JPH09168148A - Dynamic image coding method and device and recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve the coding efficiency of a dynamic image signal by recording information of a position at which a frame rate is subjected to change to a recording medium in which dynamic image sources with different frame rates are mixed and by revising coding processing through the detection of the information. SOLUTION: A VTR 2 reproduces an image signal with a frame rate 24Hz obtained from a film source through 3:2 pull-down processing. A VTR 3 reproduces an image signal whose frame rate is 30Hz picked up by a television camera. A VITC insert circuit 6 inserts the information to a signal for a vertical blanking period based on a VITC by using a user bit of, e.g. an SMPT time code as edit point information where a frame rate is switched attending the changeover of the VTRs 2,3 by a switch 4. An image coder 10 uses a VITC read circuit 12 to recognize an image obtained by the 3:2 pull-down processing based on the edit point information, a redundant field detection and elimination device 14 is inserted to the circuit to eliminate the image so that the redundant field is not coded.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a moving picture coding method and apparatus for coding a moving picture containing a redundant picture, and more particularly to a method and apparatus for optically / electrically converting an original picture source such as a movie film. The present invention relates to a moving picture coding method and apparatus for coding the obtained moving picture signal, and a recording medium for recording the moving picture signal including a redundant image.

[0002]

2. Description of the Related Art Generally, since a digital image signal has an extremely large amount of information, when it is desired to record this on a recording medium having a small size and a small amount of stored information for a long time, a means for highly efficient encoding the image signal and recording it. Is essential. In order to meet such a demand, a high-efficiency coding method utilizing the correlation of image signals has been proposed, one of which is so-called MP.
There is an EG (Moving Picture Expert Group) system. Note that MPEG means ISO (International Organization for Standardization) and IEC.
(International Electrotechnical Commission) JTC (Joint Technical Commi
ttee) 1 SC (Sub Committee) 29 WG (Workin
g Group) 11 is a common name for moving image coding methods.

In the above-mentioned MPEG system, the redundancy in the time axis direction is reduced by first taking the difference between the frames of the image signal, and then the discrete cosine transform (DCT).
This is a method for efficiently encoding an image signal by reducing the redundancy in the spatial axis direction using an orthogonal transformation method such as inetransform).

FIG. 17 shows a conventional example of a moving picture coding apparatus in which an image signal obtained from a film source having a frame rate of 30 Hz by a so-called 3: 2 pulldown process is used as an original input moving picture signal. There is. This FIG.
In encoding the image signal obtained from the film source by the 3: 2 pull-down processing, the moving image encoding apparatus described above reduces the redundancy in the time axis direction and the redundancy in the spatial axis direction and compresses the same. The redundant image is not encoded to further improve the compression efficiency.

The 3: 2 pull-down process will be briefly described. In the case of converting an image of a film source such as a movie into an interlaced scan image signal corresponding to, for example, the so-called NTSC system of a television broadcasting system (that is, telecine conversion), the image of the film source is 24 frames per second. Since the NTSC system interlaced scan image signal is 30 frames (60 fields) per second, the above-mentioned film has 2 frames per second.
In order to generate 30 frames (60 fields) per second of an interlaced scan image signal from four frames, a field number conversion process is required. Therefore, when converting an image of a film source of 24 frames per second into an interlaced scan image signal of 30 frames (60 fields) per second, as shown in FIG. 18, two continuous frames of film, for example, frames MF1 and MF2 are used. First frame MF1
Is converted into two fields of the interlaced scan image signal, and the next frame MF2 is converted into three fields. In the interlaced scan image signal obtained by such 3: 2 pulldown processing, the redundant image is a repeated field among three fields obtained from the same frame of the film source shown in FIG. Is.

A detailed description of the configuration shown in FIG. 17 will be given below.

A video tape recorder (VTR) 101 is loaded with a video tape recorded with an interlaced scan image signal converted from an image of a film source by the 3: 2 pulldown process. The reproduced image signal is sent to the redundant field detector and remover 102 as the original input moving image signal.

The redundant field detector and remover 102.
Then, the image signal corresponding to the redundant image is detected from the image signal, and the detected image signal is reduced in order not to encode the image signal of the redundant image. That is, three fields obtained from the same frame of the film source are detected from an interlaced scan image signal of 30 frames per second, and redundant repeated fields (hereinafter referred to as redundant fields) are removed from the three fields. As a result, a progressive scan frame of 24 frames / second (frame by progressive scanning) is ideal.
Will be created. FIG. 19 shows an example when the progressive scan frame of 24 frames / second can be ideally created.

As can be seen from FIG. 19, the principle of the redundant field detection algorithm is as follows. First, two consecutive fields of the first field (top field) or the second field (bottom field) are the same image (repeated). Field image) (that is, whether two consecutive fields having the same parity are the same image). Here, if two consecutive fields of the first field or the second field are repeated, then ideally these two fields should exactly match, but in reality this is not the case. That is, usually, after the 3: 2 pulldown processing, the signal level is smoothed by the smoothing filter in the time axis direction, that is, between the fields and between the frames in order to smooth the motion of the image, so that the pixel level changes. Because it has been lost.
Generally, most of the original images of movie programs supplied from post-production companies that perform telecine conversion have been smoothed.

Therefore, when detecting the redundant field described above, generally, a threshold value is set for the degree of equality between two consecutive fields of the first field or the second field to perform the redundant field determination. Has been done. For example, when the sum of the absolute values of the differences between the pixels in the continuous two fields of the first field or the second field is smaller than a predetermined threshold value, it is determined to be a redundant field. Here, if it is determined that the field is a redundant field, that field is removed from the original input image signal, and this is not coded to reduce the data. The details of the redundant field detection and removal algorithm will be described later.

Redundant field detector and remover 102
The image signal of the field sequence output from is converted by the scan converter 103 into an image signal of the frame sequence in the order of input as shown in FIG. Here, since the constructed frame corresponds to the same frame of film, it can be treated as a progressive scan frame. That is, this frame is equal to one frame of the image signal read out by progressively scanning one frame of the film.
In general, a progressive scan frame has a higher correlation between lines in the vertical direction than an interlaced scan frame, and thus has a higher redundancy and can improve frame coding efficiency.

The image signal of the progressive scan frame output from the scan converter 103 is sent to the encoder 104. In the encoder 104, for example, the scan converter 1 according to the MPEG system, which is a high-efficiency encoding system utilizing the correlation of the image signals.
The image signal of the progressive scan frame output from 03 is compression-encoded. At this time, as described above, since the image signal of the progressive scan frame has a large correlation between the lines in the vertical direction, it is possible to obtain a higher coding efficiency than that of the image signal of the interlaced scan frame.

The image signal encoded by the encoder 104 is then recorded on the recording medium 105.

As described above, in the conventional moving image coding apparatus, all the image signals recorded on the video tape are 3: 2.
If the image signal obtained by pull-down processing and the signal obtained by reproducing the video tape with the video tape recorder 101 is obtained by the 3: 2 pull-down processing, encoding of the image signal of the frame Efficiency is good and there is no problem.

[0015]

By the way, in the conventional moving picture coding apparatus shown in FIG. 17, the image signal recorded on the video tape loaded in the video tape recorder 101 is the above-mentioned 3 :. The image signal obtained by the 2: 2 pull-down process is not necessarily composed of only the image signal obtained by the 2-pull-down process. For example, the image signal obtained by the 3: 2 pull-down process and the image signal captured by the television camera (frame rate is 30 Hz) , Which may be combined by, for example, editing. That is, as an example of this, for example, a case where a program for television broadcasting in which commercial images shot by a television camera are inserted is recorded in several places in the middle of a movie program, and the like.

A video tape recorded with a mixture of image signals obtained by the 3: 2 pull-down process and image signals taken by a television camera is recorded by the conventional moving picture coding apparatus shown in FIG. Think about the case.

First, when the reproduced image signal reproduced from the video tape is a signal portion obtained from a movie program, a redundant repeated field is detected and the redundant field is detected as described above. Remove it
Must be unencoded. Further, in this signal portion, the frame rate after encoding is ideally 2
Must be 4 Hz.

Next, when the reproduced image signal is a signal portion photographed by a television camera, since the redundant field does not exist in the signal portion, the image signals of all fields must be encoded. . Also,
In this signal part, the frame rate after encoding is 30H.
must be z.

However, as described above, when detecting a redundant field, the degree of equality of signals between two fields, for example, the difference of each pixel between two consecutive fields of the first field or the second field. Since a redundant field is determined by comparing the sum of absolute values of the above with a predetermined threshold value, even if the reproduced image signal is an image signal portion captured by a television camera, for example If it is an image signal of a small scene, it may be erroneously determined as a redundant field.

In general, as the smoothness of the smoothing process applied to the image obtained by the 3: 2 pulldown process becomes stronger, that is, the smoothing filter converts the image signal in the time direction (between fields and frames). As the degree of filtering applied increases, the pixel value (pixel level) changes between the two fields even in the repeated field, and the sum of absolute values of the differences between the two fields increases. In such a case, it is difficult to determine the repeated field as the redundant field. Therefore, in order to efficiently determine the redundant field, the absolute value of the difference for performing the redundant field determination is usually used. Increase the threshold for the sum.

However, if the threshold value is increased, even if the reproduced image signal is a signal portion photographed by a television camera, it may be erroneously determined as a redundant field in the image signal of a scene in which the motion of the image is small. The problem arises that the quality becomes higher. As described above, when the reproduced image signal is a signal portion captured by the television camera, if it is erroneously determined as a redundant field, the field determined to be the redundant field is an originally necessary field. Nonetheless, it is removed, and the frame rate of encoding is also 30H.
This is less than or equal to z, resulting in a problem that the motion of a moving image obtained later becomes unnatural. Further, the correlation between fields and frames is also disturbed, which causes a problem that the coding efficiency of the image signal of the frame is lowered.

On the contrary, in order to prevent the redundant field from being erroneously determined for the image signal captured by the television camera, the redundant field determination criterion is made strict, that is, the difference between the two fields. If the threshold of the sum of absolute values is made too small, the detection efficiency of the redundant field with respect to the image signal by the 3: 2 pulldown process will be lowered. Further, if the redundant field detection efficiency decreases in this way, the frame rate for encoding the image signal by the 3: 2 pull-down processing also becomes 24 Hz or higher, and thus the encoding efficiency of the image signal of the frame also decreases. Occurs.

Therefore, the present invention has been made in view of such circumstances, and it is like a signal in which an image signal obtained by 3: 2 pulldown processing and an image signal taken by a television camera are mixed. A moving image coding method and device capable of efficiently coding a moving image signal in which image signals obtained from moving image materials having different coding frame rates are mixed in a sequence of image signals, and a recording medium. The purpose is to provide.

[0024]

A moving picture coding method and apparatus according to the present invention are provided when an image signal obtained from moving picture materials having different coding frame rates is mixed in a sequence of image signals. The position information in which the coding frame rate changes is detected in the sequence of, and the above-described problem is solved by changing the coding process of the moving image signal based on the position information.

In the recording medium of the present invention, a moving image signal in which image signals obtained from moving image materials having different coding frame rates are mixed in a sequence of image signals and a sequence of image signals are encoded. By recording the position information indicating the position where the digitized frame rate changes, the above-mentioned problem can be easily solved.

[0026]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will be described below with reference to the drawings.

First, a first embodiment for realizing the moving picture coding method of the present invention will be described with reference to FIG. FIG. 1 shows the configuration of an image coding system to which the first moving image coding apparatus of the present invention is applied.

The image coding system of FIG. 1 is roughly classified into
A video editing device 1 that edits image signals from two video tape recorders (VTRs 2, 3), and an image that encodes the image signal from the video editing device 1 as an input image signal to generate encoded data. And an encoding device 10.

The video editing apparatus 1 reproduces the signal (encoding frame rate is 24 Hz) from a video tape recorded with an image signal obtained by 3: 2 pulldown processing from a film source (for example, a movie source). V
TR2, a VTR3 that reproduces the signal (encoding frame rate is 30 Hz) from a video tape on which an image signal captured by a television camera is recorded, and V of these two units.
A switch 4 for switching and selecting signals reproduced from TRs 2 and 3, and an image signal S12 selected by the switch 4.
Is recorded on a video tape, and a video editing controller 5 for controlling the switch 4.

The video editing controller 5 sends a switching flag S11 to the switch 4 to control the switching of the switch 4 and also sets the switching flag S11 to V.
It is also output to the TR7 side. The flag S11 is recorded on the video tape in the VTR 7 together with the image signal S12 as header information of the corresponding image signal S12 selected by the switch 4. That is, the image signal S12 and the flag S11 are recorded on the video tape 9 by the VTR 7 in one-to-one correspondence with each image selected by the switch 4.

Here, as a recording method of the flag S11 to be recorded together with the image signal S12, for example, SMPT
There is a method of recording on the tape together with the image signal by using the user bit of the E time code. In addition, SMPT
E time code is American standard (C98.12: time and co
ntrol code for video and audio tape for 525/60 tel
evision system) is the abbreviation of the code specified in. I
According to the EC standard, Publication 461 (time and control code
625/50 television as for video tape recordings)
It is generalized as a standard including system. In particular,
The flag S11 is added to so-called VITC (vertical interval time code) and LTC (longitudinal time code).
To record. The VITC is a time code in the vertical blanking period. In the 1H period, 64 bits of the time code data are divided into 8-bit units and synchronization bits (“1”, “0”) are added, and 8-bit CR
It is configured by 90 bits as shown in FIG. 2 to which a C code is added, and such a signal is inserted in 2H not adjacent to each other in the vertical blanking period of each field. In particular, in the case of an NTSC television signal, the SMPTE is 1
We recommend putting them on the 4th and 16th lines. As the position of the horizontal line signal, as shown in FIG. 3, it starts from a position 10 μs behind the leading edge, and is 50.286 before 3.269 μs before the next leading edge.
Recorded within μs. Signal level is "0" of data
Is 0 IRE (Institude for Radio Engineers), ”
1 "is recorded at a level of 80 IRE. Also, L
TC is a time code recorded in the tape longitudinal direction, and is composed of 64-bit time information and 80-bit data of user bits and 16-bit sync bits as shown in FIG. It is recorded on the time code track. As shown in FIGS. 2 and 4, the VITC and LTC include user bits in addition to the time code, and the flag S is added to the user bits.
11 can be recorded.

In order to do this, the image signal S12 obtained by editing and combining the image signal obtained by the 3: 2 pull-down process from the switch 4 and the image signal taken by the television camera is sent to the VTR 7. To the VITC insertion circuit 6. At the same time, the flag S11 is also supplied to the VITC insertion circuit 6, and the flag S11 is used for editing point (position) information of the image signal obtained by the 3: 2 pulldown processing and the image signal captured by the television camera. As the user bit of, for example, VITC of the image signal S12.

From the above, on the video tape 9 recorded by the VTR 7, the image signal by the 3: 2 pull-down process and the image signal taken by the television camera are edited and combined (S12). Then, the image signal S13 in which the signal (flag S11) instructing the edit point (position) information is inserted into the user bit of VITC is recorded.

The VITC and LTC may have a bit configuration other than the formats shown in FIGS. 2 and 4, and in this case, it is sufficient that the recording side and the receiving side have a correspondence. Further, the image signal S12 and the flag S11 described above do not have to be recorded on the same recording medium, but may be recorded on different media. For example, the floppy disk 8 shown in FIG. 1 may be recorded. In this case, for example, the image signal S12 is recorded on the video tape in the VTR 7, and the flag S11 (that is, edit point information) is recorded on the floppy disk 8. You can do things like record.

The video tape 9 on which the signals as described above are recorded is reproduced by the VTR 11, and the image signal S7 obtained by the reproduction by the VTR 11 is the image encoding device 10.
Sent to The image signal S7 reproduced from the video tape 9 is the same as the image signal S13.

The image coding apparatus 10 which has received the image signal S7 from the VTR 11 receives the image signal S7 from the image signal S7.
An image signal (the image signal S12) obtained by editing and combining the image signal obtained by the 3: 2 pull-down processing and the image signal captured by the television camera is read out, and the user bit of the SMPTE time code of the image signal, for example, the VITC. The edit point information (the flag S11) of the image recorded in (1) is read, and the encoding process at the time of encoding the image signal is controlled based on the edit point information. In other words, the edit point information is the encoding control information, and the image encoding device 10
In the above, in accordance with the edit point information (encoding control information), separate encoding processing is performed on the image signal by the 3: 2 pulldown processing and the image signal captured by the television camera. I have to.

The image signal S7 supplied from the VTR 11 to the image encoding device 10 is first sent to the VITC reading circuit 12. In the VITC reading circuit 12,
From the image signal S7, the edit point information (that is, the flag S11) arranged in the user bit of the SMPTE time code, for example, VITC is read out, and the above-mentioned 3:
The image signal obtained by the 2 pulldown processing and the image signal captured by the television camera are separated from the edited and combined image signal (the same signal as the image signal S12), and the separated image signal is used as the image signal S1 in the switch 13 and thereafter. Send to the route. On the other hand, the VITC reading circuit 12 determines whether the image signal S1 is an image signal by the 3: 2 pulldown process or an image signal captured by a television camera, based on the edit point information read from the VITC. Of the flag S2 indicating the signal type (the flag S11
Flag corresponding to) is output. The flag S2 is sent to the switches 13 and 15 as a switching control signal and also to the encoder 17. Further, the VITC reading circuit 12 outputs a field synchronization signal (S608) described later together with the image signal S1.

The switch 13 is for switching the switched terminals c or d according to the flag S2. For example, the switch 13 indicates that the image signal S1 is captured by the television camera. When the flag S2 indicates, the terminal is switched to the switched terminal c side, and when the flag S2 indicates that the image signal S1 is due to the 3: 2 pulldown process, the terminal is switched to the switched terminal d side. Further, the switch 15 also switches the switched terminals e or f in accordance with the flag S2, like the switch 12, and the image signal S
When the flag S2 indicates that 1 was taken by the television camera, the flag S2 indicates that the image signal S1 is switched to the switched terminal e side and the image signal S1 is due to the 3: 2 pulldown process. , The terminal is switched to the switched terminal f side. The switched terminal c of the switch 13 and the switched terminal e of the switch 15 are directly connected, the switched terminal d of the switch 13 is an input terminal of the redundant field detector and remover 14, and the switched terminal f of the switch 15 is connected. Is redundant field detector and remover 1
4 is connected to the output terminal. The output signal from the switch 15 is input to the scan converter 16 as an image signal S5. Therefore, the image signal S1 when the flag S2 indicates that the image was taken by the television camera is directly sent to the scan converter 16 as the image signal S5, and the 3: 2 pull-down process is performed. The image signal S1 when the flag S2 indicates that it is due to the above is processed by the redundant field detector and remover 14 and then sent to the scan converter 16 as the image signal S5.

The redundant field detector / remover 14 detects an image signal corresponding to a redundant image from the image signal S1 obtained by the 3: 2 pull-down process selected by the switch 13, and detects the image of the redundant image. This detected image signal is reduced in order not to encode the signal. The redundant field detector / remover 14 outputs an image signal after the redundant image (redundant field) is removed, and a redundant field detection flag (S611), which will be described later, indicating the redundant image (redundant field). To be done. The specific configuration and operation of the redundant field detector / remover 14 will be described later.

In the scan converter 16, the image signal photographed by the television camera and the image signal and the redundant field detection flag (S611) after being processed by the redundant field detector / remover 14 are the flags. The field sequence image signal S5 switched and supplied in S2 is converted into a frame sequence image signal in the order of input. From the scan converter 16, together with the image signal of the frame sequence, a flag S101 indicating the start timing of a pair of fields constituting a frame and the frame starting from the first field or the second field, which will be described later, are given. And a flag S103 indicating that the frame is composed of two fields obtained by removing one field from three fields originally generated from the same frame of the film source. Detailed operation of the scan converter 16 will be described later.

The image signal S6 of the frame sequence output from the scan converter 16 and the flags S101, S102, S103 are sent to the encoder 17. In the encoder 17, when encoding the image signal S6 of the frame sequence, the image signal obtained by the 3: 2 pulldown process and the image signal captured by the television camera are used on the basis of the flag S2 and the flag S101. For each frame, a different encoding process is performed. Further, the encoder 17 uses the flags S2 and S102.
Also, S103 and S103 are encoded as described later, and are output together with the encoded image signal. The specific configuration and operation of the encoder 17 will be described later.

The encoder 17 encodes the image signal with high efficiency, and at the same time, the flags S2, S102, S103.
Coded data obtained by encoding (bit stream)
Is then recorded on the recording medium 18.

For example, when the flag S11 (editing point information) is recorded on the floppy disk 8 shown in FIG. 1, the VITC reading circuit 12 corresponds to the flag S11 read from the floppy disk 8. It is also possible to output the flag S2. Of course, in this case, instead of outputting the flag S2 from the VITC reading circuit 12, the flag S11 read from the floppy disk 8 may be sent as it is to the respective units (the switches 13, 15 and the encoder 17) as the flag S2. Good.

Next, the configuration and operation of the main part of the redundant field detector and remover 14 will be described with reference to FIG.

In FIG. 5, the input terminal 501 has
The image signal S1 by the 3: 2 pull-down process selected by the switch 13 is supplied. The image signal S1
Is a signal having a field rate of 60 Hz reproduced from the video tape 9 on the VTR 11.

The image signal S1 obtained by the 3: 2 pulldown processing is delayed by the delay devices 502 and 503.
The addition signal S60 is added to the adder 504 after being delayed by the field.
Entered as 2. Further, the adder 504 has an undelayed image signal S1 from the input terminal 501.
Is input as a subtraction signal, and therefore, the adder 504 outputs the image signal S delayed by the two fields.
Calculation is performed to subtract the undelayed image signal S1 from 602 for each pixel.

The difference value S603 calculated by the adder 504 for each pixel is sent to the absolute value calculator 505, where the absolute value of the difference value S603 is calculated. Absolute value S6 calculated by the absolute value calculator 505
04 is sent to the accumulator 506, where the cumulative sum per field is calculated. The cumulative value S605 calculated by the accumulator 506 is sent to the comparator 507.

A predetermined threshold value S606 stored in advance in the memory 508 is supplied to the comparator 507, and the comparator 507 compares the threshold value S606 with the cumulative value S605. When the accumulated value is smaller than the threshold value in the comparison by the comparator 507, "1" is set as the flag S607 in the output of the comparator 507. The output terminal of the comparator 507 is a 2-input AN.
It is connected to one input terminal of the D calculator 512.

On the other hand, the input terminal 509 is supplied with the field sync signal S608 of the image signal S1 of the field rate input to the input terminal 501. Although not shown in FIG. 1, the field synchronization signal S608 is supplied from, for example, the VITC reading circuit 12.

The field synchronization signal S608 from the input terminal 509 is sent to the field counter 510. The field counter 510 counts the number of fields of the image signal input to the redundant field detector / remover 14 by counting the field synchronization signal S608. The field counter 51
The count value j of the number of fields obtained by counting at 0 is sent to the comparator 511 as the output signal S609.

In the comparator 511, the output signal S6 is output.
When the count value j indicated by 09 is an odd number of 5 or more, "1" is set as the flag S610 at the output of the comparator 511. The output terminal of the comparator 511 is connected to the other input terminal of the 2-input AND operator 512.

In the AND operator 512, both the flags S610 and S607 to the two input terminals are "1".
When becomes, "1" rises in the output. This AN
When "1" rises in the output of the D calculator 512, the output of the AND calculator 512 indicates that the field of the image signal S1 currently input is a redundant field in which the field is duplicated by the 3: 2 pull-down process. The redundant field detection flag S611 is output from the terminal 513.

When "1" is set as the redundant field detection flag S611, the redundant field detection and removal circuit 14 removes the image signal of the field. For example, as a specific configuration for removing the image signal of the redundant field, based on the switch 514 to which the image signal S1 is supplied via the input terminal 501, the redundant field detection flag S611, and the field synchronization signal S608. A configuration including a control circuit 516 that outputs a switching control signal of the switch 514 and the like can be given as an example. That is, when "1" is set in the redundant field detection flag S611, the control circuit 516 controls the switch 514 to be OFF only during the redundant field based on the field synchronization signal, so that the current supply is performed. It is possible to remove the image signal of the existing field as the image signal of the redundant field. The image signal from which the redundant field has been removed is output from the terminal 515 as an output image signal of the redundant field detector / remover 14. Of course, the configuration for removing the image signal in the redundant field is not limited to the configuration including the switch 514 and the control circuit 516, but may be another configuration.

The output image signal and the redundant field detection flag S611 obtained as described above are sent to the scan converter 16 via the switch 15 of FIG.

The output terminal of the AND operator 512 is also connected to the clear terminal of the field counter 510, and therefore the field counter 510 concerned.
Indicates that the count value is cleared when "1" is set in the flag S611.

In addition, the judgment standard in the comparator 511 is
As described above, the reason why the count value j is an odd number of 5 or more is that it is guaranteed that the redundant field detection cycle by the 3: 2 pulldown process always operates regularly for the following reason. Because it is not done. That is, firstly, the pattern in which a redundant field appears in a 5-field cycle cannot be guaranteed due to, for example, video editing after the 3: 2 pulldown process. Also the second
In addition, during the 3: 2 pulldown process, the smoothing process is performed in the time axis direction, that is, between the fields and between the frames, so that it is difficult to detect the redundant field depending on the pattern of the image. This is because, for example, even if the field is actually a redundant field, "1" may not be set in the flag S607 in the comparison by the comparator 507.
Therefore, by using the above-described determination standard in the comparator 511, even if the 3: 2 pull-down processing pattern is not guaranteed, the comparator 511 can continue to determine the redundant field.

Next, referring to FIG. 6, the VITC of FIG.
The operation of the configuration from the reading circuit 12 to the scan converter 16 will be described.

In FIG. 6, the image signal S1 of the field rate output from the VITC reading circuit 12 of FIG. 1 and whether the image signal S1 is obtained by the 2: 3 pulldown process or in a television camera. The flag S2 indicating the signal type indicating whether it is obtained by photographing.
And the redundant field detection flag S611 described in FIG.
1 and the output image signal S5 from the switch 15 of FIG. 1 and the start timing of the pair of fields ("1" rise) when the scan converter 16 converts the field rate into the frame rate. ) Indicating that the frame is the first field
(top field) or the second field (bott
om field), and a flag S102 that is "1" when starting from the first field, and 2 fields obtained by removing 1 field from 3 fields originally generated from the same frame of the film source. Flag S1 which becomes "1" when the frame is composed of
03 is shown. In addition, the image signal S1 in FIG.
, The letter F or f is the image signal of the film source subjected to the 3: 2 pulldown process, and the capital letter F is the first.
Represents the field, lowercase f represents the second field,
The same number among the letters F and the numbers of the subscripts of f indicates the fields read from the same frame of the film. Further, in the image signal S1 in FIG. 6, the letter V or v is the image signal captured by the television camera, the uppercase letter V represents the first field, the lowercase letter v represents the second field, and these letters V and v The same numbers in the subscripts of indicate the pairs that make up the frame.

As can be seen from FIG. 6, the image signal S1
Is a film source subjected to 3: 2 pulldown processing, the flag S2 is "1", and the image signal S
When 1 is captured by the television camera, the flag S2 is "0". Therefore, in the configuration of FIG. 1, when the flag S2 is "1", the switched terminals d and f of the switches 13 and 15 are selected, and the redundant field detector and remover 14 for the image signal S1 is selected. Is applied, and the switched terminals c and e of the switches 13 and 15 are selected, and the redundant field detector and remover 14 is not used for the image signal S1.

Further, as can be seen from FIG. 6, the flag S611 of FIG. 5 becomes "1" when the field of the image signal S1 is a redundant field. That is, the flag S
When 2 is "1", redundant field detector and remover 14
Detects a redundant field read in 3 fields from the same frame of the film source, the flag S
611 is set to "1". Then, of the fields of the image signal S1, the field in which the flag S611 is "1" is removed from the image signal S1, and the switch 15 outputs the image signal S5 of the field sequence. On the other hand, while the flag S2 is "0", the redundant field detector / remover 14 is not used.

Further, in FIG. 6, the image signal S of the frame rate is output from the scan converter 16 of FIG.
6 and the above-mentioned flags S102 and S103 attached thereto.
Is also output to the encoder 17, and these image signals S6
And flags S102 and S103 are input. When the encoder 17 performs encoding according to the so-called MPEG2 standard, the flags S102 and S103 are the top field first (TFF) and the repeat first defined in MPEG2, respectively.・ Field (repeat_first_field:
RFF). In the case of a frame structure, the top field first is information indicating whether the first field is higher or lower, and the repeat first field is 2: 3.
This is information used when pulling down.

Next, the configuration and operation of the encoder 17 of FIG. 1 will be described below.

The encoder 17 is a so-called MPEG2.
It is assumed that hybrid coding processing is performed by combining motion compensation predictive coding and discrete cosine transform (DCT), which are widely known in (ISO / IEC 13818-2) and the like.

In MPEG2, the image of each frame is an I picture (Intra coded picture), a P picture (Predictive coded picture), and a B picture (Bidire).
ctioally predictive coded picture), and the signals of these pictures are compression-encoded.

That is, in MPEG2, as shown in FIG. 7, for example, image signals of 17 frames from frames F1 to F17 are grouped into a group of pictures (Group Of P).
icture: GOP) as one unit of processing.

For example, the first frame F1 of the GOP is set to I
It is processed as a picture, the second frame F2 is processed as a B picture, and the third frame F3 is processed as a P picture. Hereinafter, the fourth and subsequent frames F4 to F17 are alternately processed as a B picture or a P picture. Note that, in FIG. 7, the arrow from the picture to the picture indicates the direction of prediction (the same applies hereinafter).

More specifically, in the I picture, the image signal for one frame is directly encoded and transmitted. In the P picture, basically, as shown in FIG. 7A, the difference from each pixel of the I picture or P picture that is temporally past is obtained, and this difference signal is encoded and transmitted. In the B picture, basically, as shown in FIG. 7B, the difference between the average value of each pixel of both the past and future frames is obtained, and this difference signal is encoded. To transmit.

The principle of the method for encoding the image signal of the moving image in this way will be described with reference to FIG.

In FIG. 8, since the first frame F1 is processed as an I picture, the transmission data F1 remains unchanged.
X is transmitted to the transmission path (intra-picture coding).

On the other hand, the second frame F2 is B
Since it is processed as a picture, the difference between the average value of the frame F1 temporally in the past and the average value of the frame F3 temporally in the future is calculated, and the difference data is calculated as the transmission data F.
It is transmitted as 2X. However, there are four types of processing as the B picture, which will be described in more detail. The first process is to transmit the data of the original frame F2 as it is as the transmission data F2X as indicated by an arrow SP1 in the figure (that is, intra coding), and is the same process as in the I picture. The second process is to calculate a difference from the frame F3 that is temporally future and transmit the difference data as indicated by an arrow SP2 in the figure (that is, backward prediction encoding). The third process is to calculate a difference with respect to the temporally past frame F1 and transmit the difference data as indicated by an arrow SP3 in the figure (that is, forward predictive coding). Further, in the fourth processing, as indicated by an arrow SP4 in the drawing, a difference between the temporally past frame F1 and the average value of the future frame F3 is generated, and this difference data is transmitted as transmission data F2X. (Ie bidirectional predictive coding). Among these four methods, the method that minimizes the amount of transmission data is adopted. When transmitting the difference data, an image of the frame (prediction image) for which the difference in forward prediction is calculated
And a motion vector x1 (motion vector between frames F1 and F2), or a motion vector x2 (motion vector between frames F3 and F2) in backward prediction, or a motion vector x1 in bidirectional prediction. Both x2 are transmitted with the difference data.

Further, the frame F3 of the P picture is a frame F1 which is temporally past as indicated by an arrow SP3 in the figure.
Is used as a prediction image, the difference from this frame and the motion vector x3 are calculated, and these are transmitted as transmission data F3X (that is, forward predictive coding). Alternatively,
As indicated by an arrow SP1 in the figure, the data of the original frame F3 is directly transmitted as the transmission data F3X (intra-coding). Which of these methods is used for transmission is B
As with pictures, the one with less transmitted data is selected.

Next, referring to FIG. 9, the encoder 17
The specific configuration of will be described.

In FIG. 9, the image signal S6 of the frame rate from the scan converter 16 of FIG. 1 and the flags S101, S102, S are input terminal 74.
103 is input, and the input terminal 75 is input with the flag S2 indicating whether it is the image signal of the film source subjected to the 3: 2 pulldown processing or the image signal captured by the television camera. . This flag S2 is used for a motion vector detection circuit 50, a prediction mode switching circuit 52, a DCT mode switching circuit 55, which will be described later.
It is sent to the variable length coding circuit 58.

The image signal S6 supplied through the input terminal 74 and the flags S101, S102, S103.
Is input to the image coding type designation / image coding order rearranger 70. Here, first, it is designated which of the I picture, the P picture, and the B picture each frame of the image signal S6 of the sequentially input frame rate is to be processed. For example, as shown in FIG. 7, the group of pictures composed of the frames F1 to F17 are I pictures, B pictures, P pictures, B pictures, P pictures, ...
In order to process the B picture and the P picture in order, the image coding type is designated for each frame. The designated image coding type is written in the header of the image signal of each frame.

Next, when the image coding type of each frame is specified as described above, the image coding type is specified and
The image coding order rearranger 70 rearranges the image signals of the respective frames in the order in which they are coded according to the designated image coding type. This is because the B picture needs backward prediction and cannot be decoded unless an I picture or P picture as a backward predicted image is prepared in advance. That is, since the I-picture or P-picture image signal must be encoded first before the B-picture image signal is encoded, the circuit 70 rearranges the order of the frames. For example, in the example of FIG. 7, when the image coding type is specified, the order of frames is frame F1, frame F3, frame F
2, frame F5, frame F4, ...

The rearranged image signal S502 output from the image encoding type designation / image encoding order rearranger 70 is input to the scan converter 71. From the image coding type designation / image coding order rearranger 70, the flags S101 and S1 are sent.
02 and S103 are also output, and as will be described later, the flag S101 is sent to the frame memories 51 and 63, the motion vector detecting circuit 50, and the variable length coding circuit 58, and the flags S102 and S103 are variable length coding. Sent to circuit 58.

The scan converter 71 converts an image signal input by raster scan into a block format signal in MPEG. That is, FIG.
As shown in (A) of 0, the image signal input by the raster scan is frame format data in which V dots of H dots are collected per line. The scan converter 71 divides this 1-frame signal into N slices with 16 lines as a unit, as shown in FIG. Then, each slice is divided into M macroblocks as shown in FIG. Each macroblock is composed of luminance components corresponding to 16 × 16 pixels (dots), as shown in FIG. 10C, and this luminance component is further 8 × 8.
It is divided into blocks Y [1] to Y [4] each having a dot unit. Then, the 16 × 16 dot luminance component includes 8 × 8 dot Cb component blocks Cb [5] and 8
The block Cr [6] of the Cr component of × 8 dots is associated.

On the other hand, from the image coding type designating / image coding order rearranger 70, in order to predict the motion of the image signal S502 of the currently coded frame, the image signal S504 which becomes the reference image thereof. Are sent to the motion vector detection circuit 50. Further, in the motion vector detection circuit 50, a flag S101 indicating the start timing of the pair of fields that form the frame is sent from the image coding type designation / image coding order rearranger 70.
The image coding type information synchronized with each frame of the image signal S502 and the motion vector detection circuit 50 are supplied.
Processes the image signal of each frame as an I picture, a P picture, or a B picture, based on the judgment result of the prediction judgment circuit 54 described later, the flag S101, the image coding type information, and the flag S2. A frame processed as an I picture (for example, frame F1)
Image signal of the frame is transferred from the motion vector detection circuit 50 to the front original image storage unit 51a of the frame memory 51 and stored therein, and the image signal of the frame (for example, frame F2) processed as a B picture is the original image storage unit 51b. The image signal of the frame (for example, the frame F3) that is transferred to and stored in the rear original image storage unit 51c is stored in the rear original image storage unit 51c. The timing of storing the image signal in the frame memory 51 is based on the flag S101.

At the next timing, B
When an image signal of a frame to be processed as a picture (for example, frame F4) or a P picture (for example, frame F5) is input to the motion vector detection circuit 50, the first original image stored in the backward original image storage unit 51c until then. P
The image signal of the picture (frame F3) is transferred to the front original image storage unit 51a, and the next B picture (frame F3) is transmitted.
The image signal of 4) is stored (overwritten) in the original image storage unit 51b, and the image signal of the next P picture (frame F5) is stored (overwritten) in the backward original image storage unit 51c.
Such an operation is sequentially repeated.

Next, the macroblock signal S503 read from the scan converter 71 is sent to the prediction mode switching circuit 52, and as will be described later, as will be described later.
Frame prediction mode processing or field prediction mode processing is performed based on the judgment result of the prediction judgment circuit 54.
Further, the macroblock signal S 503 via the prediction mode switching circuit 52 is sent to the calculation unit 53.
In the calculation unit 53, one of intra-picture prediction, forward prediction, backward prediction, and bidirectional prediction is performed based on the determination result of the prediction determination circuit 54. Of these processes,
Which process is to be performed is determined corresponding to the prediction error (the difference between the reference image that is the target of the process and the predicted image for this). The prediction error is obtained by the motion vector detection circuit 50 as described later.

Here, the prediction mode switching circuit 52
The frame prediction mode processing and field prediction mode processing performed based on the determination result of the prediction determination circuit 54 will be described.

When the frame prediction mode is set by the prediction determination circuit 54, the prediction mode switching circuit 52 changes the four luminance component blocks Y [1] to Y [4] supplied from the scan converter 71 to It is output as it is to the arithmetic unit 53 in the subsequent stage. That is, in this case, as shown in FIG. 11A, the block Y of each luminance component
The signal of the line of the first field and the signal of the line of the second field are mixed from [1] to Y [4]. Therefore, in this frame prediction mode, prediction is performed in blocks of four luminance components, and one motion vector is associated with a unit of blocks of these four luminance components, that is, a macroblock. It

On the other hand, when the field prediction mode is set in the prediction determination circuit 54, the prediction mode switching circuit 52 changes the signal supplied from the scan converter 71 in the configuration shown in FIG. As shown in FIG. 11B, four luminance component blocks Y [1] to Y [Y]
Of the [4], the blocks Y [1] and Y [2] are configured only from the dots of the line of the first field, and the other two blocks Y [3] and Y [4] of the luminance components are It is composed of the data of the lines of two fields and is output to the arithmetic unit 53. In this case, two blocks Y
One motion vector is associated with [1] and Y [2], and another one motion vector is associated with the other two blocks Y [3] and Y [4]. Attached.

The color difference components of the Cb component and the Cr component are
In the frame prediction mode, as shown in FIG. 11A, the signals of the first field lines and the signals of the second field lines are mixed and supplied to the calculation unit 53.

On the other hand, in the field prediction mode, as shown in FIG. 11B, the block Cb of each color difference component
The upper half of [5], Cr [6] (that is, 4 lines) is
The first corresponding to the blocks Y [1] and Y [2] of the luminance component
The lower half (that is, 4 lines) is used as the color difference component of the field, and is used as the color difference component of the second field corresponding to the blocks Y [3] and Y [4] of the luminance component.

In order to do the above, the motion vector detection circuit 50 obtains the sum of absolute values of the prediction errors in the frame prediction mode and the sum of absolute values of the prediction errors in the field prediction mode. The value sum signal is output to the prediction determination circuit 54. Prediction determination circuit 5
4 compares the sum of absolute values of prediction errors in the frame prediction mode and the field prediction mode, selects a prediction mode having a small sum of absolute values of the prediction errors from the comparison result, and based on this prediction mode, the prediction mode The switching circuit 52 is controlled. The prediction mode switching circuit 52 is
The processing corresponding to the prediction mode in which the sum of absolute values of the prediction errors is small is applied to the macroblock signal S503 read from the scan converter 71. The processed signal is sent to the arithmetic unit 53.

When the flag S2 indicating that the image signal S6 is a 3: 2 pulldown processed signal is set, the image signal S6 has a progressive scan frame structure, and therefore the prediction mode switching circuit 5
The prediction mode in 2 is fixed to the frame prediction mode.

Here, the motion vector detection circuit 50
Generates the sum of absolute values of prediction errors for determining whether to perform intra-picture prediction, forward prediction, backward prediction, or bidirectional prediction in the prediction determination circuit 54 as follows.

That is, the motion vector detection circuit 50
The macroblock signal Aij of the reference image and the macroblock signal Aij are calculated as the sum of the absolute values of the prediction errors of the intra-picture prediction.
Then, the sum Σ | Aij-Aav | of the absolute value of the difference from the average value Aav of is calculated. Also, as the sum of absolute values of the prediction errors of forward prediction,
Sum Σ | Aij−Bij | of the absolute value of the difference between the signal Aij of the input macroblock and the signal Bij of the macroblock of the predicted image
Ask for. Further, the sum of absolute values of prediction errors between the backward prediction and the bidirectional prediction is also obtained in the same manner as in the case of forward prediction (changing the predicted image into a different predicted image from that in forward prediction).

The sum of these absolute values is sent to the prediction judgment circuit 54. The prediction determination circuit 54 selects the smallest sum of absolute values of prediction errors of forward prediction, backward prediction, and bidirectional prediction as the sum of absolute values of prediction errors of inter prediction. Furthermore, the sum of absolute values of the prediction error of this inter-prediction and the sum of absolute values of the prediction error of intra-picture prediction are compared,
The smaller one is selected, and the mode corresponding to the selected sum of absolute values is selected as the prediction mode. That is, if the sum of absolute values of prediction errors in intra-picture prediction is smaller, the intra-picture prediction mode is set. If the sum of absolute values of prediction errors in inter prediction is smaller, the mode in which the corresponding sum of absolute values is the smallest is set among the forward prediction, backward prediction, and bidirectional prediction modes. It should be noted that the sum of squares of the prediction error may be used for the judgment in the prediction judgment circuit 54.

In this way, the motion vector detection circuit 50
Detects the motion vector between the prediction image and the reference image corresponding to the prediction mode selected by the prediction determination circuit 54 among the four prediction modes, and outputs the motion vector information to the variable length coding circuit 58 and the motion compensation. Output to the circuit 64. As described above, the motion vector that minimizes the sum of absolute values of the corresponding prediction errors is selected.

When the image signal of the frame to be processed as the I picture is input, the prediction judgment circuit 54 sets the intra-frame prediction mode (mode without motion compensation) as the prediction mode, and the calculation unit 53 of the calculation unit 53. Switch 53
d is switched to the switched terminal a side. As a result, the image signal of the I picture is input to the DCT mode switching circuit 55.

As shown in FIG. 12A or 12B, this DCT mode switching circuit 55 sends the signals of the four luminance component blocks [1] to Y [4] to the line of the first field. The DCT is set to either the state in which the lines of the second field and the second field are mixed (frame DCT mode) or the state in which they are separated (field DCT mode).
Output to the circuit 56.

That is, the DCT mode switching circuit 55
Compares the coding efficiency in the case where DCT processing is performed with the data of the first field and the second field mixed and the coding efficiency in the case where DCT processing is performed in the separated state, and finds a mode with good coding efficiency. select.

For example, as shown in FIG. 12A, the input signal has a structure in which the lines of the first field and the second field are mixed, and the signal of the line of the first field vertically adjacent to the signal of the second field is mixed. The difference between the signals on the field lines is calculated, and the sum of the absolute values (or the sum of squares) is calculated. As shown in FIG. 12B, the input signal has a structure in which the lines of the first field and the second field are separated, and the difference between the signals of the lines of the vertically adjacent first field and The difference between the signals of the two field lines is calculated, and the sum (or sum of squares) of the absolute values of the two fields is calculated. Further, both (sum of absolute values) are compared, and the DCT mode corresponding to a smaller value is set. That is, if the former is smaller, the frame DCT mode is set, and if the latter is smaller, the field DCT mode is set.

Then, a signal having a configuration corresponding to the selected DCT mode is output to the DCT circuit 56, and a DCT flag indicating the selected DCT mode is output to the variable length coding circuit 58 and the motion compensation circuit 64.

When the flag S2 indicating that the 3: 2 pull-down process is performed is set, the image signal S6 has a progressive scan frame structure, so that the DCT mode is fixed to the frame DCT mode.

The prediction mode in the prediction mode switching circuit 52 (FIG. 11) and this DCT mode switching circuit 5
As is clear from comparison between the DCT modes in FIG. 5 (FIG. 12), regarding the blocks of the luminance component, the data structures in both modes are substantially the same.

The image signal of the I picture output from the DCT mode switching circuit 55 is input to the DCT circuit 56, subjected to DCT (discrete cosine transform) processing, and converted into DCT coefficients. This DCT coefficient data is stored in the quantization circuit 5
7 is quantized by a quantization step corresponding to the data storage amount (buffer storage amount) of the transmission buffer 59, and then input to the variable length coding circuit 58.

The variable length coding circuit 58 also transmits the image coding type from the information of the frame header and the information of the top field force and repeat first field.

Further, the variable length coding circuit 58 corresponds to the quantization step (scale) information supplied from the quantization circuit 57 and quantized DCT coefficient data (from the quantization circuit 57). In this case, I picture DC
The T coefficient data) is converted into a variable length code such as a Huffman code and output to the transmission buffer 59.

Further, the variable length coding circuit 58 receives the information of the quantization step (scale) from the quantization circuit 57, and the prediction judgment circuit 54 predicts the prediction mode (intra-picture prediction,
Information indicating which of forward prediction, backward prediction, and bidirectional prediction has been set), information about a motion vector from the motion vector detection circuit 50, and a prediction flag (frame prediction mode or field prediction) from the prediction mode switching circuit 52. Flag indicating which of the modes was set)
Is further output by the DCT mode switching circuit 55
CT flag (frame DCT mode or field DC
Flag indicating which of the T modes has been set), the flags S102 and S103 from the image coding type designation / image coding order rearranger 70, and the flag S2 supplied to the terminal 75. Has been entered,
In the variable length coding circuit 58, a flag S1 indicating the start timing of the pair of fields that make up the frame is described.
Based on 01, these flags and the like are also variable-length coded.

However, when the flag S2 indicating that the image signal has undergone 3: 2 pulldown processing is set, both the prediction flag and the DCT flag are fixed values in the frame mode, and therefore these are variable length coded. No output from the circuit 58. Instead, information that the flag S2 is set (information that the input frame has a progressive scan frame structure) is transmitted.

The transmission buffer 59 temporarily stores the encoded data supplied from the variable length encoding circuit 58, and outputs information corresponding to the accumulated amount to the quantization circuit 57 as a quantization control signal.

That is, the transmission buffer 59 increases the quantization scale of the quantization circuit 57 by the quantization control signal when the data storage amount is increased to the allowable upper limit value in which the data can be stored. Reduce the amount of data output from. Also, on the contrary,
The transmission buffer 59 uses the quantization control signal to quantize circuit 5 when the amount of accumulated data decreases to the lower limit.
By reducing the quantization scale of 7, the amount of data output from the quantization circuit 57 is increased. In this way, overflow or underflow of the transmission buffer 59 is prevented.

Then, the encoded data accumulated in the transmission buffer 59 is read out at a predetermined timing and output to the transmission line via the output terminal 79.

On the other hand, the quantized DCT coefficient data of the I picture output from the quantization circuit 57 is also sent to the dequantization circuit 60, where the quantization step information supplied from the quantization circuit 57. Inverse quantization is performed corresponding to. The output data of the inverse quantization circuit 60 is IDCT (inverse D
CT) circuit 61, and after inverse DCT processing, the frame / field DCT block switching circuit 65 switches the DCT block according to the frame / field DCT flag. It is supplied and stored in the forward prediction image storage unit 63a of the memory 63.

Next, when the scan converter 71 outputs the image signal of the frame to be processed as the P picture, the motion vector detection circuit 50 is operated in the same manner as in the above case.
Then, the sum of absolute values of prediction errors (differences between frames) in macroblock units is supplied to the prediction determination circuit 54. As a result, the prediction determination circuit 54 sets the frame / field prediction mode, the intra-picture prediction, or the forward prediction prediction mode in accordance with the sum of absolute values of the prediction errors of the macroblocks. Therefore, the prediction mode switching circuit 52
It operates based on the set prediction mode.

When the intra-frame prediction mode is set, the arithmetic unit 53 switches the switch 53d to the switched terminal a side as described above. Therefore, this image signal is
Similar to the image signal of the I picture described above, the DCT mode switching circuit 55, the DCT circuit 56, the quantization circuit 57,
The data is transmitted to the transmission line as encoded data via the variable length encoding circuit 58 and the transmission buffer 59. Also at this time, the DCT coefficient data output from the quantization circuit 57 is the inverse quantization circuit 60, the IDCT circuit 61, the frame / frame.
Field DCT block switching circuit 65, calculator 6
The backward prediction image storage unit 6 of the frame memory 63
3b is supplied and stored.

Here, in the forward prediction mode, the switch 53d is switched to the switched terminal b, and the image signal stored in the forward prediction image portion 63a of the frame memory 63 (in this case, the image signal of the I picture). ) Is read out and sent to the motion compensation circuit 64, where motion compensation is performed according to the motion vector information output from the motion vector detection circuit 50.

The predicted image signal output from the motion compensation circuit 64 is supplied to the calculator 53a. The calculator 53a is
The predicted image signal corresponding to the macroblock supplied from the motion compensation circuit 64 is subtracted from the signal of the macroblock of the reference image supplied from the prediction mode switching circuit 52, and the difference (prediction error) is output. This difference signal is supplied to the DCT mode switching circuit 55 and the DCT circuit 5
6, the quantization circuit 57, the variable length coding circuit 58, and the transmission buffer 59 are transmitted to the transmission line as encoded data. In addition, this difference signal is the inverse quantization circuit 60, the IDC.
It is locally decoded by the T circuit 61 and the frame / field DCT block switching circuit 65 and input to the arithmetic unit 62.

However, when the flag S2 indicating that the signal has undergone 3: 2 pulldown processing is set, both the prediction flag and the DCT flag are fixed values in the frame mode, and therefore these are variable length coding circuits. No output from 58. Instead, it transmits information that the flag S2 is set (information that the input frame has a progressive scan frame structure).

The same signal as the predicted image signal supplied to the calculator 53a is also supplied to the calculator 62. The calculator 62 adds the predicted image signal output by the motion compensation circuit 64 to the difference signal output by the IDCT circuit 61 and switched by the frame / field DCT block switching circuit 65. As a result, the locally decoded image signal of the P picture is obtained. The image signal of the P picture is stored in the backward prediction image storage unit 6 of the frame memory 63.
3b is supplied and stored.

Then, when the image signal of the frame to be processed as a B picture is output from the scan converter 71, the motion vector detection circuit 50, as in the case described above.
From, the sum of absolute values of prediction errors (differences between frames) in macroblock units is sent to the prediction determination circuit 54.
Thereby, in the prediction determination circuit 54, the frame / field prediction mode or the prediction mode is set to the intra-frame prediction mode, the forward prediction mode, the backward prediction mode, or the bidirectional prediction mode in accordance with the sum of the absolute values of the prediction errors of the macroblocks. Set to either. Therefore, the prediction mode switching circuit 52 operates based on the set prediction mode.

As described above, in the intra-frame prediction mode or the forward prediction mode, the switch 53d is switched to the switched terminals a and b, respectively. At this time, the same processing as in the P picture is performed and the data is transmitted.

On the other hand, when the backward prediction mode or the bidirectional prediction mode is set, the switch 53d is switched to the switched terminals c and d, respectively.

In the backward prediction mode in which the switch 53d is switched to the switched terminal c, the backward prediction image storage unit 6
The image signal (in this case, the image signal of the P picture) stored in 3b is read out, and the motion compensation circuit 64 performs motion compensation corresponding to the motion vector information output from the motion vector detection circuit 50.

The predicted image signal output from the motion compensation circuit 64 is supplied to the calculator 53b. The calculator 53b is
The prediction image signal supplied from the motion compensation circuit 64 is subtracted from the image signal of the input macroblock supplied from the prediction mode switching circuit 52, and the difference is output. This difference signal is transmitted to the transmission line as encoded data via the DCT mode switching circuit 55, the DCT circuit 56, the quantization circuit 57, the variable length encoding circuit 58, and the transmission buffer 59. In addition, this difference signal is the inverse quantization circuit 60, I
The DCT circuit 61 and the frame / field DCT block switching circuit 65 locally decode the data, and the arithmetic unit 62
Is input to

The same signal as the predicted image signal supplied to the calculator 53b is also supplied to the calculator 62. The arithmetic unit 62 outputs from the IDCT circuit 61, and further the frame / field DCT block switching circuit 6
The motion compensation circuit 64 is added to the differential signal switched by
The predicted image signals output by are added. As a result, a locally decoded B picture image signal is obtained.

In the bidirectional prediction mode in which the switch 53d is switched to the switched terminal d, the image signal (in this case, the I picture image signal) stored in the forward predicted image storage unit 63a and the backward predicted image are stored. The image signal (in this case, the image signal of the P picture) stored in the storage unit 63b is read out, and the motion compensation circuit 64 performs motion compensation corresponding to the motion vector information output from the motion vector detection circuit 50. It

The predicted image signal output from the motion compensation circuit 64 is supplied to the calculator 53c. The calculator 53c is
The average value of the prediction image signal supplied from the motion compensation circuit 64 is subtracted from the image signal of the input macroblock supplied from the prediction mode switching circuit 52, and the difference is output. This difference signal is supplied to the DCT mode switching circuit 55,
DCT circuit 56, quantization circuit 57, variable length coding circuit 5
8. The data is transmitted to the transmission line as encoded data via the transmission buffer 59. In addition, this difference signal is used for the inverse quantization circuit 60, the IDCT circuit 61, the frame / field DCT.
The data is locally decoded by the block switching circuit 65 and input to the arithmetic unit 62.

The same signal as the predicted image signal supplied to the calculator 53c is also supplied to the calculator 62. The arithmetic unit 62 outputs from the IDCT circuit 61, and further the frame / field DCT block switching circuit 6
The predicted image signal output from the motion compensation circuit 64 is added to the difference signal switched in 5. As a result, a locally decoded B picture image signal is obtained.

Here, when the flag S2 is set to indicate that the signal is obtained by the 3: 2 pull-down processing, both the prediction flag and the DCT flag are fixed values in the frame prediction mode, so these are variable length codes. It is not output from the digitizing circuit 58. Instead, information that the flag S2 is set (information that the input frame has a progressive scan frame structure) is transmitted.

The B-picture image signal is not stored in the frame memory 63 because it is not used as a predicted image of another image.

In the frame memory 63, the forward predicted image storage section 63a and the backward predicted image storage section 63b are
Bank switching is performed as necessary, and one stored in one or the other with respect to a predetermined reference image can be switched and output as a forward prediction image signal or a backward prediction image signal.

In the above description, the block of the luminance component is mainly described, but the block of the color difference component is similarly processed and transmitted in units of the macro blocks shown in FIGS. 11 and 12. In addition, as a motion vector when processing a block of a color difference component, a motion vector of a block of a corresponding luminance component that is halved in each of the vertical direction and the horizontal direction is used.

The bit stream of the encoded data generated by the encoder 17 of FIG. 1 as described above is recorded on the recording medium 18.

As is clear from the above description, in the image coding system of FIG. 1 to which the moving image coding method and apparatus of the present invention are applied, the coding frame rate differs in the sequence of the input image signal. When encoding an image signal sequence in which moving image materials are mixed, a recording medium in which position information, for example, edit point information, at which the encoding frame rate changes in the image signal sequence is stored in advance is prepared. The coding frame rate can be changed, that is, whether or not the input image is coded can be controlled according to the information read from the.

In addition, the present system is, for example, an image signal (encoding frame rate 24
Hz) and an image signal obtained by edit-combining an image signal (encoding frame rate 30 Hz) captured by a television camera are very effective in efficiently encoding. That is, when encoding an image sequence, it is possible to know the image signal portion by the 3: 2 pulldown process from the image signal sequence by referring to the edit point information prepared in advance corresponding to the image signal sequence. Therefore, the redundant field detection and removal method by 3: 2 pulldown processing can be applied only to that portion, and the frame coding efficiency can be improved. Further, in the image signal portion photographed by the television camera, the above redundant field detection and removal method can be applied so that the problem of erroneously removing the originally necessary field does not occur, and the motion of the moving image does not occur. There is no problem of unnaturalness and no problem of reduction in frame coding efficiency.

Next, a second embodiment for realizing the moving picture coding method of the present invention will be described with reference to FIG.
FIG. 13 shows the configuration of an image coding system to which the second moving image coding apparatus of the present invention is applied.

In the configuration of FIG. 13, an image of a film source 21 of 24 frames per second such as a motion picture film is subjected to 3: 2 pull-down processing to obtain, for example, N in a television broadcasting system.
A telecine device 22 for performing telecine conversion into an image signal having a field rate of 30 Hz, such as the TSC system, and recording this image signal on a video tape, for example, with a VTR, and an image signal output from the telecine device 22 are input image data. An image encoding device 30 that encodes as a signal to generate encoded data.

Here, the image signal obtained by the 3: 2 pull-down process has a coding frame rate of 2/60 seconds when one frame is read in two fields, and when one frame is read in three fields. It will change to 3/60 seconds.

The telecine device 22 determines whether the image signal S30 of the field sequence obtained by the 3: 2 pulldown process and each field of the image signal S30 read one frame of the film 21 in three fields. A flag S31 indicating whether or not it is output. The flag S31 indicates that the image signal S30 corresponds to the film 21.
Is read out in 3 fields, and is set to "1" when the telecine device 22 outputs the repeated field of the 3 fields.

The flag S31 indicates the image signal S30.
Along with this, it is sent to the VTR 24 and recorded on the video tape in the VTR 24 as the header information of the image signal S30. That is, the image signal S3 of the field sequence
0 and the flag S31 are 1 to 1 for each field.
Correspondingly, it is recorded on the video tape in the VTR 24.

Thus, the video tape 25 recorded by the VTR 24 has the image signal (S30) of the field sequence by the 3: 2 pull-down process, and each field included in the sequence has three film frames.
Information (flag S31) indicating whether or not the data is read in the field is recorded.

The image signal S30 and the flag S31 are
It does not need to be recorded in the same recording medium, and may be recorded in different media. For example, the floppy disk 26 shown in FIG. 12 may be recorded. In this case, for example, the image signal S30 may be recorded on the video tape in the VTR 24 and the flag S31 may be recorded on the floppy disk 26. You can do it.

Here, as a recording method of the flag S31 which is recorded together with the image signal S30, for example, as in the case of FIG. 1, the user bits of the SMPTE time code are used and the flag is recorded on the tape together with the image signal. A possible method is to keep a record. Therefore, the configuration of FIG.
A VITC insertion circuit 23 that operates in the same manner as in FIG. 1 is provided, and the image signal S30 and the flag S31 are supplied to the VITC insertion circuit 23, and the flag S31 is set to the user bit of VITC in the image signal S30. The inserted image signal S32 is formed. Also, the VITC and LTC may have a bit configuration other than the formats shown in FIGS. 2 and 4 as described above, and in this case, it suffices that the recording side and the receiving side have a correspondence. Note that products that record information indicating the relationship of 3: 2 pull-down processing at the time of telecine to VITC include those manufactured by AATON and Everts. These products have traditionally been used to facilitate non-linear film editing.

The video tape 25 on which the above-mentioned image signal S32 is recorded is reproduced by the VTR 31, and the VT
The image signal S23 obtained by the reproduction in R31 is sent to the image encoding device 30. In addition, this video tape 25
The image signal S23 reproduced from
Is the same as

Upon receiving the image signal S23 from the VTR 31, the image coding device 30 reads the image signal S30 from the image signal S23 and records it in the user bit of the SMPTE time code of the image signal, for example, in the VITC. Read the flag S31 (that is, information indicating the relationship of 3: 2 pulldown processing)
The redundant field of the image signal is removed based on the flag S31, and then the image signal is encoded.

In order to do this, the VTR3
Image signal S23 supplied from 1 to the image encoding device 30
Are first sent to the VITC reading circuit 32. The V
The ITC reading circuit 32 reads the flag S31 arranged in the user bit of the SMPTE time code, for example, VITC from the image signal S23, separates it from the image signal S30 by the 3: 2 pull-down process, and separates it. The image signal S30 is sent to the path after the redundant field remover 33 as the image signal S20. On the other hand, the VITC reading circuit 32 is
Flag S corresponding to flag S31 read from the ITC
21 is output. The flag S21 is sent to the redundant field remover 33 as a control signal indicating whether to remove the redundant field.

In the redundant field remover 33, the VITC reading circuit 3 is based on the flag S21.
The image signal corresponding to the redundant image is removed from the image signal S20 supplied from 2 by the 3: 2 pull-down processing.
As a result, the image signal of the redundant image is not encoded.

The image signal S22 of the field sequence output from the redundant field remover 33 is input to the scan converter 34 together with the flag S21. The scan converter 34 is similar to the scan converter 16 in FIG. That is, in the scan converter 34, the redundant field remover 3
The image signal S22 of the field sequence after the redundant field is removed in 3 is converted into an image signal of the frame sequence in the order of input.

The image signal of the frame sequence output from the scan converter 34 is sent to the encoder 35 together with the same flags S101, S102 and S103 as described above. In the encoder 35, similarly to the above, the image signal of the frame sequence is highly efficiently encoded, and the flag S
102 and S103 are encoded. The encoder 3
5 is the same as the encoder 17 in the basic configuration except that the encoding process using the flag S2 in the encoder 17 in FIG. 1 is not controlled. That is, in the encoder 35 in the configuration example of this FIG.
Since the image signal input to the encoder 35 is an image signal subjected to 3: 2 pulldown processing from the film source, the same operation as when "1" is set in the flag S2 in the example of FIG. Becomes

The coded data (bit stream) obtained by high-efficiency coding the image signal in the encoder 35 and coding the flags S102 and S103 is then recorded in the recording medium 36.

For example, when the flag S31 is recorded on the floppy disk 26 shown in FIG. 13, the VITC reading circuit 32 outputs the flag S21 corresponding to the flag S31 read from the floppy disk 26. It is also possible to make it happen. Of course, instead of outputting the flag S21 from the VITC reading circuit 32, the flag S31 read from the floppy disk 26 may be directly sent to the redundant field remover 21 as the flag S21.

Next, referring to FIG. 14, the VI of FIG.
The operation of the configuration from the TC reading circuit 32 to the scan converter 34 will be described.

In FIG. 14, the image signal S20 of the field rate output from the VITC reading circuit 32 of FIG. 13 and a flag S21 indicating whether the image signal S20 is a repeating field by 3: 2 pulldown processing. And the image signal S22 after the redundant field is removed by the redundant field remover 33 in FIG.
As in the case of FIG. 6, when converting the field rate into the frame rate by the scan converter 34 of FIG. , A flag S102 that indicates whether the frame starts from the first field (top field) or the second field (bottom field) and becomes “1” when the frame starts from the first field, and the frame is originally a film. The flag S103 is "1" when the frame is composed of 2 fields obtained by removing 1 field from 3 fields generated from the same frame of the source. Also, regarding the image signal S20 in FIG.
Alternatively, f is an image signal of a film source that has undergone 3: 2 pulldown processing, an uppercase letter F represents the first field, a lowercase letter f represents the second field, and the same numbers among these letters F and the subscript numbers of f It represents a field read from the same frame of film.

As can be seen from FIG. 14, the image signal S
When 20 is a repeat field by 3: 2 pulldown processing, the flag S21 is "1", and when it is not a repeat field, the flag S21 is "0". Therefore, when the flag S21 is set to "1", the redundant field remover 33 determines that the field of the image signal S20 from the VITC reading circuit 32 is a redundant field and sets the redundant field to the image signal. Remove from S20.

As in the case of FIG. 1, the scan converter 34 having the configuration of FIG. 13 outputs the image signal of the frame rate together with the flags S101, S102, S103 of FIG. 14 attached thereto. Therefore, the image signal and the flag S101,
S102 and S103 are input. Further, when the encoder 35 performs the encoding according to the MPEG2 standard in the same manner as described above, the flags S102 and S103 are the top field first (TFF), which is defined by the MPEG2, respectively. Repeat first field (repeat_first_field:
RFF).

As is clear from the above description, in the image encoding system of FIG. 13 to which the moving image encoding method and apparatus of the present invention are applied, for example, an image signal by 3: 2 pulldown processing is highly efficient encoded. It is very effective when doing. That is, when the image signal sequence is encoded, the repetition is performed by referring to the information, which is prepared in advance corresponding to the image signal sequence, whether each field is one frame of film read out three fields. Since it is possible to know the position of the redundant field that is present, the redundant field can be removed efficiently, and therefore the frame coding efficiency can be improved. Further, there is no problem of erroneously removing an originally necessary field that is not a redundant field, there is no problem that motion of a moving image is unnatural, and there is no problem that frame coding efficiency is reduced.

From the above, the moving picture coding system of FIGS. 1 and 13 to which the moving picture coding method and apparatus of the present invention are applied has a very large effect in practice.

As described above, according to the present invention, the editor or the like records the image coding control information (flags S11 and S31) together with the image signal on a recording medium such as a video tape in advance. There is. In the above description, for example, the above encoding control information is recorded in the user bit of the SMPTE time code corresponding to the image, for example, VITC or LTC, and the encoding control information is read on the image encoding device side, and based on it. As a control of the image coding method, the method of controlling the removal of the repeated field by the 2: 3 pulldown has been mainly described, but various methods can be considered as the control of the image coding method. To be

For example, a video editor sets a scene change flag in VITC or LTC in the field for scene change. Then, on the side of the image encoding device, the scene change flag recorded in VITC or LTC is read out, and based on this, the first field of the scene change or the field including it is encoded with an I picture.

Further, for example, when the video editor wants to make the coded image quality high for a certain scene, a high image quality instruction flag is set in VITC or LTC of the field. On the image encoding device side, VITC and L
The high image quality instruction flag recorded in the TC is read out, and based on this, the encoding bit rate of the scene is increased and encoding is performed.

Further, for example, when the video editor pulls down one frame of the film by 2: 2, the V of the first field is
A head field flag is set in ITC or LTC. Here, the 2: 2 pulldown is a telecine conversion method which is widely used when converting a film image of 24 frames into an image signal of 25 frames per second (50 fields per second). This is a method of reading one frame of film by interlaced scanning (interlaced scanning) with two fields of an image signal. Then, on the side of the image encoding device, the first field flag recorded in VITC or LTC is read out, two fields read out from the same frame constitute one frame, and the frame is encoded.

Thus, along with the image, the user bit of the SMPTE time code corresponding to the image, for example, VIT
The information recorded in C or LTC is read out, and based on the information, various uses can be made for controlling the image encoding method on the image encoding device side.

Next, a third embodiment for realizing the moving picture coding method of the present invention will be described. The third embodiment is an example in which the recommendation of SMPTE called Video Index Information is used as a recording method of the flag S11 to be recorded together with the image signal S12 as shown in FIG.

The above video index information is recorded in the blanking of a so-called 4: 2: 2 component digital video signal, and in a system using 512 lines such as the NTSC system, the color difference signal of the 14th line and the 277th line ( Cb, Cr). The luminance signal (Y) of the same line is D
VITC (Digital Vertical Interval Time Code) is transmitted.

The video index information is represented by 90 bytes of information using the color difference signal of 720 samples of the effective portion of one line. This is shown in FIG.
This will be described with reference to FIG. The effective portion of one line of a 4: 2: 2 component digital video signal is Cb, Y, C.
1440 in the order of r, Y, Cb, Y, Cr, Y, ...
There is a sample. Each sample is 10 bits long. In general, the 2 bits on the LSB (least significant bit) side are always "0", and the effective bit length is 8 bits. Also, the value 204h (hexadecimal notation) of the sample of the color difference signal of 10-bit length represents the binary "1",
The value 200h represents a binary number "0".

The first color difference signal (sample word No. 0
Cb) is an 8-bit video index word 0
LSB (bit 0) of The second color difference signal (Cr in sample word No. 2) represents bit 1 of video index word 0. The same applies thereafter, and the eighth color difference signal (Cr of sample word No. 14) is
It represents the MSB (bit 7) of video index word 0. After that, the last color difference signal (Cr in sample word No. 1438) represents the MSB (bit 7) of the video index word 89. In this way, 90-byte video index information of video index words 0 to 89 is represented.

The data format of the video index information is defined in the above-mentioned recommendation (Proposal
ed SMPTE recommended practice RP-186 "Vide
o Index Information Coding for 525 and 625 Line Te
levision Systems ”(August20, 1995)) and is specified as shown in Fig. 16. This video index information is mainly for recording information of the source before becoming digital video. belongs to.

When the third embodiment is made to correspond to the first embodiment described above, the video index word No. 1
The lower 4 bits of 4 are used to transmit information indicating whether the video signal is an image by 3: 2 pulldown processing or an image signal captured by a TV camera. Then, the moving picture coding apparatus detects the information and controls the coding method.

As a concrete configuration when the third embodiment is made to correspond to the first embodiment, the VITC insertion circuit 6 and the VITC reading circuit in FIG. 1 which are the configurations of the first embodiment are described. The video data 12 may be replaced with the video index information insertion circuit 6 and the video index information reading circuit 12, respectively.

When the third embodiment is made to correspond to the second embodiment, the video index word N
o. 12 upper 4-bit video fields, film frame data (fil)
m frames data) is used to transmit information indicating whether the field of the 3: 2 pulldown video signal is a repeated field by 3: 2 pulldown processing, and the moving picture coding apparatus uses the information. To control the encoding method.

As a concrete configuration when the third embodiment is made to correspond to the second embodiment, the VITC insertion circuit 23 and the VITC reading circuit 32 in FIG. 13 are respectively the video index information insertion circuit. 6 and the video index information reading circuit 12 is changed.

Note that, for example, when using information designating a field for scene change, the information is the video index word No. Lower 4 bits of 14 (upper 4bi
The transmission is performed using the source flag data of t), and the moving image encoding device detects the information and controls the encoding method.

When the information indicating the leading field of the 3: 2 pulldown film is used, the information is the video index word No. 12 upper 4-bit video fields,
It is transmitted using film frames data, and the moving picture coding device detects the information and controls the coding method.

[0168]

As is apparent from the above description, in the moving picture coding method and apparatus of the present invention, the image signals obtained from the moving picture materials having different coding frame rates are mixed in the sequence of the image signals. At this time, the position information in which the coding frame rate changes is detected in the sequence of the image signal, and the coding process of the moving image signal is changed based on the position information to obtain the 3: 2 pull-down process. It is possible to efficiently encode a moving image signal in which moving image materials having different encoding frame rates are mixed in a sequence of image signals, such as a signal in which image signals and image signals captured by a television camera are mixed. It will be possible.

[Brief description of the drawings]

FIG. 1 is a block circuit diagram showing a schematic configuration of an image coding system of a first embodiment to which a moving image coding method and apparatus of the present invention are applied.

FIG. 2 is a diagram for explaining a VITC time code.

FIG. 3 is a diagram for explaining a signal position of a horizontal line of an image signal of digital data.

FIG. 4 is a diagram for explaining an LTC time code.

FIG. 5 is a block circuit diagram showing a specific configuration of a redundant field detection and removal circuit.

6 is a timing chart when an image signal obtained by 3: 2 pulldown processing and an image signal obtained by editing an image signal captured by a television camera are processed by the image encoding system in FIG. 1. FIG.

[Fig. 7] Fig. 7 is a diagram for describing a picture coding type.

FIG. 8 is a diagram for explaining the principle of an image signal coding method.

FIG. 9 is a block circuit diagram showing a specific configuration of an encoder.

FIG. 10 is a diagram for explaining the structure of an image signal.

FIG. 11 is a diagram for explaining a frame / field prediction mode.

FIG. 12 is a diagram for explaining a frame / field DCT mode.

FIG. 13 is a block circuit diagram showing a schematic configuration of an image encoding system of a second embodiment to which the moving image encoding method and apparatus of the present invention is applied.

FIG. 14 shows an image signal by 3: 2 pulldown processing.
3 is a timing chart when processed by the image coding system of No. 3 in FIG.

FIG. 15 is a diagram used for explaining video index information.

FIG. 16 is a diagram showing a data format of video index information.

FIG. 17 is a diagram for explaining 3: 2 pulldown processing.

FIG. 18 is a block circuit diagram showing a schematic configuration of a conventional image encoding system.

FIG. 19 is a diagram for explaining processing for detecting a redundant field by 3: 2 pulldown processing and removing it.

[Explanation of symbols]

 1 video editing device 2,3,7,11,24,25 VTR 4,13,15 switch 5 video editing controller 6,23 VITC insertion circuit 9,25 video tape 10,30 image coding device 12,32 VITC reading circuit 14 Redundant field detector and remover 16,34 Scan converter 17,35 Encoder 18,36 Recording medium 33 Redundant field remover

Claims (21)

[Claims] 1. A moving picture coding method for coding a moving picture signal in which a picture signal obtained from moving picture materials having different coding frame rates are mixed in a sequence of picture signals. A moving picture coding method, which detects position information in which a coding frame rate changes in a sequence, and changes the coding process of the moving picture signal based on the detected position information. 2. The moving image signal comprises a film image 3:
2. The moving image according to claim 1, wherein the 2 pulldown-processed moving image signal and a moving image signal captured by a television camera are edited and combined, and the edit point information of the edited combination is detected as the position information. Encoding method. 3. The moving image signal comprises a film image 3:
Detecting, as the position information, information indicating a position where one frame of a film image is converted into a three-field image or a position where one frame is converted to a two-field image, which is composed of a moving image signal subjected to 2 pulldown processing. The moving picture encoding method according to claim 1. 4. When the moving image signal subjected to the 3: 2 pulldown is a signal obtained by converting one frame of a film image into an image of three fields, a redundant field is removed from the three fields. The moving image encoding method according to claim 2. 5. The moving picture coding method according to claim 3, wherein when one frame of the film image is converted into an image of three fields, redundant fields among the three fields are removed. 6. The moving picture coding method according to claim 2, wherein the moving picture signal photographed by the television camera is subjected to frame / field adaptive motion compensation predictive coding. 7. The moving picture coding method according to claim 1, wherein the position information arranged in a vertical blanking period of the moving picture signal is detected. 8. The moving image signal is reproduced from a recording medium on which the moving image signal is recorded, and the position information is reproduced from a recording medium on which the position information is recorded. Video coding method. 9. A moving picture coding apparatus for coding a moving picture signal in which a moving picture material obtained from moving picture materials having different coding frame rates are mixed in a sequence of picture signals, A detection means for detecting position information in which the coding frame rate changes in the sequence, and a coding means for changing the coding process of the moving image signal based on the detected position information. Video encoding device. 10. The moving image signal is an edited combination of a moving image signal obtained by subjecting a film image to 3: 2 pulldown processing and a moving image signal taken by a television camera, and the detecting means uses the position information as the position information. 10. The moving picture coding apparatus according to claim 9, wherein edit point information of edit combination is detected. 11. The moving image signal comprises a moving image signal obtained by subjecting a film image to 3: 2 pull-down processing, and the detecting means detects a position where one frame of the film image is converted into an image of three fields as the position information. , Or 1 frame is 2
10. The moving picture coding apparatus according to claim 9, wherein information indicating a position converted into a field image is detected. 12. When the moving image signal subjected to the 3: 2 pull-down processing is a signal obtained by converting one frame of a film image into an image of three fields, a redundant field removing means for removing a redundant field of the three fields is provided. The moving picture coding apparatus according to claim 10, wherein the moving picture coding apparatus is provided. 13. A redundant field removing means for removing a redundant field among the three fields when one frame of the film image is converted into an image of three fields, according to claim 11. Video coding device. 14. The moving picture coding apparatus according to claim 10, wherein said coding means performs frame / field adaptive motion compensation predictive coding on the moving picture signal captured by said television camera. 15. The moving picture coding apparatus according to claim 9, wherein the detecting means detects the position information arranged in a vertical blanking period of the moving picture signal. 16. The encoding means encodes the moving image signal reproduced from the recording medium on which the moving image signal is recorded, and the detecting means reproduces from the recording medium on which the position information is recorded. The moving image coding apparatus according to claim 9, wherein the position information is detected. 17. A moving image signal in which image signals obtained from moving image materials having different encoding frame rates are mixed in a sequence of image signals, and the encoding frame rate changes in the sequence of image signals. A recording medium characterized by recording position information indicating a position. 18. The moving image signal comprises a film image 3:
18. The recording medium according to claim 17, wherein the moving image signal subjected to the 2 pulldown processing and the moving image signal captured by the television camera are edited and combined, and the position information is edit point information of the edited and combined. 19. The moving image signal comprises a film image 3:
It is composed of a moving image signal subjected to 2 pulldown processing, and the position information is information indicating a position where one frame of a film image is converted into a three-field image or one frame is converted into a two-field image. 17. The method according to claim 17,
The recording medium described. 20. The recording medium according to claim 17, wherein the position information is recorded during a vertical blanking period of the moving image signal. 21. The recording medium according to claim 17, wherein the position information is recorded in an area in which a time code is recorded, other than an area in which the moving image signal is recorded.