JPH09182019A - Video signal recording device and reproducing device - Google Patents

Abstract

(57) [Abstract] [Problem] Extracting a representative screen from the received video signal,
Providing a device for recording and / or reproducing a digest. A broadcast signal received by an antenna 1 is received by a video channel tuner 2 and then converted into a digital signal, and a representative screen detection circuit 5 detects, for example, scene change information. Based on the detected scene change information, a video signal is recorded in the storage device 22 or the stored information is read out to perform the digest recording / reproducing. The scene change information for detecting the representative screen is performed by the motion detection of the video frame, but in the screen where the scene changes gently, for example, when the screen is faded in / out, many detection outputs are obtained, which is wasteful. The representative screen is extracted. This recording / reproducing apparatus obtains scene change information from the last detection signal when the detection signal is continuous for a predetermined period,
Prevent unnecessary recording and playback.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a recording apparatus for recording a large number of programs, checking the contents of each program, and efficiently watching the selected program. The present invention relates to a recording device that enables efficient grasping of broadcast information, such as wanting to grasp the contents of news or a specific program in a short time. The present invention also provides
In particular, the present invention relates to a recording / reproducing device for recording and reproducing a large amount of broadcast signals, and using a hard disk / optical disc / semiconductor memory or the like as a recording medium, a device for efficiently recording / reproducing by utilizing accessibility. Is suitable for.

[0002]

2. Description of the Related Art Conventionally, VT has been used as a broadcast signal recording device.
R is known. Further, as a business use, a recording device for a magneto-optical disk / optical disk has been put into practical use. Therefore, a representative screen is extracted from the broadcast signal and, for example, a moving screen or a screen corresponding to an audio signal is recorded to reduce the capacity of the recording medium, and from the broadcast signal of one channel or multiple channels. A recording device capable of grasping the contents of a program to some extent has been considered.

[0003]

That is, a large amount of video signals are recorded while reducing the recording capacity by detecting the change / movement of the screen from the video signal itself and recording the digest. In the case of a recording apparatus of such a system, for example, a motion detection technique or a system of detecting a representative screen with change / motion using a pixel value histogram of the screen has sufficient detection capability, and two scenes are used. When switching in frame (or field) units, the method as it is is sufficiently practical.

However, when the scenes are gradually switched over a period of several frames to several seconds, such as fade-in / fade-out, it is necessary to set the criteria for the presence / absence of change to be sensitive to change / movement. ,Fade-in/
All frames (or fields) that are out are detected as points with large changes / movements, and all the changing screens that do not seem to be valid information are recorded (or reproduced and displayed), which is unnecessary. Results in significant redundancy.

If the criterion for determining the presence or absence of a change is set to detect only a case where a change between adjacent frames is large, it is possible to switch to another screen in a few frames to a few seconds, but it is instantaneous. Compared with the screen switching of
There is a problem that fade-in / out with a small amount of change cannot be detected in the adjacent frame (or field) question.

[0006]

In order to solve such a problem, the present invention detects information relating to a screen change from the video signal itself by using means such as scene change information detection, and based on this information, a representative An image is selected. At the time of this selection, the change detection judgment is performed especially when the fade-in /
If a change criterion that is sensitive enough to detect changes between out-of-frames is adopted, and if multiple changes are detected within a set period of time, the previously detected change point is Invalidate and set the representative screen only for the last detected change point.

[0007]

[Function] In a digest recording device or a reproducing device for recording a representative screen, it is possible to detect a screen change due to fade-in / out without failing, and to eliminate the recording of a less effective image during fade-in / out. By doing so, it is possible to avoid recording unnecessary information. Therefore, the capacity of the recording medium can be used maximally and effectively.

Further, when performing multi-channel recording, it is possible to avoid fading in / out of one channel from interfering with recording of another channel when configuring a mechanical device for performing digest recording. Simultaneous recording of channels can be realized with higher quality and easily.

On the other hand, the information is not deleted during recording,
Even in a device that performs digest playback during playback display, there is no possibility of failing to detect a screen change due to fade-in / out, and there is no possibility of displaying an ineffective image during fade-in / out. Digest playback status can be realized. Further, when used as a pre-process in the editing apparatus, the same effect can be provided in automatic scene / cut extraction.

[0010]

FIG. 1 shows a typical block diagram of a recording apparatus as an embodiment of the present invention. In the following, from the recording system, the broadcast signal that has entered the receiving antenna shown in (1) is divided into the video signal tuner shown in (2) and the audio signal tuner shown in (13). Is entered.

The video signal is the LP shown in (3).
After passing through the F (low pass filter), it is digitized by the AD converter shown in (4). Further, although the sync separation circuit (10) performs the sync separation, the position of the sync separation shown in this figure is an example, and the signal after digitizing may be sync separated. The sync signal shown by (11) obtained from the sync separation circuit (10) is sent to the first microcomputer 1 shown by (20) to produce a sw control signal (12) described later. The signal digitized by the AD converter (4) is input to the representative screen detection circuit (5) to detect information for the representative screen detection.

Although the above-mentioned video signal receiving means is shown for only one channel, a plurality of tuners and selected signal processing circuits for a plurality of channels are provided side by side to simultaneously generate a video signal for a plurality of channels. On the other hand, it is configured to extract representative screens, and while supplying these representative screens to a storage means described later while selecting them in a specific priority order or in a time-division manner, to a recording medium such as an optical disk or a hard disk drive. It may be recorded.

As the information detecting method, a scene change detecting technique using a motion detecting technique, a scene change detecting technique using a histogram of pixel values on the screen, or the like can be used. (As an example of the information detection method, a case of using a motion detection technique will be described with reference to FIG. 4 and subsequent figures.) The extracted information (6) obtained by the representative screen detection circuit (5) is
Similar to the synchronization signal, it is sent to the microcomputer (20) as information for creating the Sw control signal (12). Microcomputer (2
0), the sync signal (11), the representative screen detection circuit (5)
Based on the extracted information (6) obtained in step 1 and the timer etc. in itself, it is judged whether to record or thin out in units of each field (or frame) of each channel. The output Sw control signal is output to control the switch (7) for turning on and off the input signal of the encoder (8).

When the switch (7) is turned on, the video signal is input to the video encoder (8) and the encoded signal output (9) is supplied to the multiplexer (18). The encoding method is not specified, but in the case where one field or one frame is independently selected and controlled to perform slide show reproduction of a still image, in-plane completion processing is required. . Also,
When recording a short-time moving image, it is required to use a coding method that is completed within the screen only for the first frame (or field), but from the next frame (or field), compression in the time direction is performed. It is also possible to adopt an encoding method using.

When the switch (7) is off, no signal is input to the encoder (8) and therefore the encoded signal output (9) is also not output. In this control, it is possible to select independent 1 field or 1 frame as a representative screen to perform slide show-like reproduction of a still image, but select a signal for several continuous fields to several minutes. It is also possible to realize a playback method in which short videos are switched and displayed.

Although it is possible to record only the representative screen, in order to display a smooth screen at the time of reproduction, a screen other than the representative screen is added and recorded within the range of the recording information amount. It is also possible. When selecting the screen to be added, the method of selecting at every fixed time interval is the simplest, but the representative screen detection circuit (5)
By using the extracted information (6) obtained from (3), if there are a lot of screens with relatively large movements, the number of screens to be recorded is reduced by reducing the number of screens to be thinned out, and the scenes with less motions are closer to still images. On the other hand, it is more desirable to control the number of screens to be thinned and the number of screens to be recorded to be reduced. By this processing, the video signal information, which occupies most of the recorded signal, is considerably reduced while maintaining the characteristic of not compromising the understanding of the video content.

Next, the audio signal processing of the recording system will be described. The audio signal detected by the audio tuner (13) passes through the LPF (14), is digitized by the AD converter (15), is then encoded by the audio signal encoder (16), and is the encoded audio signal (17). Becomes The audio signal may be separated from the channel selected via the video tuner (2).

The encoded video signal (9) and the audio signal (17) are input to a multiplexer (18) and multiplexed. At this time, the selection information (time / Sw ON / OFF information: 21a) regarding the selection of the representative screen of the video signal is also sent from the first microcomputer (20) to the multiplexer (18) and recorded as additional information regarding recording / reproduction. . At the same time, information regarding the selection of the representative screen of this video channel (time / Sw on / off information:
21b) is sent to the memory (38) and stored in a form corresponding to the address on the storage medium. This enables program selection at the time of reproduction and synchronous reproduction of video and audio signals.

The multiplexed signal is subjected to signal processing such as ECC (error collector code) addition and channel coding in a recording circuit (19) for recording in a storage device (storage). , Are recorded in the storage device (22).

Next, the reproducing system will be described. The user operation input circuit (35) receives the operation desired by the user and inputs the information (36) to the second microcomputer (37). The second microcomputer (37) receives from the memory (38) the recording medium address and the information (21c) relating to the recording of the audio and video, and sends it to the control / drive circuit (40) of the storage so that the information can be obtained. The access control signal (39) is output. Access control signal (39)
Received storage control / drive circuit (40)
Reads out the information requested by the user from the storage medium by the signal (41) for controlling the storage, the read signal is subjected to channel coding and ECC by the storage reproducing circuit (23), and the demultiplexer (24) )
Is input to

The demultiplexer control signal (42) is also sent from the microcomputer (37) to the demultiplexer to synchronize the audio / video signal at the time of reproduction and to obtain the correct time relationship of the thinned video signal. Enables realization of reproduction and so on.
The video signal output from the demultiplexer (24) is
The video signal is decoded by the video signal decoder 1 (25), passes through the video signal processing circuit (26), passes through the DA converter (27), and is output as an analog signal (28). Of course, depending on the specifications of the receiver that receives the signal, a DA converter (2
It is also possible to output a digital signal (29) that does not pass 7). The video signal processing circuit (26) is a microcomputer 2
In accordance with the control signal (43) from (37), graphic processing such as on-screen display function is performed.

On the other hand, the audio signal output from the demultiplexer (24) passes through the audio signal decoder (30), the audio signal processing circuit (31), and the DA converter (32) to obtain the analog audio signal ( 33) is output. Digital output (3
4) can be provided in the same manner as the video signal.
In this example, the first and second microcomputers are described as being provided, but this is for convenience of description, and the microcomputer (20) and the microcomputer (37) are processed by one microcomputer. It is also possible to divide the microcomputer into several pieces and use two or more chips.

FIG. 2 shows a multi-channel (n-channel) video tuner (2a to 2n) and an audio tuner (13a to 1).
3n) is provided to show a multi-channel digest recording device, and the numbers excluding the subscripts a to n indicate the same functional circuits as in the case of FIG. In the case of this embodiment, the number of encoders (8a to 8b) is m (n> m), and the image of the channel from which the representative screen is extracted from the image signal of the n channel is switched (7a to 7m). It is selected so that it can be recorded. When performing the digest recording, the channel selection information and the extraction information of the representative screen extracted from the selected channel are supplied to the recording means via the microcomputer (20) and linked to the video signal and the audio signal. It is supplied to the information recording means.

FIG. 3 shows an example of a video signal recording / reproducing apparatus capable of reading out a digest of a 1-channel broadcast program as in FIG. The basic horizontal composition is shown in Figure 1.
Although it is not so different from the digest recording device of (1), it is different in that all the video / audio information is recorded together with the time code at the time of recording (not thinned out at the time of recording). That is, the switch (7) for selecting a recording screen is omitted, and a large amount of recording information is recorded while reducing the information amount by the MPEG system or other information compression decoder, and digest reproduction can be performed during reproduction to shorten the time. There is one that makes it possible to grasp the recorded contents. Therefore, at the time of recording, only the extraction information obtained from the representative screen detecting means (5) for detecting the change point of the screen is set, and this extraction information can be used as a read control signal at the time of reproduction for generating the representative screen. ing.

In a system using a removable medium, the read time information of the representative screen is the TOC (Table Of Coten) of the recording medium at the end of a series of recording.
ts) area is written. Further, in a system that uses a storage medium (22) such as an HDD that is a non-removable recording medium as a recording medium, a TOC area can be provided in the medium for writing, but a memory outside the medium (for example, 38 in FIG. 2). It is also possible to write it in) and use it at the time of reproduction.

At the time of reproduction, in a system in which the selection information of this representative screen is recorded in the TOC area of the medium, this TOC information is read prior to the video signal / audio signal,
It is stored in the memory (38) connected to the microcomputer (37). In the system in which the representative screen information is written and stored in the memory outside the medium like the example of the HDD described above, the reading process from the medium can be omitted.

When the user operation input circuit (35) receives a digest reproduction request from the user, the operation signal (36) is sent to the microcomputer (37), and the microcomputer (37) outputs from the memory (38) Information on the start point / end point of the representative screen is read, and based on this information, a control signal (39) for instructing medium access is sent to the medium control /
Input to the drive circuit (40). The medium control / drive circuit (40) controls the storage device (22) with a control signal (41) in accordance with an instruction from the microcomputer (37),
Necessary information is sequentially read from the storage device (22).

Next, the storage reproducing circuit (23) solves the channel coding and ECC, and the demultiplexer (24) synchronizes the video signal and the audio signal, and continuously displays two representative screen questions. Synchronize and separate video signal and audio signal.
5), send to the audio decoder (30). There are many variations in the method of presenting the video / audio of the digest reproduction.

As a display method with the least display information, there is a method of presenting the representative screen at a high speed with the representative screen = still image (without audio), and the method with the most display information is the representative screen = ( (Video for a certain period + voice of the question)
As a method, a method of sequentially presenting representative screens that are distant in time can be considered. In fact, besides this,
Many methods are conceivable, such as a presentation method using a combination of a still image and audio for a certain period of time, a method in which a moving image is used as a representative screen, and the representative screen including audio is presented at n-times speed.

The control signals (42, 43, 44) are sent from the microcomputer (37) to the demultiplexer (24) in accordance with the representative screen presentation method actually set as the design specification.
Also, it is sent to the video signal processing circuit (26) and the audio signal processing circuit (31), and the digest reproduction presentation system of various specifications is finally realized by the processing in these three blocks.

Note that the digest video signal recording /
Even in the case of the reproducing apparatus, it is possible to provide a receiving means capable of coping with multiple channels. Also, in FIG.
Although the D converter (4) and encoder (8) were directly connected,
It is also possible to adopt a configuration in which the representative screen detection circuit (5) is placed on the video signal line and its output is input to the microcomputer (20).

FIG. 4 shows a basic configuration of a motion detection circuit which can be used as a representative screen detection circuit. The input signal is sent to the frame memory (45) and the motion detection circuit (46). The current input signal and the signal one frame before stored in the frame memory are input to the motion detection circuit (46) for comparison. And
The compared result is sent to the microcomputer (20) as representative screen detection information (6). Further, in this example, a configuration example in which a signal delayed by one frame from the input signal output from the frame memory (45) is output to the encoder, but the encoder system output should output the input signal as it is. You can also However, the presence or absence of motion can be judged only when a certain 1-frame signal is input. Therefore, if a signal delayed by 1 frame is sent to the encoder, only necessary frames are encoded and thinned (not recorded) frames. Can be set so that it is not encoded, which is desirable in terms of power consumption reduction.

Next, a motion detection technique will be described as an example of a technique for detecting a screen change and obtaining information for selecting a representative screen. First, the principle of the motion detection technique will be described with reference to FIG. 5, but the motion detection is essentially a technique already put to practical use in MPEG or the like. As shown in FIG. 5, one screen (50) is divided into blocks of s × s or 2s × s pixels. X in the block in the screen
Bpq (51) for the p-th direction and the q-th direction in the y direction
Will be expressed as Also, within the block Bpq, i in the x direction
The value of the j-th pixel (52) in the y-th and y-direction is A (i,
j).

Usually, as the value of S for one screen, a power of 2 such as 4, 8 or the like is selected. Although the block size of S × S is illustrated in the description of FIG. 5, the same algorithm can be applied even when blocks having different numbers of pixels in the vertical and horizontal directions are used, and generality is not lost. It is just an example. Further, in the following description, in order to simplify the description, the description will be given as (screen = frame).
Even when (screen = field), the same algorithm can be applied, and generality is not lost.

Next, the motion detection method will be described with reference to FIG. As shown in FIG. 6, a block Bo: pq (56) in the current frame (55) for which motion detection is desired,
The difference from the inspection block (58) including the number of pixels of s × s in the reference frame (57) is calculated. This inspection block (58) is normally found in the search range (5
As shown as 9), a certain range including the position corresponding to the position of the motion detection block position of the current frame on the reference frame is set.

Within this search range (59), the inspection block (58) is shifted in the x and y directions pixel by pixel, and the block (56) in which the motion detection is to be performed in the current frame and the reference frame are performed. The difference from the inspection block (58) composed of s × s pixels is calculated. In the inspection block within the search range (59), the position in the screen corresponding to the position of the motion detection block (56) of the current frame is the origin, and the shift in the X direction and the y direction. Are represented by k and 1, respectively, and the value of the i-th pixel (53) in the x direction and the j-th pixel (53) in the y direction in the motion detection block is at the position of Ao (i, j), (k, L). When the value of the i-th pixel (54) in the x direction and the j-th pixel (54) in the y direction in the inspection block is expressed as Ar: kl (i, j),
The difference Spq (k, l) between the motion detection block and the inspection block at the position (k, l) is

[Equation 1] Becomes

This calculation is performed for all (k, l) within the initially set search range (59), and the smallest value among the calculated Spq (k, l) is shown (k,
The value (K, L) of 1) is the value of the "motion vector" for this motion detection block Bpq between the current frame and the reference frame. The values of these motion vectors K and L, and the value of Spq (k, l), Spq within one screen
It is possible to judge the change / movement of the screen from the number of (k, l) values below (or above) a certain value.

For example, when a still image continues, for all motion detection blocks of the current frame,
K = 0 and L = 0, and the value of Spq (k, l) at this time becomes zero. Also, if you are panning horizontally,
Similarly, K = C (const) for all blocks,
L = 0, and the value of Spq (k, l) at this time becomes zero if the object to be photographed is stationary, and the time between the reference frame and the current frame is short even if the object to be photographed is slowly moving. If so, the amount of deformation is small, so the price is small. Spq of the part with large deformation in a short time
Only (k, l) has a large value, but the number is small. On the other hand, when the screens are switched, there is no matching screen in the two frame questions, so Spq (k, l) takes a large value for all the values of (i, j).

This motion detection may be performed only for the luminance signal, but it may be performed only for the color difference signal,
There is also a method of performing both for the luminance signal and the color difference signal.
In addition to this, as a method of detecting a change in the screen of a video signal, a method of using a histogram of pixel values of the screen is known, and of course, this method can also be used.

In the case of the present invention, unnecessary motion detection can be invalidated in addition to such motion detection. Hereinafter, the usefulness of invalidating unnecessary motion detection output will be described with reference to FIG. 7. In FIG. 7, the scene (a) of the program in the top row is an example of program switching, and can be easily understood by comparing with the interval of the frame (b) of the video signal in the second row. It is shown that the scene A is switched to the scene B, or the scene E is switched to the scene F instantaneously, while the scene C is switched to the scene D gradually about 10 frames. There is. This switching method can be seen in scene switching using fade-in / out, wipe, and the like.

Usually, when the motion is detected as described above by taking the difference from the previous frame (or several frames) or the change is detected by the histogram, the motion detection output shown in FIG. As described above, when the scene A is switched to the scene B or the scene E is switched to the scene F, one change detection output (bl, f1) is obtained for one scene change, but the scene C to the scene D is obtained.
In the case of gradually switching over a plurality of frames, such as when switching to, a large number of change detection outputs (dl to dlO) are obtained for one scene switching.

For example, considering the case of recording (or reproducing and displaying) a still image as the representative screen, the representative screen of scene B is recorded (or reproduced and displayed) based on the detection signal bl, and the detection signal f1 is displayed. Based on scene F
The recording (or reproduction display) of the representative screen of is an operation as originally intended. However, it is not desirable that 10 representative frames of the scene D that are close to each other are recorded (or reproduced and displayed) based on the detection signals of dl to dlO, which is merely redundant. Therefore, as shown in (d) of FIG. 7, even when the scene C is switched to the scene D, only the first frame (d10) of the scene is the same as b1 for the scene B and f1 for the scene F. It needs to be detected.

That is, in the case of fade-in / out in which the screen gradually changes, when a plurality of changes / motions are detected within a fixed short time, the last of the series of detection outputs is detected. It is necessary to process the detection output so as to regard it as a complete start point of a new scene.

In the following, when a plurality of detection outputs continue,
A process for enabling only the last detection output and invalidating the preceding unnecessary detection output and performing digest recording (reproduction) will be described.

First, the timing at which change / motion detection is performed will be described with reference to FIG. FIG. 8A shows an explanation when a motion detection method is used for detection, and FIG. 8B shows an operation timing when a change is determined by comparing a pixel histogram. In the motion detection method of (A), as the input signal, the residual of the motion vector / block error is calculated for each block, using the signal of one frame before written in the frame memory as the reference signal. As described with reference to the drawing, the change / motion is determined using the vector value and the residual value. Therefore, it is determined whether or not there is a change / motion with respect to the frame 2 of the input signal.
It becomes clear only when the signal is input.

The result is the same when the histogram shown in FIG. Since the determination is made by comparing the histogram of the preceding frame (frame 1) and the histogram of the current frame (frame 2), the determination of frame 2 can be made only when the input of the current frame (frame 2) is completed.

In the present invention, when there are a plurality of changing points within a certain period, the preceding changing point is judged to be invalid. This judgment is made by the microcomputer 1 (20). As the simplest example, an example in which two frames are selected as the fixed period will be described. This means that if there is a continuous change, the preceding change point is invalidated. That is, the determination on the frame 2 is made on a separate sheet from information on the presence / absence of two changes, that is, the presence / absence of a change in the frame 1 of the frame 2 and the presence / absence of a change in the frame 2 of the frame 3.

[Table 1] It is judged like.

As can be seen from the timing relationship described with reference to FIG. 8, it is possible to determine whether or not there is a change in the frame 3 only when the input of the frame 3 is completed. Therefore,
The change information validity of the frame 2 becomes clear only when the input of the next frame (frame 3) is completed. With respect to the change of a certain frame (frame 2) as described above, the validity / invalidity as the change point is first determined when the next frame (frame 3) is completed. Information is also obtained that questions 2 and 3 are not change points (frames 2 and 3 are a series of consecutive scenes).

9 to 11 show examples of the configuration and recording timing of the representative screen detection circuit when performing digest recording while determining whether the change is valid or invalid. 9 to 11 show a case where there are 10 frames in which information of two scenes is mixed when switching from the scene B to the scene C, but FIG. 9 shows that frame detected as an effective change point. Is recorded as a representative screen as a still image (or as a moving image including at least that frame). The input signal is delayed by two frames by the two frame memos (45) and (46) and output to the encoder. This is because, as described with reference to FIG. 8 and Table 1, the validity / invalidity of the motion of a certain frame is found at the end time point of the next frame, and therefore it is preferable to use a delay of two frames.

FIG. 9 shows the input screen (input), the timing screen of the frame memory (1) and the output screen of the frame memory (2), and the time when motion is detected is indicated by a thick vertical line. Points that were not detected are indicated by crosses. Then, thick vertical lines BC1 to BC10 are fade-in / out regions, and it is determined that the screen B is changed to the screen C by disabling this part (marked by X). Therefore, effective detection points are indicated by the thick vertical line of C1.

FIG. 10 shows another embodiment in which the detection points are invalidated, and since the waveform chart is similar to that in the case of FIG. 9, its detailed description is omitted, but in this embodiment, the frame memory (2 ) Is omitted and a signal is output from the frame memory (1) to the encoder. As described in the description of Table 1, when the motion of a certain frame is determined to be valid, it means that there is no scene change between that frame and the next frame. Therefore, there is no functional problem in using the next frame in which an effective change / motion is detected as the representative screen.

FIG. 11 shows an example in which motion detection is performed by histogram detection. That is, the input video signal is alternately supplied to the two histogram counters (60) 61) for each frame (or field) via the switch means SW, and the count output is compared by the change point detection circuit (62). By doing so, the detection output (6) of the representative screen is obtained. Also in this case, motion detection outputs BC1 to BC10 (thick vertical lines) are obtained during the fade-in / out period in which the screen changes from the screen B to the screen C, but if the motion detection during this period is disabled, it can be seen at the bottom of the figure The effective motion detection output C1 can be obtained after the C screen is displayed. Then, when Cl is determined to be effective as a change point, a method of recording an image after C2 is shown. As a result, the frame memory can be reduced to one frame.

By the way, in general, one scene is rarely as short as 2-3 frames. Therefore,
Even when an effective change point is detected, it is sufficient to make specifications such that the change point (cl) and the next frame (c2) are not recorded and the subsequent points (c3 to) are adopted as the representative screen. Conceivable. Therefore, in general, FIG.
The input signal can be sent to the encoder as it is, as shown in, and the frame memory can be eliminated. As described in the description of FIG. 11, in general, one scene rarely ends in only a few frames, and as a representative screen, a still image (or a number of frames) several frames after the effective change / motion detection point is displayed. It is sufficiently practical to select a movie that starts after the frame).

By utilizing this fact, it is possible to make two consecutive judgments.
It is possible to improve the certainty of the determination by performing the determination using the detected outputs from more frames without being limited to the detection output from the frame. For that purpose, the determination criterion may be set to "if a change in one regular question (question in N frames) is not detected after the change is detected, the change point is set as an effective change point".

As an example, FIG. 12 exemplifies a case where the determination criterion is N = 3. In this example, scene B
Transition from scene to scene C over time is uneven,
As an example, an extremely bad condition is shown in which the degree of change temporarily decreases during the transition (marked with x). Also in this example, the change / motion detection output of each frame is known when the input of that frame is completed. The control method in which the change / motion information of each frame is reset when the next change / motion information is detected is the same as in the examples of FIGS.

Since N = 3 in the embodiment, effective motion detection is as follows. (A) Immediately after the frame input is completed, it is determined that a frame for which no change / motion is detected is not an effective change / motion point (“effective change / motion” at the bottom in the figure).
(Shown as (A)) (b) Regarding the frame in which the change / motion is detected at the time when the input of the frame is completed, the motion information regarding the frame is again detected at a time of 3 frames or more and 4 frames or less after the detection time. Judge by looking at the condition. (In the figure, a middle broken line L with an arrow indicates a state in which judgment is made 3.5 frames later, and is shown as (B) at the bottom stage).

As described above, if the next change / motion is detected within 3 frames, the change / motion information for that frame has been reset. After the detection, the change / motion information remains only when all the subsequent three frames have no change / motion, and in this determination process, the change point C1 is determined to be a valid change point C1.

In this embodiment, only the case of N = 3 is shown, but N is not limited to 3 and can be set to any value according to the design specifications. For any value of N, as described above, the timing of the judgment after the delay is
It may be selected such that “after N frames <t <(N + 1) frames after”.

Further, the present invention is characterized in that unnecessary change points are invalidated. However, even if the change points are not change points, the screen is recorded at appropriate intervals to improve the quality of the presented video during digest reproduction. It goes without saying that this may be applied when performing such processing.

In the above-described representative screen detection method, all the information is recorded while detecting the change / movement of the image at the time of recording, and the change / movement information detected at the time of recording is used when performing the digest reproduction. " The same process can be implemented in the digest reproducing device.

As the determination method, a method using information of two consecutive frames as described in Table 1 and FIG.
A method of exemplifying only a case where no change / motion is detected for a certain period (question of N frames) is detected after detecting a certain change / motion point, as described with reference to FIG. Further, in this case, as shown in the description of FIG. 3, it is not necessary to selectively record the signal at the time of recording. Therefore, change /
The motion information is not required to be synchronized with the video signal and real-time, and is recorded in the TOC area in the medium or the memory outside the medium by the end of the program recording operation.

A flow of detection and determination of effective change / motion information common to digest recording / digest reproduction will be described.
As shown in FIG. As shown in this flow, the change / detection detected by the representative screen detection circuit in step (100)
When the motion information is input to the microcomputer, it is first written in a predetermined memory as detection information, and at the same time, the previously written information is reset (101).

On the other hand, the detection time is set as the starting point, and the determination time after a fixed period is set (102). When the judgment time comes, the contents of a predetermined memory are read out based on the timer of the microcomputer. When the detection information is continuously input, the detection information in the memory is rewritten, and the determination time is reset at the same time (104), so that the determination time does not come easily. The judgment time comes and the judgment instruction is issued only when certain detection information is input and then one regular inquiry detection input does not come (105).

[0064]

The video signal recording device and the recording / reproducing device of the present invention detect the representative screen for the digest recording device, and thin out and record the representative screen so that the contents are grasped at the time of reproduction. The amount of recorded information can be reduced without losing the required screen, and therefore, the required capacity of the storage medium can be reduced, and when the capacity of the recording medium is constant, it is possible to perform recording for a longer time.

Further, by the control method which more positively utilizes the intermittent recording, it is possible to eliminate the number of expensive video encoders even in the simultaneous recording of multiple channels. However, due to the effect of lowering the band, the recording circuit such as ECC is not required to have a particularly wide band and expensive, and the cost can be reduced because it can be configured by one system.

Further, since the transfer rate required for the recording medium at the time of recording / reproducing is also reduced in proportion to the deterioration of the band, it can be realized at a low price without requiring a special recording medium driving device. Particularly, in the device of the present invention, even when a scene is switched over a plurality of frames for motion detection, one frame can always be designated as a start point of a new scene. One representative screen can be correctly set for one scene change regardless of the situation, unnecessary recording can be eliminated, and the capacity of the medium can be used more effectively.

When performing multi-channel recording,
Even if there is a scene change over multiple frames on a certain channel, one effective change point can be correctly set for each scene change, so that it is possible to avoid unnecessarily hindering recording on other channels. And so on.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an embodiment of a recording apparatus of the present invention.

FIG. 2 is a block diagram showing another embodiment of the present invention.

FIG. 3 is a block diagram of a video signal recording / reproducing apparatus of the present invention.

FIG. 4 is a block diagram illustrating a motion detection circuit.

FIG. 5 is an explanatory diagram showing a motion detection area.

FIG. 6 is an explanatory diagram for detecting movement between frames.

FIG. 7 is a diagram showing a motion detection signal in the case where scene change is performed smoothly.

FIG. 8 is an explanatory diagram showing a timing at which a motion detection output is obtained.

FIG. 9 is a motion detection waveform diagram when scene change is faded in / out.

FIG. 10 is a motion detection waveform diagram (one frame before) when scene change is faded in / out.

FIG. 11 is an explanatory diagram when performing motion detection using a histogram.

FIG. 12 is an explanatory diagram for performing detection when there is motion detection over three frames.

FIG. 13 shows a flow chart of a microcomputer for detecting motion.

[Explanation of symbols]

 1 Antenna 2 Video Channel Tuner 4 AD Converter 5 Representative Screen Detection Circuit 13 Audio Tuner 18 Multiplexer 20, 37 Microcomputer 22 Storage Device 24 Demultiplexer 25 Video Signal Decoder 25 Audio Signal Decoder 38 Memory 40 Storage / Drive Circuit

Claims (9)

[Claims] 1. A representative screen to be recorded from the received video signal on the basis of a detection means for receiving a video signal, a motion detection means for detecting a motion of the screen from the received video signal, and a detection result of the motion detection means. And a recording means for recording the video signal obtained from the representative screen extracting means together with the extracted information, wherein the motion detecting means detects a plurality of changes within a predetermined period. The video signal recording apparatus according to claim 1, further comprising means for invalidating the previously detected change point and validating the last change point within the period. 2. The video signal recording apparatus according to claim 1, wherein the representative screen is a still image in units of at least one independent field (frame) selected by the motion detecting means. 3. The video signal recording apparatus according to claim 1, wherein the representative screen is one in which moving images of several consecutive fields are selected by the motion detecting means. 4. The motion detecting means is set so as to reduce the number of thinned-out screens when a large-movement screen continues and lengthen the thinned-out interval when a small motion screen continues. The video signal recording device according to claim 1, 2, or 3. 5. The video signal recording device according to claim 1, wherein the receiving device is a device capable of simultaneously receiving broadcast signals of multiple channels. 6. A representative screen selected from the received video signal on the basis of the detection means of the reception means for receiving the video signal, the motion of the screen from the received video signal, and the detection result of the motion detection means. A control means for generating extraction information indicating the recording position of the recording medium, a recording means for recording the video signal together with an audio signal and the extraction information, and a reproduction means capable of reading the information recorded in the recording means by a predetermined operation. A video signal recording / reproducing apparatus, characterized in that the information read by the reproducing means is configured to be a representative screen selected by the extracted information. 7. The video signal recording / reproducing according to claim 6, wherein the representative screen is a still image in units of at least one field (frame) selected by the motion detecting means. apparatus. 8. The video signal recording / reproducing apparatus according to claim 6, wherein the representative screen is one in which moving images of several consecutive fields are selected by the motion detecting means. 9. The representative screen is set so as to reduce the number of thinned-out screens when a large motion screen continues, and to lengthen the thinned-out interval when a small motion screen continues. The video signal recording / reproducing apparatus according to claim 6, 7, or 8.