JPH06309733A - Reserving device for video recording and recognizing device for contents of video - Google Patents

Abstract

PURPOSE:To enable reserving of video recording of a desired program by recognizing contents of a broadcast of a program from video information and delaying a start time and a finish time of recording operation if it is not contents of a broad-cast previously at the time when the prescribed time comes. CONSTITUTION:For example, when a program is a movie from 9 o'clock to 11 o'clock and it is assumed that a relay broadcast of a baseball is extended to 9:30, since a recording start time set to a register 8 and an output of a timepiece circuit 9 coincide with each other, a recording start command is outputted from a timer 6 at the time point of 9 o'clock and inputted to a recording time control means 16. The recording time control means 16 waits until information from a video contents recognizing means 14 becomes information other than a baseball, and when information becomes information other than a baseball at 9:30 of thirty minutes after, the recording start command is outputted to a recording section. Also, a recording finish command is outputted from the timer 6 at 11 o'clock, but it is delayed by thirty minutes in the same way as the recording start time, the recording finish command is outputted to the recording section 5 at 11:30.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a video recognition method for recognizing video content from various characteristics of the video content, and a recording reservation apparatus using the video recognition method.

[0002]

2. Description of the Related Art FIG. 9 is a block diagram of a conventional recording reservation device. In the figure, TV radio waves from a TV broadcasting station are received by an antenna 1 and amplified by an intermediate amplifier 3 via a high frequency amplifier / mixer 2. The intermediate frequency signal from the intermediate amplifier 3 is further detected by the detection circuit 4, and a video signal and an audio signal are obtained and input to the recording unit 5.

When a TV program is recorded by the recording reservation function, if information such as start time, end time, channel, recording mode of the TV program to be recorded is set in the timer section 6,
The recording unit 5 is configured to automatically start the recording of the TV program of the set channel at the setting start time in the setting mode even when not in the absence, and automatically end the recording operation at the setting end time. Give orders.

The operation of the timer unit 6 will be described in detail. The timer setting unit 7 sets the start / end time information of the TV program, the recording mode and the channel in the register 8. On the other hand, the time information at the present time is generated in the clock circuit 9, and this time information is compared with the start / end time information from the register 8 in the comparison circuit 10. When the output of the clock circuit 9 matches the start time information, a recording start command in the set recording mode and channel is output to the recording unit 5, and when the output of the clock circuit 9 matches the end time information, the recording end command is issued. Is output to.

Further, recently, there is a device in which the setting of the recording start / end time and the channel is coded and the timer can be easily set by inputting a numeral.

[0006]

However, in these cases, since the recording of the TV program in the absence is solely dependent on the timer recording function, after the TV program recording by the function is reserved and set. If, for some reason, the program start / end time is changed in the program program, the change cannot be dealt with.

[0007] In particular, in the case of a baseball relay, if the game does not end,
Programs are often extended. When the baseball broadcast is extended, later programs will be delayed by the extension time and broadcast. Therefore, if you reserved a recording of a movie or drama that is a program after the baseball broadcast, if the baseball broadcast is extended, the baseball broadcast for the extended time will be actually recorded,
On the contrary, it means that the last extended time of the program reserved for recording is not recorded.

This situation will be specifically described with reference to the example of FIG. Broadcast schedule is from 7:30 to 9 as shown in Fig. 10 (A).
When the movie is reserved from 9 am to 11 pm and the movie is reserved from 9 am to 11 pm, the recording start time is set to 9:00 and the recording end time is set to 11:00 as shown in FIG. . At this time, when the baseball relay is extended by 30 minutes, the actual broadcast content is postponed from 9:30 to 11:30 in the baseball relay until 9:30, as shown in FIG.
Therefore, the video content actually recorded on the tape is shown in FIG.
As shown in (D), there was a problem that the baseball broadcast was recorded for the first 30 minutes, and the important movie was not recorded for the last 30 minutes, which is the climax.

The present invention has been made to solve the above-mentioned problems, and can record a program such as extension from the content recognized by the video content recognition device and the program information set in the recording reservation. The purpose is to provide a reservation device.

Another object of the present invention is to provide a video content recognizing device which extracts some feature from a video signal with a small circuit scale and recognizes a genre or content of a program with high accuracy.

[0011]

According to a first aspect of the present invention, there is provided a recording reservation apparatus, a feature detecting means for detecting a feature of a video signal, a knowledge base means having knowledge about a feature of a video content, and a detected feature. And a knowledge base, a video content recognizing means, a program information setting means for setting information about a program to be recorded, a recognition result of the video content recognizing means, and the program information setting means. A recording time control means for collating the set information and controlling the start and end times of the recording operation is provided.

A video content recognition apparatus according to a second aspect of the present invention is a feature detection means for detecting a feature of brightness pattern information of a video signal, a knowledge base means having knowledge about the brightness pattern information, and a detected feature and knowledge base. And a feature matching means for matching with.

According to a third aspect of the present invention, there is provided a video content recognizing device, which comprises a feature detecting means for detecting a feature of chromaticity pattern information of a video signal, a knowledge base means having knowledge about the chromaticity pattern information, and a detected feature. It is provided with a feature matching means for matching with a knowledge base.

A video content recognition apparatus according to a fourth aspect of the present invention comprises a feature detection means for detecting a feature of a high frequency component of a video signal or edge pattern information, and a knowledge base means having knowledge of a high frequency component or edge pattern information.
A feature matching means for matching the detected feature with the knowledge base is provided.

A video content recognizing device according to a fifth aspect of the present invention is a feature detecting means for detecting a feature of edge pattern information of a telop display inserted in a video signal, and knowledge about the inserted edge pattern information of the telop display. And a feature matching unit for matching the detected feature with the knowledge base.

A video content recognizing apparatus according to a sixth aspect of the present invention is a feature detecting means for detecting a feature of vertical edge pattern information of a telop display inserted in a video signal, and a feature detecting means for inserting a telop display in the vertical direction. A knowledge base means having knowledge about edge pattern information and a feature matching means for matching the detected feature with the knowledge base are provided.

According to a seventh aspect of the present invention, there is provided a video content recognizing device which detects a feature of a relative positional relationship of edge pattern information of a telop display inserted in a video signal, and an inserted telop display edge. A knowledge base means having knowledge about the relative positional relationship of the pattern information and a feature matching means for matching the detected feature with the knowledge base are provided.

The video content recognition apparatus according to the eighth invention is different from the fifth invention to the seventh invention in that the video content to be recognized is a baseball relay and the telop display for detecting a feature is "S".
These are “B” and “O”.

According to the ninth aspect of the invention, in addition to the means of the second to eighth aspects of the invention, once the image content is recognized, the recognition information is retained for a predetermined time even when the image content changes. The recognition information holding means is provided.

[0020]

According to the first aspect of the present invention, with the above configuration, the broadcast content of the program is recognized from the video information, and if the broadcast content is not preset even after a predetermined time, the broadcast content is set until the preset broadcast content is reached. Delay the start and end time of the recording operation by the time.

According to the second aspect of the invention, the characteristics of the luminance pattern information of the video signal and the characteristics held as knowledge of the luminance pattern information of the predetermined program in advance are collated.

According to the third aspect of the invention, the characteristics of the chromaticity pattern information of the video signal and the characteristics held as knowledge of the chromaticity pattern information of a predetermined program in advance are collated.

According to the fourth aspect of the invention, the characteristics of the high frequency component or edge pattern information of the video signal are compared with the characteristics that are previously held as knowledge of the high frequency component or edge pattern information of the predetermined program.

In the fifth aspect of the invention, the feature of the edge pattern information of the predetermined telop display inserted in the video signal is compared with the feature which is previously known as the knowledge of the edge pattern information of the telop display of the predetermined program.

According to the sixth aspect of the invention, the characteristics of the vertical edge pattern information of the predetermined telop display inserted in the video signal and the characteristics held in advance as the knowledge of the vertical edge pattern information of the telop display of the predetermined program are included. To match.

According to the seventh aspect of the present invention, as the knowledge of the relative positional relationship of the edge pattern information of the predetermined telop display inserted in the video signal and the relative positional relationship of the edge pattern information of the telop display of the predetermined program in advance. Match the features you have.

In the eighth invention, in the telop display,
In particular, the characteristics of "S", "B", and "O" of baseball broadcast are compared.

In the ninth aspect, when the characteristic image is collated, the collation result is held for a predetermined time.

[0029]

【Example】

Example 1. First Embodiment FIG. 1 is a block diagram of a recording reservation device according to a first embodiment of the present invention. The TV radio wave from the TV broadcasting station is received by the antenna 1 and is steeped by the steep wave circuit 4 via the high frequency amplifier / mixer 2 and the intermediate frequency amplifier 3 to obtain a video signal and an audio signal, and to the recording unit 5. Inputting, setting the start time of the TV program or the like from the timer setting unit 7 to the register 8, comparing the output of the clock circuit 9 and the output of the register 8 with the comparison circuit 10, and issuing a recording start command and a recording end command. The operation of the timer unit 6 that outputs is the same as that of the conventional example.

The difference from the prior art is that the feature detecting means 11 detects a feature from the video signal, the knowledge base means 12 has knowledge about the feature of the video content, and the feature detected by the feature detecting means 11 and the knowledge base means 12. When there is an instruction to start or end recording from the video content recognition means 14 including a feature matching means 13 for checking the characteristics of the above, a program information setting means 15 for setting information about a program to be recorded, and a timer section 6 Since the program information from the video content recognizing means 14 and the program information from the program information setting means 15 are collated with each other, the recording time control means 16 for delaying the recording start and recording end commands by the time until they match is added. is there.

FIG. 2 is a diagram for explaining the operation of the recording reservation apparatus according to the first embodiment. As in the case of the description of the conventional example, the broadcast schedule starts from 7:30 as shown in FIG. 2 (A). 9
The case where the baseball relay is up to the time and the movie is from 9:00 to 11:00 will be described. The desired channel, desired driving mode, recording start time at 9 o'clock and recording end time at 11:00 are the same as before, but in addition to these, the contents to be recorded are, for example, programs other than "baseball". The information is set by the information setting means 15.

At this time, if the actual broadcast is extended to 9:30 pm as shown in FIG. 2 (C), then 9
At the time point of time, the recording start time set in the register 8 and the output of the clock circuit 9 coincide with each other from the timer unit 6, so that a recording start command is output and input to the recording time control means 16.

On the other hand, since the baseball relay is extended for the actual broadcast content, the information of the video content recognition means 14 is "baseball".
Therefore, it does not match the information "other than baseball" from the program information setting means 15. Therefore, the recording time control means 16
Waits until the information from the image content recognizing means 14 becomes "other than baseball", and outputs a recording start command to the recording unit 5 when it becomes "other than baseball" at 9:30 after 30 minutes. Also,
Although the recording end command is output from the timer unit 6 at 11:00,
The recording is delayed by 30 minutes, which is the same time as the start of recording, and a recording end command for the recording unit 5 is output at 11:30.

As a result, as shown in FIG. 2D, desired reservation recording can be realized even if the broadcast program is changed due to extension.

Next, the image content recognition means will be described. In addition to baseball, there are various categories of video content such as various sports such as skiing, soccer, drama, news, weather forecast, etc. For the extraction of features, luminance signals and chromaticity signals are used. Alternatively, various combinations such as using predetermined frequency components or edges can be considered.

Example 2. As an example of the image content recognition means, an example of recognizing a baseball relay using the features of the edge pattern detected from the luminance signal will be described.

FIG. 3 is a block diagram of an image content recognition method according to the second embodiment of the present invention. In the figure, 301 is a register for delaying an input luminance signal for a predetermined time, 302 is a register.
Is a subtraction circuit for calculating the difference between the input brightness signal and the brightness signal delayed by the register 301, 303 is an absolute value circuit for calculating the absolute value of the output of the subtraction circuit 302, 30
Reference numeral 4 is a comparator for comparing the output of the absolute value circuit 303 with a reference level, 305 is a register for delaying the output of the comparator 304 for a predetermined time, 306 is pattern data having information on a telop display pattern of baseball relay, and 307 is A predetermined number of edge patterns and pattern data 3 delayed by a predetermined time unit by the register 305
It is a coincidence detection circuit for collating a predetermined number of edge patterns from 06.

Generally, in baseball relay, the score, progress,
Telop information such as count information is periodically displayed. Although the display position and display font are slightly different depending on the broadcasting station, almost the same telop information is displayed. When the telop information is mixed with the original video, the change in the brightness level becomes large at the position of the telop information and can be detected as an edge signal. A series of edge patterns of this telop display is stored in the pattern data 306.

When the difference between the input luminance signal and the luminance signal delayed by the register 301 is calculated by the subtraction circuit 302 and the absolute value of this output is obtained by the absolute value circuit 303,
A large output can be obtained where the change in the brightness level is large. An edge can be detected by comparing this output with the reference level by the comparator 304. This edge is delayed a predetermined number of times for a predetermined period of time, and this series of edge signals is input to the match detection circuit 307 as an edge pattern, which is matched with the edge pattern from the pattern data 306.

Example 3. Next, a case where a vertical edge pattern is used as a matching pattern will be described. FIG. 4 is a block diagram of an image content recognizing method according to the third embodiment of the present invention, and shows a configuration example of vertical edge pattern matching. In the figure, the subtraction circuit 302, the absolute value circuit 303, the comparator 304, and the coincidence detection circuit 307 are the same as those in FIG. The difference from FIG. 3 is that the registers 301 and 305 for delaying by a predetermined time are set to 1
It is to be replaced with the line memory 401 and the line memory 402 which delays by the horizontal period, and the pattern data 306 has edge pattern data in the vertical direction of telop display.

Generally, a received TV signal has a ghost component or is subjected to a low-pass filter by various processes. For such a signal, the edge component in the vertical direction can be detected more accurately than in the horizontal direction. Therefore, the line memory 401 detects a vertical edge by delaying one horizontal period, and the line memory 402 obtains a series of vertical edge patterns in which edge information is delayed by one horizontal period, and the vertical direction of the pattern data 306 is obtained. And the edge pattern information of.

Example 4. Next, an example of recognizing a baseball relay from the characteristics of the luminance signal will be described. FIG. 5 is a diagram for explaining a video content recognition device according to Embodiments 4 and 5 of the present invention, and shows one characteristic video pattern of a baseball broadcast. In the figure, an area A is a pitcher, an area B is a catcher, and an area C is a batter, which are often white uniforms. The area D is a ground area and is often green or brown on the lawn, and the area E is a back net area and often dark.

In such a video pattern, when projection processing is performed in the vertical direction on a pixel having a high luminance signal level, as shown in FIG. 5 (a), it is located at a position corresponding to the vertical direction in the areas A, B and C. High value. An example of recognition using this feature will be described below.

FIG. 6 is a block diagram of a video content recognition apparatus according to a fourth embodiment of the present invention. In the figure, 601 is the first
First for detecting a luminance signal higher than the reference level of
, 602 is a vertical direction projection circuit for projecting the number of high-intensity pixels detected by the first comparator 601 in a one-dimensional manner, 603 is a vertical method for displaying a characteristic image of a baseball broadcast as shown in FIG. Pattern data subjected to projection processing, 604 is a cumulative addition circuit of differences for cumulatively adding a predetermined number of data differences between corresponding values of data from the vertical projection circuit 602 and the pattern data 603, and 605 is a cumulative addition circuit of differences. 60
It is a second comparator that compares the output of No. 4 with a second reference level to determine the recognition result.

First for the video pattern very similar to FIG.
When a high-luminance pixel is detected by the comparator 601 and the projection data is obtained by the vertical projection circuit 602, data similar to that in FIG. 5A is obtained. The difference between the corresponding data of the data of FIG. 5A and the data very similar to FIG. 5A is small, and the value obtained by cumulatively adding a predetermined number of these difference components is also small. Therefore, when the cumulative addition result of the differences is smaller than the predetermined reference level, the pattern data is considered to match and the characteristic image can be recognized.

Example 5. Next, an example of recognizing a baseball relay from the characteristics of the chromaticity signal will be described. In a video pattern as shown in FIG. 5, when projection processing is performed in the horizontal direction on pixels having a high green component as a chromaticity signal, as shown in FIG. 5 (a), it is high at the position corresponding to the horizontal direction in the area D. It becomes a value. An example of recognition using this feature will be described below.

FIG. 7 is a block diagram of a video content recognition apparatus according to a fifth embodiment of the present invention. In the figure, 701 is a green component extraction circuit that outputs “H” when the green component of the chromaticity signal input by table lookup is high, and outputs “L” when it is low, and 702 is a green component extraction circuit 701. A horizontal projection circuit for projecting the detected high pixel number of the green component in one dimension is shown in FIG.
Pattern data obtained by performing projection processing on a characteristic video of a baseball broadcast as shown in (a) in a horizontal method, 704
Is a cumulative addition circuit of differences for cumulatively adding a predetermined number of data differences between corresponding values of data from the horizontal projection circuit 702 and the pattern data 703, and 705 compares the output of the cumulative addition circuit 704 of differences with a reference level. Then
It is a comparator that determines the recognition result.

When a green component extraction circuit 701 detects a pixel having a high green component in a video pattern very similar to that shown in FIG. 5 and horizontal projection circuit 702 obtains projection data, it is very similar to that shown in FIG. Data is obtained. The difference between the data of FIG. 5A and the corresponding data of the data similar to FIG. 5A is small, and the value obtained by cumulatively adding a predetermined number of these difference components is also small. Therefore, when the cumulative addition result of the differences is smaller than the predetermined reference level, the pattern data is considered to match and the characteristic image can be recognized.

Example 6. Next, the recognition information holding means for holding the recognition information for a predetermined time once the video content is recognized will be described.

FIG. 8 is a diagram for explaining the circuit configuration and operation of the sixth embodiment of the present invention, and shows the case where the recognition information holding means is constituted by a counter. The value that sets the delay amount with the recognition signal is loaded, the inversion of the carry is output as the recognition delay signal that holds the recognition information for the predetermined time, and the count-up is enabled by the logical product of the predetermined timer output pulse and the recognition delay signal. To do.

The operation will be described assuming that the delay amount is set to 3 pulses by the timer output pulse. When the recognition signal becomes "H",
The value corresponding to the delay amount of 3 pulses is loaded into the counter,
Since the carry becomes "L", the recognition delay signal outputs "H". When the timer output pulse is counted by 3 pulses after the recognition signal becomes "L", overflow occurs, the carry becomes "H", and the recognition delay signal becomes "L".

With the above-described structure, a recognition delay signal which is turned on immediately and turned off after a lapse of a predetermined time is obtained as a recognition signal. With this, even if the characteristic images are not continuous, it is sufficient if the characteristic images can be detected periodically.
It can also be used for commercials.

In the above example, the case of the baseball relay as the video content is explained, but other sports programs such as sumo and soccer may be used, and various categories such as drama and news are conceivable. Further, the knowledge base means may or may not be able to learn, and the low level knowledge base means can be realized by a decoding circuit of a specific pattern.

[0054]

As described above, according to the first aspect of the present invention, even if a program is extended, a desired program can be reserved and recorded by detecting the extension from the video information and delaying the reserved recording time. .

According to the second aspect of the invention, the contents of the video can be recognized by utilizing the characteristics of the luminance pattern information of the video signal.

In the third invention, the contents of the video can be recognized by utilizing the characteristics of the chromaticity pattern information of the video signal.

In the fourth aspect of the invention, the contents of the image can be recognized by utilizing the high frequency component of the image signal or the characteristics of the edge pattern information.

According to the fifth aspect of the invention, the contents of the video can be recognized with a small circuit scale and high accuracy by utilizing the feature of the edge pattern information of the predetermined telop display inserted in the video signal.

According to the sixth aspect of the invention, by utilizing the characteristics of the edge pattern information in the vertical direction of the predetermined telop display inserted in the video signal, the contents of the video can be recognized accurately even with respect to ghost and jitter. You can

In the seventh invention, by utilizing the characteristic of the relative positional relationship of the edge pattern information of the predetermined telop display inserted in the video signal, the difference in the display position of the telop display due to the difference of the broadcasting station can be realized. On the other hand, the contents of the image can be recognized.

In the eighth aspect of the invention, the "S" particularly for baseball broadcasting is used.
By detecting the telop display of "B" [O], it is possible to recognize the baseball broadcast to various broadcasting stations with a small circuit scale and high detection accuracy.

In the ninth aspect of the invention, even if the characteristic image is not continuous, the content of the image can be recognized if it can be detected periodically, and it becomes strong against the disturbance against image recognition such as commercial.

[Brief description of drawings]

FIG. 1 is a block diagram of a recording reservation device according to a first embodiment of the present invention.

FIG. 2 is a diagram for explaining the operation of the recording reservation device according to the first embodiment of the present invention.

FIG. 3 is a block diagram of a video content recognition apparatus according to a second exemplary embodiment of the present invention.

FIG. 4 is a block diagram of a video content recognition device according to a third exemplary embodiment of the present invention.

FIG. 5 is a diagram illustrating a video content recognition device according to embodiments 4 and 5 of the present invention.

FIG. 6 is a block diagram of a video content recognition device according to a fourth exemplary embodiment of the present invention.

FIG. 7 is a block diagram of a video content recognition device according to a fifth exemplary embodiment of the present invention.

FIG. 8 is a diagram for explaining a circuit configuration and operation of a sixth embodiment of the present invention.

FIG. 9 is a block diagram of a conventional recording reservation device.

FIG. 10 is a diagram for explaining the operation of a conventional recording reservation device.

[Explanation of symbols]

 11 Feature Detecting Means 12 Knowledge Base Means 13 Feature Matching Means 14 Video Content Recognition Means 15 Program Information Setting Means 16 Recording Time Control Means

─────────────────────────────────────────────────── ───

[Procedure amendment]

[Submission date] February 2, 1994

[Procedure Amendment 1]

[Document name to be amended] Statement

[Name of item to be corrected] Claim 5

[Correction method] Change

[Correction content]

[Procedure Amendment 2]

[Document name to be amended] Statement

[Name of item to be corrected] Claim 6

[Correction method] Change

[Correction content]

[Procedure 3]

[Document name to be amended] Statement

[Name of item to be corrected] Claim 7

[Correction method] Change

[Correction content]

[Procedure amendment 4]

[Document name to be amended] Statement

[Name of item to be corrected] 0015

[Correction method] Change

[Correction content]

The video content recognizing apparatus according to a fifth invention comprises a feature detection means for detecting a feature of the pattern information of the telop display inserted into the video signal, knowledge of the telop display of pattern information inserted And a feature matching unit for matching the detected feature with the knowledge base.

[Procedure Amendment 5]

[Document name to be amended] Statement

[Correction target item name] 0016

[Correction method] Change

[Correction content]

The sixth video content recognizing apparatus according to the present invention includes: a feature detection means for detecting a characteristic of the vertical pattern information of the telop display inserted into the video signal, the telop display is inserted in the vertical direction and knowledge base means having a knowledge of the pattern information is obtained by a feature matching means for matching the detected feature and the knowledge base.

[Procedure correction 6]

[Document name to be amended] Statement

[Correction target item name] 0017

[Correction method] Change

[Correction content]

The video content recognizing device according to a seventh aspect of the present invention, the feature detection means for detecting a characteristic of the relative positional relationship between the pattern information of the telop display inserted into the video signal, the telop display of paths that are inserted The knowledge base means has knowledge about the relative positional relationship of the turn information, and the feature matching means for matching the detected feature with the knowledge base.

[Procedure Amendment 7]

[Document name to be amended] Statement

[Name of item to be corrected] 0024

[Correction method] Change

[Correction content]

[0024] In the fifth invention, it collates the features of predetermined telop display of pattern information inserted in the video signal, a feature that has a knowledge of a telop display of pattern information in advance certain program.

[Procedure Amendment 8]

[Document name to be amended] Statement

[Name of item to be corrected] 0025

[Correction method] Change

[Correction content]

[0025] In the sixth invention, the features of the vertical pattern information of a predetermined telop display that is inserted into the video signal,
Matching feature to have a knowledge in advance of the predetermined program telop display of the vertical path coater <br/> down information.

[Procedure Amendment 9]

[Document name to be amended] Statement

[Correction target item name] 0026

[Correction method] Change

[Correction content]

[0026] In the seventh aspect of the invention, the characteristic of the relative positional relationship between the predetermined telop display of pattern information inserted in the video signal, as knowledge of the relative positional relationship between the telop display of pattern information in advance certain program Match the features you have.

[Procedure Amendment 10]

[Document name to be amended] Statement

[Name of item to be corrected] 0058

[Correction method] Change

[Correction content]

[0058] In the fifth aspect of the present invention, by utilizing the characteristics of the pattern information of a predetermined telop display that is inserted into the video signal, can recognize the contents of the video with a small circuit scale, high precision.

[Procedure Amendment 11]

[Document name to be amended] Statement

[Correction target item name] 0059

[Correction method] Change

[Correction content]

[0059] In the sixth invention, by utilizing the characteristics of the vertical pattern information of a predetermined telop display that is inserted into the video signal, to recognize the contents of accurately image against ghosts and jitter You can

[Procedure Amendment 12]

[Document name to be amended] Statement

[Correction target item name] 0060

[Correction method] Change

[Correction content]

[0060] In the seventh invention, by utilizing the characteristic of the relative positional relationship between the pattern information of a predetermined telop display that is inserted into the video signal, the difference of the display position of the display a telop by the difference or the like of the broadcasting station On the other hand, the contents of the image can be recognized.

Claims (9)

[Claims] 1. A video content recognizing means comprising a feature detecting means for detecting a feature of a video signal, a knowledge base means having knowledge about a feature of a video content, and a feature collating means for collating the detected feature with a knowledge base. And a program information setting means for setting information on a program to be recorded, the recognition result of the video content recognition means and the information set by the program information setting means are collated to control the start and end times of the recording operation. A recording reservation device comprising a recording time control means. 2. A feature detecting means for detecting a feature of luminance pattern information of a video signal, a knowledge base means having knowledge about the luminance pattern information, and a feature collating means for collating the detected feature with the knowledge base. A video content recognition device characterized by: 3. A feature detecting means for detecting a characteristic of chromaticity pattern information of a video signal, a knowledge base means having knowledge about the chromaticity pattern information, and a feature collating means for collating the detected feature with a knowledge base. An image content recognition device comprising: 4. A feature detecting means for detecting a feature of a high frequency component of a video signal or edge pattern information, a knowledge base means having knowledge of a high frequency component or edge pattern information, and a collation of the detected feature with a knowledge base. A video content recognition device comprising: a feature matching means for performing. 5. A feature detecting means for detecting a feature of the edge pattern information of the telop display inserted in the video signal, a knowledge base means having knowledge about the inserted edge pattern information of the telop display, and the detected feature. And a feature matching means for matching the knowledge base with the knowledge base. 6. A feature detecting means for detecting a feature of vertical edge pattern information of a telop display inserted in a video signal, and a knowledge base having knowledge about vertical edge pattern information of the inserted telop display. An image content recognition apparatus comprising: a means and a feature matching means for matching the detected feature with a knowledge base. 7. A feature detecting means for detecting a feature of a relative positional relationship of edge pattern information of a telop display inserted in a video signal, and knowledge of a relative positional relationship of edge pattern information of the inserted telop display. A video content recognition device, comprising: a knowledge base unit having the above; and a feature matching unit for matching the detected feature with the knowledge base. 8. The video content recognition device according to claim 5, wherein the video content to be recognized is a baseball relay, and the telop display for detecting a feature is "S", "B" or "O". . 9. A video content recognizing device comprising: recognition information holding means for holding the recognition information for a predetermined time even when the video content is changed by the feature matching means.