unit 105 to determine whether or not they are the same. Comparison determination means 1
The operation of Step 04 will be described with reference to the flowchart of FIG.

(Step 701) Telop area extraction unit 1
02 or the image and the identification ID sent from the flip area extraction unit 103 are received.

(Step 702) Among the identification IDs, the one with the largest value of the identification number is taken out from the image storage means 105 as the ID of the latest stored image.

(Step 703) The value of the image attribute of the identification ID received in S701 and the identification ID extracted in S702
It is determined whether the values of the image attributes are the same. If it is determined that they are the same, the procedure goes to S704, and if they are different, the procedure goes to S708.

(Step 704) The image corresponding to the identification ID extracted in S702 is extracted from the image storage means 105.

(Step 705) The difference between the pixel of the image received in S701 and the pixel of the image extracted in S704 is obtained at a predetermined sampling point.

(Step 706) The difference value of each sample point calculated in S705 is squared and added.

(Step 707) It is determined whether or not the sum of the squares of the difference values of each sample taken in S706 is larger than a threshold value. If it is larger than the threshold, the two images are regarded as different images, and the process proceeds to S708. If it is smaller than the threshold value, the two images are regarded as the same image, and the image and the identification ID received in S701 are discarded.

(Step 708) The image and the identification signal received in S701 are stored in the image identification means 105.

FIG. 8 shows a configuration example of the identification ID. The identification number is a number assigned to each image in S201, and the newer the image, the larger the value. However, the value is reset at regular intervals. The image attribute indicates whether the image is an original image, a telop image, or a flip image. The value takes 0 when the image is the original image, 1 when the image is a telop image, and 2 when the image is a flip image. The group number is a number assigned to each of a plurality of flip areas on the same image by the flip area extracting means. When the flip region is single, the value is 0.

FIG. 9 shows a screen display example of the character recognition target selecting means 106. Numbers are assigned to the images stored in the image storage unit 105 in order from 1 and the numbers and the reduced images are displayed on the screen.

FIG. 10 shows a case where an emergency telop broadcast is played.
4 shows a telop broadcast screen and an output screen after telop recognition. The upper and lower areas show examples of earthquake warning telop and flood warning telop, respectively.

The upper left figure of each area shows the original image of the telop broadcast, the lower figure shows an example of a screen display of symbols and English characters converted according to the contents of the telop broadcast, and the upper right figure shows a telop broadcast. An example of a screen display of a place and a place name of an appearing area converted into Roman characters is shown. The color of the frame of the conversion screen corresponds to the urgency of the telop content.

The process of recognizing an emergency text broadcast in the case where an emergency telop broadcast is entered by the TV video text information extraction / conversion device of FIG. 1 will be described with reference to the flowchart of FIG.

(Step 801) When the emergency telop broadcast speech recognition means recognizes the emergency telop broadcast start speech, the procedure goes to step 802.

(Step 802) Because of the feature that the emergency telop broadcast is displayed only on the upper part of the television screen, a one-third area image is extracted from the image sent from the video storage unit 101 in FIG. The telop portion is extracted from the extracted image by using the telop area extracting means 102.

(Step 803) The extracted image is sent to the emergency broadcast image storage means together with the identification ID given in S201. The identification ID is put in the character recognition processing waiting queue.

(Step 804) The identification IDs are extracted from the character recognition processing standby queue in the order of FIFO, and the corresponding images are extracted from the emergency broadcast image storage means.
Characters in the image are recognized using the character recognition unit 107 and converted into text data. A text data ID is assigned to the extracted text data. The text data and the text data ID are stored in text data storage means.
D is put into a symbol conversion processing standby queue. Character recognition means 1
As soon as the process 07 is completed, the ID in the character recognition processing standby queue is taken out again, the corresponding image is taken out from the emergency broadcast image storage means, and the process is continued.

(Step 805) Text data IDs in the symbol conversion processing standby queue are fetched in FIFO order, and the corresponding text data is fetched from the text data storage means. Text data and emergency terms
Emergency terms and regional terms in the text data are extracted by pattern matching with terms in the regional term database. At the same time, if a term indicating a danger such as “alarm” or “advisory” is detected, the danger is set to “1 (safety)” or “2”.
(Needs attention) It is classified into four levels, "3 (slightly dangerous)" and "4 (very dangerous)." Even if a numerical value is attached such as "seismic intensity 4," the degree of risk is determined by the numerical value The text data ID, the group ID, the English characters, the urgent term / regional term, and the risk are referred to as an urgent term.
Store in local term storage means. If several region names are detected, group IDs are sequentially assigned to each region from 1
Allocate. If there is only one area name, a value of 0 is entered for the group ID.

(Step 806) The design, English character, and local graphic corresponding to the emergency term and the local term extracted in S805 are extracted from the design, English character, and local graphic database.

(Step 807) The symbols and English characters extracted in S806 are displayed on the screen as shown in FIG.

The symbol is surrounded by a frame, and the color of the frame corresponds to the degree of danger. For example, the colors are classified into blue (1 = safe), green (2 = needs attention), yellow (3 = somewhat dangerous), and red (4 = very dangerous). The frame blinks from level 2, and the blinking speed increases as the degree of danger increases. Further, the screen display of the symbols and English characters, the map showing the dangerous area and the screen display of the area name are alternately performed at regular time intervals as shown in FIG.

If the group ID is 1 or more in step S805, the screen display method of the symbols and English characters is the same, and the screen display of the dangerous area displays all areas collectively on one map. The danger areas are displayed in colors corresponding to the respective danger levels. In this case, the frame provides a display corresponding to the highest degree of danger in the extracted danger area.

FIG. 12 is a configuration diagram of the emergency telop character recognition conversion display device. Reference numeral 916 denotes an emergency telop character recognition conversion display device. Emergency telop character recognition conversion display device 9
Reference numeral 901 which constitutes 16 is an emergency telop speech recognition means;
902 is an emergency broadcast image storage unit, 903 is a telop area extraction unit, 904 is an emergency broadcast telop image storage unit, 90
5 is a character recognition processing standby queue, 906 is character recognition means,
907 is a text data storage unit, 908 is a symbol conversion processing standby queue, 909 is an emergency term detection unit, 910 is an emergency term database, 911 is an emergency term / region term storage unit, 912 is a design / English character / region figure extraction unit, 91
3 is a symbol / English character / region graphic storage means, 914 is a converted image screen display means, and 915 is a symbol / English character / region graphic database.

The emergency telop speech recognition means 901 recognizes emergency telop speech. When the emergency telop voice is recognized, the video storage unit 101 is instructed to transmit the image to the emergency broadcast image storage unit 902.

The emergency broadcast image storage means 902 takes out one third of the area image from the image sent from the video storage section 101 and stores it.

The telop area extraction unit 903 extracts the telop area, transmits the identification ID given in S201 and the telop area extracted by the telop area extraction unit 903 to the emergency broadcast telop image storage unit 904, and outputs the emergency broadcast telop image. The storage unit 904 stores the identification I given in S201.
D and the telop area extracted by the telop area extraction unit 903 are accumulated.

The character recognition standby queue 905 stores the identification IDs in the order in which they were sent, and when there is a request from the character recognition means 906 to retrieve them, retrieves them in the order in which they were stored.

The character recognition means 906 takes out the identification ID from the character recognition processing standby queue 905, takes out the corresponding image from the emergency broadcast telop image storage means 904,
Perform character recognition and convert to text data.

The text data storage means 907 stores text data output from the character recognition means 906 and its text data ID.

The symbol conversion processing standby queue 908 stores the text data IDs in the order in which they were sent, and retrieves them in the order in which they were stored.

The emergency term detecting means 909 takes out the text data ID from the symbol conversion processing standby queue 908 and stores the corresponding text data in the text data storing means 9.
07, the extracted text data and the terms in the emergency term database are searched and extracted by pattern matching.

The emergency term database 910 is a database in which emergency terms, area names, and numerical values frequently appearing in telop emergency broadcasts are collected.

The emergency term / region term storage unit 911 stores the text data ID, the group ID, the emergency term / region term, and the risk in the emergency term / region term storage unit.

The pattern / English character / region graphic extracting means 912 retrieves data from the emergency term / region term storage means 911 in the order of FIFO, and converts the pattern / English character / region graphic corresponding to the retrieved data into the pattern / English character / region. Detect from the figure database.

The design / English character / region graphic storage means 913 stores the design / English character / region graphic detected by the design / English character / region graphic extraction means 912.

The converted image screen display means 914 displays the design / English character / region graphic stored in the design / English character / region graphic storage means 913 on the screen.

Pattern / English / Regional figure database 91
Reference numeral 5 denotes a database in which symbols, English characters, and regional figures corresponding to the emergency terms in the emergency term database 910 are collected.

[0098]

As described above, the present invention has an excellent effect that, by extracting only necessary information, information can be stored without leaking even with a limited storage capacity.
It is intended to provide a V-video character information extraction and conversion device.

[Brief description of the drawings]

FIG. 1 is a configuration diagram of a video character information extraction device that is an embodiment of a video character information extraction device according to the present invention.

FIG. 2 is a flowchart for explaining the operation of the video character information extraction device.

FIG. 3 is a diagram showing a specific example of an extracted image at each stage of extracting a telop area from an original image including a telop area in a telop area extraction unit in the video character information extraction apparatus according to the present invention.

FIG. 4 is a flowchart illustrating an operation of extracting a telop area in the video character information extraction device according to the present invention.

FIG. 5 is a diagram showing a specific example of an extracted image at each stage of extracting a flip region from an original image including a flip region in the flip region extraction unit in the video character information extraction device according to the present invention.

FIG. 6 is a flowchart illustrating an operation of extracting a flip region in the video character information extraction device according to the present invention.

FIG. 7 is a flowchart illustrating an operation of determining whether the latest accumulated image and the extracted image are the same or different in the video character information extracting apparatus according to the present invention.

FIG. 8 is a diagram showing a configuration example of an identification ID used in the video character information extracting device according to the present invention.

FIG. 9 is a diagram showing a display example when an image to be subjected to character recognition is displayed on a screen in the video character information extracting apparatus according to the present invention.

FIG. 10 is a diagram showing a display example of an emergency telop broadcast and an example of a screen display for converting the emergency broadcast into symbols and symbols in the video character information extracting apparatus according to the present invention.

FIG. 11 is a diagram illustrating an example of a video character information extracting apparatus according to the present invention.
Flowchart for explaining the operation of converting the pattern into English characters and displaying it on the screen

FIG. 12 is a configuration diagram of an emergency telop broadcast character recognition device in the video character information extraction device according to the present invention.

[Explanation of symbols]

Reference Signs List 101 video storage unit 102 telop region extraction unit 103 flip region extraction unit 104 extracted image comparison determination unit 105 image storage unit 106 character recognition target selection unit 107 character recognition unit 108 text data storage unit 109 text data editing / processing unit 110 TV video characters Information extraction conversion device 301 Example of original image including telop area 302 Example of detecting high-luminance part of 301 303 Example of detecting edge part by taking high-frequency part of 301 301 Example of combining results of 302 and 303 305 304 Example of telop area extracted from the result 501 Example of original image including flip area 502 Example of extracting light color area of original image 501 501 Example of dividing image and detecting segment including light color area 504 To detect flip area Center point Example of the next segment when the center segment of 505 504 is determined to be the flip region Example of the next sampling segment when the 505 505 sampling segment is determined to be the flip region 507 The sample segment of 506 is flipped Example of the next sampling segment when it is determined to be a region 508 Example of a flip region extracted when the sampling segment of 507 is not determined to be a flip region 901 Telop voice recognition means 902 Emergency broadcast image storage means 903 Extraction of telop area Part 904 emergency broadcast telop image storage means 905 character recognition processing standby queue 906 character recognition means 907 text data storage means 908 symbol conversion processing standby queue 909 emergency term detection means 910 emergency term data Database 911 Emergency term / regional term storage means 912 Design / English character / regional figure extraction means 913 Design / English character / regional figure storage means 914 Conversion image screen display means 915 Design / English character / regional figure database 916 Emergency telop character recognition conversion Display device

Claims (5)

[Claims] 1. A video storage unit for storing a video signal, a telop region extraction unit for extracting an image of a region composed of telops from the video signal, and whether there is no region to be extracted by the telop region extraction unit And a flip region extracting means for extracting an image of a flip region from the video signal when the region does not exist, an image of the region extracted by the telop region extracting device or an image of the region extracted by the flip region extracting device And an output pixel which is a pixel of the image of the area extracted by the telop area extracting means or a pixel of the image of the area extracted by the flip area extracting means, and the image stored in the image storing means. Take the difference with the latest pixel, which is the pixel of the most recently stored image, and if the sum of the absolute values of the difference values is greater than the threshold, An extracted image comparison / determination unit that determines that the latest pixel is different, and instructs to store the output pixel; and a character recognition target that selects a character recognition target from the images stored in the image storage unit. A TV video character information extracting / converting device comprising: a selecting unit; a character recognizing unit for recognizing a character of the selected image to convert it to text data; and a text data storing unit for storing the text data. 2. The image processing apparatus according to claim 1, wherein the flip region extracting unit detects a region composed of white or light color pixels from the pixels of the video signal, and detects a pixel level of the video signal in the region detected by the detecting unit. 2. The TV video character information extracting and converting apparatus according to claim 1, wherein the flip area is extracted by using a uniform level area detecting means for detecting a substantially uniform level area by comparing peripheral pixel levels. 3. The telop area extracting means includes: a high-frequency area detecting means for extracting a pixel having a frequency component higher than a threshold value in an arbitrary image plane of an image signal and detecting a high frequency band; A uniform level area detecting means for comparing a level and a peripheral pixel level to detect a substantially uniform level area, and an extracting means for extracting a specific area based on each output of the high frequency area detecting means and the uniform level area detecting means. The TV video character information extraction and conversion apparatus according to claim 1, further comprising: 4. An emergency telop broadcast voice recognition means for recognizing the voice of the emergency telop broadcast, and a telop voice reception communication indicating that the voice has been recognized when the voice is recognized by the emergency telop broadcast voice recognition means. A telop area extracting means for extracting a telop area according to claim 3 for an upper layer portion of a screen where the telop emergency broadcast flows when it is transmitted that the telop emergency broadcast has entered by the telop voice receiving communication means; The image of the telop area extracted by the telop area is subjected to character recognition using the character recognition means and converted into text data, and the text data is compared with the emergency term database to extract emergency terms and regional terms from the text data. , Patterns and English characters composed of extracted emergency terms and their corresponding images, English characters and regional figures The regional graphic database, the character recognition means and the pattern / English / local graphic database are compared and converted into corresponding images, English characters and regional graphics, and terms representing the degree of urgency are extracted and stored as one group. A pattern / English character / region graphic storage means; and a converted image screen display means for displaying the image / English word / region graphic stored by the pattern / English character / region graphic storage means in a frame on a screen. 2. The TV video character information extracting and converting apparatus according to claim 1, wherein: 5. The conversion image screen display means according to claim 4, wherein said conversion means displays a color corresponding to the degree of danger.
2. The TV video character information extracting / converting apparatus according to claim 1, further comprising: a blinking speed means for associating the degree of risk with a blinking speed with yellow for attention and red for danger.