JPH04306093A - Method for detecting movie portion of television broadcast program during television broadcast in partial selection video recording device - Google Patents

Abstract

PURPOSE:To discriminate a genuine video image portion accurately and easily in a TV program when the sum of correlation values obtained through the comparison of a data of one frame and a data of a succeeding frame for each picture element of the same position indicates a movie film. CONSTITUTION:A horizontal vertical synchronizing signal detection means 2 detects a horizontal and a vertical synchronizing signal from a video signal, gives a timing signal via a timing detection means 3 to a sampling means 1 to sample and digitize a video signal and stores a picture element data for each frame to a storage means 3. A picture element unit correlation detection means 4 compares a data by one frame from the means 3 with a data by a succeeding frame for each picture element of the same position to obtain a correlation value, an adder means 6 adds correlation values by one frame, a correlation change pattern discrimination means 8 compares the sum of correlation values with a correlation value change pattern of a movie film and outputs a section to a VTR when it is discriminated that the both are coincident.

Description

[Detailed description of the invention]

0001

[Industrial Field of Application] The present invention is directed to a television (hereinafter referred to as
The present invention relates to a method for detecting genuine movie parts in a broadcast program (abbreviated as television) when recording the program.

0002

[Prior Art and Problems to be Solved by the Invention] When recording a movie program to be broadcast on television, unnecessary parts such as movie commentary and next previews are removed and only the necessary parts, that is, only genuine movie parts are recorded. You may want to extract and record it. In particular, when the broadcast is a commercial broadcast, commercials are often inserted in the middle of the broadcast, so it would be ideal if these commercials could also be removed and recorded.

[0003] In order to meet this demand, a conventional method uses audio identification signals (two languages, stereo) included in television broadcast waves to detect when a video tape recorder is recording when a stereo broadcast signal is received. (VTR) and removes commercials. This removal method was devised based on the fact that most commercial broadcasts are stereo broadcasts, but it is still incomplete in extracting and recording only genuine movie parts.

[0004] This is because everything other than commercials, such as commentary before the movie starts, announcements such as ``The broadcast up to this point has been provided by the sponsor you are watching'' during the program, and previews for the next movie after the movie ends, etc. This is because it will be recorded. For example, when recording is started at the time set by a timer device, it is rare for the target movie to start suddenly, and usually starts with a commercial or movie commentary. Since this commentary will be broadcast in bilingual mode, it will be recorded if the VTR that is recording is paused when a stereo broadcast signal is received. In addition, announcements during the program and previews for the next program as described above are also broadcast in bilingual mode, so in the conventional system, these would also be recorded.

[0005] As described above, in the conventional method, a VTR that is recording is temporarily stopped when a stereo broadcast signal is received by using an audio identification signal included in the television broadcast radio waves.
The drawback is that it is not possible to extract and record only the authentic movie parts while being compatible with various programs. The present invention aims to eliminate this drawback and makes it possible to accurately and easily identify genuine movie portions from programs being broadcast on television.

[0006]

[Means for Achieving the Object] In order to achieve this object, the present invention samples the video signal of the television broadcast radio wave and converts it into a digital quantity (step 2 in Fig. 1), which corresponds to one frame of television. The data from the digital signal received by the user is compared with the data from the digital signal for one frame that comes next for each pixel at the same position, and the correlation value is calculated (step 3 in Figure 1). The sum of the correlation values for each frame for one frame is sequentially calculated for each frame (steps 4 and 5 in FIG. 1). Then, it is determined whether or not these sequentially calculated total correlation values match the correlation value change pattern for a movie film (steps 7 and 8 in FIG. 1), and it is determined that they match. At the time (step 11 in FIG. 1), the video signal is output to detect a genuine movie part.

[0007]

[Operation] The operation of the present invention will be explained based on FIG. 1, which is a diagram corresponding to claims. The video signal of the television radio wave is sampled and converted into a digital quantity as shown in step 2 in FIG. This data in the form of a digital signal corresponding to one television frame is sequentially stored in each square, that is, each pixel, of the sampling data storage means shown at 3 in FIG.

The next data corresponding to one frame of television is also sequentially stored in each pixel of the sampling data storage means 3 in the same way as the previously sampled data, but as shown in step 3 in FIG. By comparing one frame's worth of data sampled last time with the next frame's worth of data sampled for each pixel at the same position, a correlation value for each pixel can be obtained. Incidentally, when sampling the video signal, it is preferable to read out one frame's worth of data sampled last time from the sampling data storage means 3 in advance, as shown in step 1 in FIG.

Then, as shown in steps 4 and 5 in FIG. 1, when the sum of the correlation values for each pixel for one frame is calculated in sequence for each frame, the total correlation value between two frames is calculated. You can get it. Once the total correlation values between two frames have been calculated in sequence, it is determined whether these values match the correlation value change pattern for a movie film, as shown in steps 7 and 8 in FIG. Then, it can be determined whether the television broadcast screen is a film or not. When the determination result is that they match, it is determined that the television broadcast screen is a film, as shown in step 11 in FIG.

[0010] If a video signal is output at this time,
It becomes possible to extract authentic movie parts from programs being broadcast on television and record them on a video tape recorder (VTR).

If it is determined in step 5 of FIG. 1 that the total correlation value for one frame has not been calculated up to a predetermined number, for example, N frames, the process proceeds to step 6.
Steps are taken to calculate the total correlation value for up to N frames. If it is determined in step 8 of FIG. 1 that the total correlation value does not match the correlation value change pattern for a movie film, the process proceeds to step 9 and sequentially examines up to a predetermined number of data, ie, N. Nevertheless, if it does not match the correlation value change pattern for movie film, it is determined that the screen is not film (
Step 12 in Figure 1). Furthermore, if it is determined in step 9 of FIG. 1 that up to N has not been checked, the register is advanced by one as shown in step 10 of FIG. 1, and the process returns to step 7 of FIG. Read it out and follow the same steps as before.

0012

[Example] The present invention will be described below with reference to Figs. 2 to 9 showing examples thereof.
This will be explained in detail based on the following.

FIGS. 2 and 3 are flowcharts showing a procedure for detecting a movie part during television broadcasting in more detail than in FIG. 1, which is an embodiment of the present invention, and FIG. 3 shows a series of steps continued from FIG. shows. FIG. 4 is a block diagram of a device for performing this series of operations. Next, embodiments of the present invention will be described in detail based on this flowchart and block diagram.

[0014] The video signals of television broadcast waves currently used in Japan are based on the NTSC system (National Telecommunication System).
levision system committee
This signal is converted into a digital quantity and sampled by a video signal sampling means shown at 1 in FIG. 4. As the video signal sampling means 1, for example, an analog/
One example is a digital converter.

The above-mentioned NTSC system is a system in which an image is converted into an optical signal on the transmitting side and transmitted by radio waves, and one TV frame (one screen) consists of a total of five signals as shown in FIG.
Although it is composed of an array of 25 scanning lines, horizontal and vertical synchronization signals are transmitted along with the video signal so that exactly the same image transmission and reception as on the transmitting side can be obtained.

Among these horizontal and vertical synchronization signals, the horizontal synchronization signal is a signal that indicates the start position of each of the 525 scanning lines, and the vertical synchronization signal is a signal that indicates the start position of one frame (one screen) of the television. It is. Therefore, for example, when a vertical synchronization signal is detected by the horizontal and vertical synchronization signal detection means shown as 2 in FIG. Move horizontally. When the horizontal synchronization signal is detected by the horizontal and vertical synchronization signal detecting means 2, the light spot is located at the top (left end) of the screen, as shown by the scanning line 1 in FIG.
, and then move sequentially to the right as in the previous case. In the NTSC system, this is repeated 525 times, and when 525 scanning lines are drawn, one frame (one screen) is constructed. Then, when the next vertical synchronization signal is detected, the next new frame (screen) starts.
The next frame (screen) will be constructed in the same way as in the previous case. In the NTSC system, this screen configuration is repeated 30 times per second.

Video signals are transmitted one after another in this manner, but when a vertical synchronization signal is detected by the horizontal and vertical synchronization signal detection means 2 (step 101 in FIG. 2), the video signal sampling means 1 The signal is converted into a digital quantity and sampled (step 103), and stored for each pixel in a frame memory, such as the one shown in FIG. 6, by the sampling data storage means 3. The pixel referred to here refers to each square of the frame memory in FIG. 6, and the data is stored one after another in the pixel corresponding to each light spot on the scanning line of the television frame (screen). For example, data at point 0 in the X direction on scanning line 0 is stored in the frame memory at m=n=0, that is, at the pixel at coordinates (0,0) in FIG. The subsequent data on scanning line 0 is m = 1, 2, that is, one for each pixel located on coordinate 0 in the Y direction, such as coordinates (1, 0) and (2, 0) in Fig. 6. Stored sequentially. After sampling m of the above signals, when a horizontal synchronization signal is detected by the horizontal and vertical synchronization signal detection means,
The next data is stored in the frame memory at m=0, n=1, that is, at the pixel at the coordinates (0, 1) in FIG. 6, and thereafter m pieces of data are sampled in the X direction. Then, each time a horizontal synchronization signal is detected, data is stored in all pixels of one frame memory by advancing n in the Y direction one by one.

In this way, data for one frame of the television is sampled and stored in the frame memory, but when the next vertical synchronization signal is detected by the horizontal and vertical synchronization signal detection means 2, the next television frame is Start sampling frame data. The data in this case is also stored sequentially in each pixel of the frame memory as in the previous case, but since the frame memory contains the previously sampled data, when storing new data in this frame memory, compares the previously sampled data for one frame with the next frame's worth of data for each pixel at the same position, calculates the correlation value, and then updates the frame memory with the new data of the next frame. (step 104). The correlation value referred to here is the absolute value of the difference between stored values at pixels (X, Y) at the same position in the frame memory, and can be expressed by the following equation. 1 pixel correlation value=|sampling value−frame memory value|This correlation value can be detected by a pixel-by-pixel correlation value detection means shown at 4 in FIG.

Note that when sampling data, storing it in the frame memory, and detecting the pixel-by-pixel correlation value, the data at addresses m and n of the frame memory are stored in advance as shown in step 102 in FIG. Follow the steps to read. The timing is controlled by timing control means shown at 5 in FIG.

Next, the obtained correlation value for each pixel (pixel unit correlation value) is added for one frame by pixel unit correlation value addition means shown at 6 in FIG. The correlation value is obtained for each interval (steps 105, 106). The sum of correlation values between frames (total correlation value S) can be expressed by the following equation 1.

[0021]

[Math 1]

If the total correlation value S is small, it means that the correlation between the two frames is high (there is no difference in the screen); conversely, if the total correlation value S is large, it means that the correlation between the two frames is high (there is no difference in the screen). , the correlation between the two frames is low (there is a difference in the screen)
It means that. The smallness and largeness of this total correlation value S are checked over multiple frames, for example from frame 0 to frame N as shown in FIG. 7, and the respective values (SO to SN) compared between each frame are shown as 7 in FIG. The total correlation value is stored in a storage means, that is, a register.

Note that if the pixel-by-pixel correlation values have not been added for all of one frame in step 106, the process proceeds to step 107, where the sampling data is stored and the read position of the frame memory is incremented one by one. Returning to step 102, correlation values in the remaining pixel portions that have not been added are added.

When comparing whether the total correlation value S is small or large from frame 0 to frame N, N pieces of data will be written in the correlation value total storage means 7. The correlation value change pattern determining means shown at 8 in FIG. 4 determines whether the total value matches the correlation value change pattern for a movie film.

The correlation value change pattern for a movie film is stored in a correlation value change pattern storage means for a film, ie, a program memory of a microcomputer, as indicated by 9 in FIG. The comparison between the total correlation value and the correlation value change pattern for a movie film is performed by the control unit B of the microcomputer according to the control program written in the program memory.

Note that if the total correlation value has not been checked up to N frames in step 108, the process advances to step 109, and the frame memory read position is set to m=n.
=0, and correlation value total storage means 7 (register)
is advanced by one, and the process returns to step 101 to calculate the total correlation value for the remaining frames, and store the calculated value in the correlation value total storage means 7 (register).

By the way, before explaining the procedure for comparing the total correlation value of a predetermined number with the correlation value change pattern in the case of a movie film, let us explain the correlation value change in the case of a movie being broadcast on television according to the NTSC system. Explain what the pattern is. This correlation value change pattern can be easily determined from a method for resolving the lack of frames when a movie is broadcast on television using the NTSC method.

[0028] When broadcasting a movie by the television method,
The original movie is made on film, and the frame advance rate is 24 frames per second. On the other hand, NTSC
Television broadcasts based on this system are broadcast at a rate of 30 frames per second. That is, the frame advance rate in television broadcasting based on the NTSC system is 30 frames per second. Therefore, when a movie is shown on TV, 6
There will be a shortage of frames, but in order to make up for this shortage, the program actually broadcasts 4 frames and then re-transmits the 4th frame for the 5th frame, 6 times per second. To explain this based on FIG. 8, the frame advance speed of movie film is 24 frames (0 to 23) per second;
Television broadcasting using the NTSC system has 30 frames (0 to 29) per second, and the 4th frame (frame number 3) of a movie film is the 4th and 5th frame (frame numbers 3 and 4) of a TV frame. The 8th frame (frame number 7) of a movie film is the 9th and 10th frame (frame numbers 8 and 9) of a TV frame, and the 12th frame (frame number 11) of a movie film is a TV frame. The 14th and 15th frames (frame number 1)
3 and 14), the 16th frame (frame number 15) of the movie film is the 19th and 20th frame of the TV frame.
The 20th frame (frame number 19) of a movie film is the 24th and 25th frame (frame numbers 23 and 2) of a TV frame.
4), the 24th frame of the movie film (frame number 23)
) are sent to the 29th frame and 30th frame (frame numbers 28 and 29) in the TV frame, and so on, and so on, and so on, and so on, and so on, and so on. In the TV frame, the same screen is displayed every fourth frame. is coming. That is, one TV frame is compensated for every four film frames, and by repeating this six times per second and broadcasting, the shortage of frames is eliminated.

As is clear from this comparison between a movie film and a television frame, when a movie film is converted to a television screen in the NTSC system, the same screen appears every fourth frame. In other words, in terms of TV frame numbers, 3, 4, 8, 9, 13, 14, 18, 19, 23
, 24 and 28, 29 are the same screen (in other words,
It can be said that the correlation between those frames that follow one another is high). When this is converted into a correlation value and shown on a graph, it becomes as shown in FIG. That is, as shown by TV frame numbers 3 and 8 in FIG. 9, the correlation value of those portions becomes small. Although not shown in FIG. 9, TV frame number 1
3, 18, 23, and 28 also have small correlation values as in the case of 3 and 8.

In this way, when a movie is broadcast on television according to the NTSC system, this pattern is constantly repeated, and in the present invention, this correlation value change pattern is stored in advance in the correlation value change pattern storage means 7. The total correlation value (described above) calculated from the video signal of the television broadcast wave and the correlation value change pattern storage means 7
Correlation value change patterns for movie films stored in advance are examined by correlation value change pattern determining means 8 of the microcomputer. If it is determined that two consecutive screens of every fourth frame in a television broadcast are the same, this broadcast can be recognized as a movie.

Next, the correlation value change pattern determination means 8 of the microcomputer compares the correlation value total value between each television frame and the correlation value change pattern stored in the correlation value change pattern storage means 9 for film. The procedure for comparison will be explained.

As shown in FIG. 3, in step 110, the value of the correlation value total storage means 7 is read out, and the value is shown in FIG.
The control unit B determines whether the value is above or below the reference value shown in (Step 1) according to the control program written in the program memory of the microcomputer.
11). If it is determined that the data read from the correlation value total storage means 7 is less than the reference value, step 11
4, the correlation value total storage means 7, ie, the register, is advanced by one and the next value is read out. If it is determined in step 115 that the value is greater than or equal to the reference value, the process proceeds to step 116, where the counter is incremented to count the four data following the data with the small correlation value. In step 117, if it is determined that the counter is 4, that is, if it is determined that data with a small correlation value is followed by four data with a high correlation value, this correlation value change pattern is the same as the change pattern (FIG. 9) stored in the correlation value change pattern storage means 9 for film, so it is determined that the television broadcast screen is a film (step 11).
9).

On the other hand, if it is determined in step 111 that the data read out from the correlation value total storage means 7 is not less than the reference value, the process advances to step 112 and the data stored in the correlation value total storage means 7 is Examine up to N and determine whether there are any places where the correlation value is small,
If it is determined that there is no small correlation value even after checking up to register N, it is determined that the television broadcast screen is not a film (step 120). In addition, if it is determined in step 112 that register N has not been checked, advance the register by one as shown in step 113, return to step 110 again, read the data of this register, and follow the same procedure as before. .

If it is determined in step 115 that the data is not equal to or greater than the reference value, the counter is set to 0 as shown in step 118, and then the procedure returns to step 111 when the determination result is NO, and steps 112 to 12 are performed.
0 or proceed to step 113. If it is determined in step 117 that the counter is not 4, the process returns to step 114 and the steps from step 115 onwards are performed.

In this way, the correlation value change pattern determination means 8 of the microcomputer uses the correlation value change pattern stored in the correlation value change pattern storage means 9 for film and the correlation value change pattern for each television frame. By comparing, it is possible to determine whether the program being broadcast on television is a movie. If a video signal is output when a program being broadcast on television is determined to be a movie, it is possible to record only the genuine movie portion on the videotape of a video tape recorder (VTR). Become.

[0036] This detection method may be carried out alone, or may be carried out in combination with a conventional selective recording method of a portion of a television broadcast program using a sound identification signal.

[0037]

[Effects of the Invention] When the detection method according to the present invention is used, it is possible to accurately and easily detect and selectively record only genuine movie parts from programs being broadcast on television, excluding unnecessary parts. effective.

[Brief explanation of drawings]

FIG. 1 is a claim correspondence diagram showing a procedure for detecting a movie part during television broadcasting according to the present invention;

FIG. 2 is a flowchart showing part of a procedure for detecting a movie part during television broadcasting according to an embodiment of the present invention;

[Figure 3
] A flowchart showing a series of steps continued from FIG.

FIG. 4 is a block diagram of a device for performing this series of operations;

[Fig. 5] A schematic diagram for explaining the screen configuration of a television according to the NTSC system,

FIG. 6 is a schematic diagram showing an example of a frame memory that stores data obtained by sampling video signals of television broadcast waves and converting them into digital quantities;

FIG. 7 is a schematic diagram illustrating a state in which small and large total correlation values of sampling data are compared from frame 0 to frame N;

FIG. 8 is a schematic diagram showing the correspondence between frames of a movie film and television frames when a movie is broadcast on television according to the NTSC system;

FIG. 9 is a schematic diagram showing an example of a correlation value change pattern when a movie is broadcast on television according to the NTSC system.

[Explanation of symbols]

1 Video signal sampling means 3 Sampling data storage means 4 Pixel-by-pixel correlation value detection means 6 Pixel-by-pixel correlation value addition means 7 Total correlation value storage means 8 Correlation value change pattern determination means

Claims (1)

[Claims] Claim 1: A video signal of a television broadcast wave is sampled and converted into a digital quantity, and the digital signal data corresponding to one frame of television and the digital signal of one subsequent frame are generated. The correlation value is calculated by comparing the data for each pixel at the same position, and the sum of the correlation values for each pixel for one frame is calculated sequentially between frames, and that value is the movie film. The television is characterized in that it determines whether or not the pattern matches the correlation value change pattern of the case, and when it is determined that the pattern matches, the video signal is outputted to detect a genuine video portion. A method for detecting a movie part during a television broadcast in a broadcast program part selection recording device.